A Comparison of a Machine Learning Model with EuroSCORE II in Predicting Mortality after Elective Cardiac Surgery: A Decision Curve Analysis.

PubMed

Allyn, Jérôme; Allou, Nicolas; Augustin, Pascal; Philip, Ivan; Martinet, Olivier; Belghiti, Myriem; Provenchere, Sophie; Montravers, Philippe; Ferdynus, Cyril

2017-01-01

The benefits of cardiac surgery are sometimes difficult to predict and the decision to operate on a given individual is complex. Machine Learning and Decision Curve Analysis (DCA) are recent methods developed to create and evaluate prediction models. We conducted a retrospective cohort study using a prospective collected database from December 2005 to December 2012, from a cardiac surgical center at University Hospital. The different models of prediction of mortality in-hospital after elective cardiac surgery, including EuroSCORE II, a logistic regression model and a machine learning model, were compared by ROC and DCA. Of the 6,520 patients having elective cardiac surgery with cardiopulmonary bypass, 6.3% died. Mean age was 63.4 years old (standard deviation 14.4), and mean EuroSCORE II was 3.7 (4.8) %. The area under ROC curve (IC95%) for the machine learning model (0.795 (0.755-0.834)) was significantly higher than EuroSCORE II or the logistic regression model (respectively, 0.737 (0.691-0.783) and 0.742 (0.698-0.785), p < 0.0001). Decision Curve Analysis showed that the machine learning model, in this monocentric study, has a greater benefit whatever the probability threshold. According to ROC and DCA, machine learning model is more accurate in predicting mortality after elective cardiac surgery than EuroSCORE II. These results confirm the use of machine learning methods in the field of medical prediction.

Statistical assessment of the learning curves of health technologies.

PubMed

Ramsay, C R; Grant, A M; Wallace, S A; Garthwaite, P H; Monk, A F; Russell, I T

2001-01-01

(1) To describe systematically studies that directly assessed the learning curve effect of health technologies. (2) Systematically to identify 'novel' statistical techniques applied to learning curve data in other fields, such as psychology and manufacturing. (3) To test these statistical techniques in data sets from studies of varying designs to assess health technologies in which learning curve effects are known to exist. METHODS - STUDY SELECTION (HEALTH TECHNOLOGY ASSESSMENT LITERATURE REVIEW): For a study to be included, it had to include a formal analysis of the learning curve of a health technology using a graphical, tabular or statistical technique. METHODS - STUDY SELECTION (NON-HEALTH TECHNOLOGY ASSESSMENT LITERATURE SEARCH): For a study to be included, it had to include a formal assessment of a learning curve using a statistical technique that had not been identified in the previous search. METHODS - DATA SOURCES: Six clinical and 16 non-clinical biomedical databases were searched. A limited amount of handsearching and scanning of reference lists was also undertaken. METHODS - DATA EXTRACTION (HEALTH TECHNOLOGY ASSESSMENT LITERATURE REVIEW): A number of study characteristics were abstracted from the papers such as study design, study size, number of operators and the statistical method used. METHODS - DATA EXTRACTION (NON-HEALTH TECHNOLOGY ASSESSMENT LITERATURE SEARCH): The new statistical techniques identified were categorised into four subgroups of increasing complexity: exploratory data analysis; simple series data analysis; complex data structure analysis, generic techniques. METHODS - TESTING OF STATISTICAL METHODS: Some of the statistical methods identified in the systematic searches for single (simple) operator series data and for multiple (complex) operator series data were illustrated and explored using three data sets. The first was a case series of 190 consecutive laparoscopic fundoplication procedures performed by a single surgeon; the second was a case series of consecutive laparoscopic cholecystectomy procedures performed by ten surgeons; the third was randomised trial data derived from the laparoscopic procedure arm of a multicentre trial of groin hernia repair, supplemented by data from non-randomised operations performed during the trial. RESULTS - HEALTH TECHNOLOGY ASSESSMENT LITERATURE REVIEW: Of 4571 abstracts identified, 272 (6%) were later included in the study after review of the full paper. Some 51% of studies assessed a surgical minimal access technique and 95% were case series. The statistical method used most often (60%) was splitting the data into consecutive parts (such as halves or thirds), with only 14% attempting a more formal statistical analysis. The reporting of the studies was poor, with 31% giving no details of data collection methods. RESULTS - NON-HEALTH TECHNOLOGY ASSESSMENT LITERATURE SEARCH: Of 9431 abstracts assessed, 115 (1%) were deemed appropriate for further investigation and, of these, 18 were included in the study. All of the methods for complex data sets were identified in the non-clinical literature. These were discriminant analysis, two-stage estimation of learning rates, generalised estimating equations, multilevel models, latent curve models, time series models and stochastic parameter models. In addition, eight new shapes of learning curves were identified. RESULTS - TESTING OF STATISTICAL METHODS: No one particular shape of learning curve performed significantly better than another. The performance of 'operation time' as a proxy for learning differed between the three procedures. Multilevel modelling using the laparoscopic cholecystectomy data demonstrated and measured surgeon-specific and confounding effects. The inclusion of non-randomised cases, despite the possible limitations of the method, enhanced the interpretation of learning effects. CONCLUSIONS - HEALTH TECHNOLOGY ASSESSMENT LITERATURE REVIEW: The statistical methods used for assessing learning effects in health technology assessment have been crude and the reporting of studies poor. CONCLUSIONS - NON-HEALTH TECHNOLOGY ASSESSMENT LITERATURE SEARCH: A number of statistical methods for assessing learning effects were identified that had not hitherto been used in health technology assessment. There was a hierarchy of methods for the identification and measurement of learning, and the more sophisticated methods for both have had little if any use in health technology assessment. This demonstrated the value of considering fields outside clinical research when addressing methodological issues in health technology assessment. CONCLUSIONS - TESTING OF STATISTICAL METHODS: It has been demonstrated that the portfolio of techniques identified can enhance investigations of learning curve effects. (ABSTRACT TRUNCATED)

Learning curves for single incision and conventional laparoscopic right hemicolectomy: a multidimensional analysis.

PubMed

Park, Yoonah; Yong, Yuen Geng; Yun, Seong Hyeon; Jung, Kyung Uk; Huh, Jung Wook; Cho, Yong Beom; Kim, Hee Cheol; Lee, Woo Yong; Chun, Ho-Kyung

2015-05-01

This study aimed to compare the learning curves and early postoperative outcomes for conventional laparoscopic (CL) and single incision laparoscopic (SIL) right hemicolectomy (RHC). This retrospective study included the initial 35 cases in each group. Learning curves were evaluated by the moving average of operative time, mean operative time of every five consecutive cases, and cumulative sum (CUSUM) analysis. The learning phase was considered overcome when the moving average of operative times reached a plateau, and when the mean operative time of every five consecutive cases reached a low point and subsequently did not vary by more than 30 minutes. Six patients with missing data in the CL RHC group were excluded from the analyses. According to the mean operative time of every five consecutive cases, learning phase of SIL and CL RHC was completed between 26 and 30 cases, and 16 and 20 cases, respectively. Moving average analysis revealed that approximately 31 (SIL) and 25 (CL) cases were needed to complete the learning phase, respectively. CUSUM analysis demonstrated that 10 (SIL) and two (CL) cases were required to reach a steady state of complication-free performance, respectively. Postoperative complications rate was higher in SIL than in CL group, but the difference was not statistically significant (17.1% vs. 3.4%). The learning phase of SIL RHC is longer than that of CL RHC. Early oncological outcomes of both techniques were comparable. However, SIL RHC had a statistically insignificant higher complication rate than CL RHC during the learning phase.

Evaluation of the learning curve for external cephalic version using cumulative sum analysis

PubMed Central

Kim, So Yun; Chang, Eun Hye; Kwak, Dong Wook; Ahn, Hyun Kyung; Ryu, Hyun Mi; Kim, Moon Young

2017-01-01

Objective We evaluated the learning curve for external cephalic version (ECV) using learning curve-cumulative sum (LC-CUSUM) analysis. Methods This was a retrospective study involving 290 consecutive cases between October 2013 and March 2017. We evaluated the learning curve for ECV on nulli and over para 1 group using LC-CUSUM analysis on the assumption that 50% and 70% of ECV procedures succeeded by description a trend-line of quadratic function with reliable R2 values. Results The overall success rate for ECV was 64.8% (188/290), while the success rate for nullipara and over para 1 groups was 56.2% (100/178) and 78.6% (88/112), respectively. ‘H’ value, that the actual failure rate does not differ from the acceptable failure rate, was −3.27 and −1.635 when considering ECV success rates of 50% and 70%, respectively. Consequently, in order to obtain a consistent 50% success rate, we would require 57 nullipara cases, and in order to obtain a consistent 70% success rate, we would require 130 nullipara cases. In contrast, 8 to 10 over para 1 cases would be required for an expected success rate of 50% and 70% on over para 1 group. Conclusion Even a relatively inexperienced physician can experience success with multipara and after accumulating experience, they will manage nullipara cases. Further research is required for LC-CUSUM involving several practitioners instead of a single practitioner. This will lead to the gradual implementation of standard learning curve guidelines for ECV. PMID:28791265

A Comparison of a Machine Learning Model with EuroSCORE II in Predicting Mortality after Elective Cardiac Surgery: A Decision Curve Analysis

PubMed Central

Allyn, Jérôme; Allou, Nicolas; Augustin, Pascal; Philip, Ivan; Martinet, Olivier; Belghiti, Myriem; Provenchere, Sophie; Montravers, Philippe; Ferdynus, Cyril

2017-01-01

Background The benefits of cardiac surgery are sometimes difficult to predict and the decision to operate on a given individual is complex. Machine Learning and Decision Curve Analysis (DCA) are recent methods developed to create and evaluate prediction models. Methods and finding We conducted a retrospective cohort study using a prospective collected database from December 2005 to December 2012, from a cardiac surgical center at University Hospital. The different models of prediction of mortality in-hospital after elective cardiac surgery, including EuroSCORE II, a logistic regression model and a machine learning model, were compared by ROC and DCA. Of the 6,520 patients having elective cardiac surgery with cardiopulmonary bypass, 6.3% died. Mean age was 63.4 years old (standard deviation 14.4), and mean EuroSCORE II was 3.7 (4.8) %. The area under ROC curve (IC95%) for the machine learning model (0.795 (0.755–0.834)) was significantly higher than EuroSCORE II or the logistic regression model (respectively, 0.737 (0.691–0.783) and 0.742 (0.698–0.785), p < 0.0001). Decision Curve Analysis showed that the machine learning model, in this monocentric study, has a greater benefit whatever the probability threshold. Conclusions According to ROC and DCA, machine learning model is more accurate in predicting mortality after elective cardiac surgery than EuroSCORE II. These results confirm the use of machine learning methods in the field of medical prediction. PMID:28060903

An appraisal of the learning curve in robotic general surgery.

PubMed

Pernar, Luise I M; Robertson, Faith C; Tavakkoli, Ali; Sheu, Eric G; Brooks, David C; Smink, Douglas S

2017-11-01

Robotic-assisted surgery is used with increasing frequency in general surgery for a variety of applications. In spite of this increase in usage, the learning curve is not yet defined. This study reviews the literature on the learning curve in robotic general surgery to inform adopters of the technology. PubMed and EMBASE searches yielded 3690 abstracts published between July 1986 and March 2016. The abstracts were evaluated based on the following inclusion criteria: written in English, reporting original work, focus on general surgery operations, and with explicit statistical methods. Twenty-six full-length articles were included in final analysis. The articles described the learning curves in colorectal (9 articles, 35%), foregut/bariatric (8, 31%), biliary (5, 19%), and solid organ (4, 15%) surgery. Eighteen of 26 (69%) articles report single-surgeon experiences. Time was used as a measure of the learning curve in all studies (100%); outcomes were examined in 10 (38%). In 12 studies (46%), the authors identified three phases of the learning curve. Numbers of cases needed to achieve plateau performance were wide-ranging but overlapping for different kinds of operations: 19-128 cases for colorectal, 8-95 for foregut/bariatric, 20-48 for biliary, and 10-80 for solid organ surgery. Although robotic surgery is increasingly utilized in general surgery, the literature provides few guidelines on the learning curve for adoption. In this heterogeneous sample of reviewed articles, the number of cases needed to achieve plateau performance varies by case type and the learning curve may have multiple phases as surgeons add more complex cases to their case mix with growing experience. Time is the most common determinant for the learning curve. The literature lacks a uniform assessment of outcomes and complications, which would arguably reflect expertise in a more meaningful way than time to perform the operation alone.

Autoshaped head poking in the mouse: a quantitative analysis of the learning curve.

PubMed

Papachristos, Efstathios B; Gallistel, C R

2006-05-01

In autoshaping experiments, we quantified the acquisition of anticipatory head poking in individual mice, using an algorithm that finds changes in the slope of a cumulative record. In most mice, upward changes in the amount of anticipatory poking per trial were abrupt, and tended to occur at session boundaries, suggesting that the session is as significant a unit of experience as the trial. There were large individual differences in the latency to the onset of vigorous responding. "Asymptotic" performance was unstable; large, bidirectional, and relatively enduring changes were common. Given the characteristics of the individual learning curves, it is unlikely that physiologically meaningful estimates of rate of learning can be extracted from group-average learning curves.

Autoshaped Head Poking in the Mouse: A Quantitative Analysis of the Learning Curve

PubMed Central

Papachristos, Efstathios B; Gallistel, C.R

2006-01-01

In autoshaping experiments, we quantified the acquisition of anticipatory head poking in individual mice, using an algorithm that finds changes in the slope of a cumulative record. In most mice, upward changes in the amount of anticipatory poking per trial were abrupt, and tended to occur at session boundaries, suggesting that the session is as significant a unit of experience as the trial. There were large individual differences in the latency to the onset of vigorous responding. “Asymptotic” performance was unstable; large, bidirectional, and relatively enduring changes were common. Given the characteristics of the individual learning curves, it is unlikely that physiologically meaningful estimates of rate of learning can be extracted from group-average learning curves. PMID:16776053

Learning Curve of the Application of Huang Three-Step Maneuver in a Laparoscopic Spleen-Preserving Splenic Hilar Lymphadenectomy for Advanced Gastric Cancer

PubMed Central

Huang, Ze-Ning; Huang, Chang-Ming; Zheng, Chao-Hui; Li, Ping; Xie, Jian-Wei; Wang, Jia-Bin; Lin, Jian-Xian; Lu, Jun; Chen, Qi-Yue; Cao, Long-long; Lin, Mi; Tu, Ru-Hong

2016-01-01

Abstract To investigate the learning curve of the application of Huang 3-step maneuver, which was summarized and proposed by our center for the treatment of advanced upper gastric cancer. From April 2012 to March 2013, 130 consecutive patients who underwent a laparoscopic spleen-preserving splenic hilar lymphadenectomy (LSPL) by a single surgeon who performed Huang 3-step maneuver were retrospectively analyzed. The learning curve was analyzed based on the moving average (MA) method and the cumulative sum method (CUSUM). Surgical outcomes, short-term outcomes, and follow-up results before and after learning curve were contrastively analyzed. A stepwise multivariate logistic regression was used for a multivariable analysis to determine the factors that affect the operative time using Huang 3-step maneuver. Based on the CUSUM, the learning curve for Huang 3-step maneuver was divided into phase 1 (cases 1–40) and phase 2 (cases 41–130). The dissection time (DT) (P < 0.001), blood loss (BL) (P < 0.001), and number of vessels injured in phase 2 were significantly less than those in phase 1. There were no significant differences in the clinicopathological characteristics, short-term outcomes, or major postoperative complications between the learning curve phases. Univariate and multivariate analyses revealed that body mass index (BMI), short gastric vessels (SGVs), splenic hilar artery (SpA) type, and learning curve phase were significantly associated with DT. In the entire group, 124 patients were followed for a median time of 23.0 months (range, 3–30 months). There was no significant difference in the survival curve between phases. AUGC patients with a BMI less than 25 kg/m2, a small number of SGVs, and a concentrated type of SpA are ideal candidates for surgeons who are in phase 1 of the learning curve. PMID:27043698

Urgent and Elective Robotic Single-Site Cholecystectomy: Analysis and Learning Curve of 150 Consecutive Cases.

PubMed

Kubat, Eric; Hansen, Nathan; Nguyen, Huy; Wren, Sherry M; Eisenberg, Dan

2016-03-01

The use of robotic single-site cholecystectomy has increased exponentially. There are few reports describing the safety, efficacy, and operative learning curve of robotic single-site cholecystectomy either in the community setting or with nonelective surgery. We performed a retrospective review of a prospective database of our initial experience with robotic single-site cholecystectomy. Demographics and perioperative outcomes were evaluated for both urgent and elective cholecystectomy. Cumulative sum analysis was performed to determine the surgeon's learning curve. One hundred fifty patients underwent robotic single-site cholecystectomy. Seventy-four (49.3%) patients underwent urgent robotic single-site cholecystectomy, and 76 (50.7%) underwent elective robotic single-site cholecystectomy. Mean total operative time for robotic single-site cholecystectomy was 83.3 ± 2.7 minutes. Mean operative time for the urgent cohort was significantly longer than for the elective cohort (95.0 ± 4.4 versus 71.9 ± 2.6 minutes; P < .001). There was one conversion in the urgent cohort and none in the elective cohort. There was one bile duct injury (0.7%) in the urgent cohort. Perioperative complications occurred in 8.7% of patients, and most consisted of superficial surgical-site infections. There were no incisional hernias detected. The surgeon's learning curve, inclusive of urgent and elective cases, was 48 operations. Robotic single-site cholecystectomy can be performed safely and effectively in both elective and urgent cholecystectomy with a reasonable learning curve and acceptable perioperative outcomes.

Learning curves for single incision and conventional laparoscopic right hemicolectomy: a multidimensional analysis

PubMed Central

Park, Yoonah; Yong, Yuen Geng; Jung, Kyung Uk; Huh, Jung Wook; Cho, Yong Beom; Kim, Hee Cheol; Lee, Woo Yong; Chun, Ho-Kyung

2015-01-01

Purpose This study aimed to compare the learning curves and early postoperative outcomes for conventional laparoscopic (CL) and single incision laparoscopic (SIL) right hemicolectomy (RHC). Methods This retrospective study included the initial 35 cases in each group. Learning curves were evaluated by the moving average of operative time, mean operative time of every five consecutive cases, and cumulative sum (CUSUM) analysis. The learning phase was considered overcome when the moving average of operative times reached a plateau, and when the mean operative time of every five consecutive cases reached a low point and subsequently did not vary by more than 30 minutes. Results Six patients with missing data in the CL RHC group were excluded from the analyses. According to the mean operative time of every five consecutive cases, learning phase of SIL and CL RHC was completed between 26 and 30 cases, and 16 and 20 cases, respectively. Moving average analysis revealed that approximately 31 (SIL) and 25 (CL) cases were needed to complete the learning phase, respectively. CUSUM analysis demonstrated that 10 (SIL) and two (CL) cases were required to reach a steady state of complication-free performance, respectively. Postoperative complications rate was higher in SIL than in CL group, but the difference was not statistically significant (17.1% vs. 3.4%). Conclusion The learning phase of SIL RHC is longer than that of CL RHC. Early oncological outcomes of both techniques were comparable. However, SIL RHC had a statistically insignificant higher complication rate than CL RHC during the learning phase. PMID:25960990

Applications of Adaptive Learning Controller to Synthetic Aperture Radar.

DTIC Science & Technology

1985-02-01

FIGURE 37. Location of Two Sub- Phase Histories to be Utilized in Estimating Misfocus Coefficients A and C . . . A8 FIGURES 38.-94. ALC Learning Curves ...FIGURES (Concl uded) FIGURE 23. ALC Learning Curve .... .................. ... 45 .- " FIGURE 24. ALC Learning Curve ......... ................. 47 FIGURE...25. ALC Learning Curve .... .................. ... 48 FIGURE 26. ALC Learning Curve ....... .... ... .... 50 FIGURE 27. ALC Learning Curve

Using learning curves on energy-efficient technologies to estimate future energy savings and emission reduction potentials in the U.S. iron and steel industry

DOE Office of Scientific and Technical Information (OSTI.GOV)

Karali, Nihan; Park, Won Young; McNeil, Michael A.

Increasing concerns on non-sustainable energy use and climate change spur a growing research interest in energy efficiency potentials in various critical areas such as industrial production. This paper focuses on learning curve aspects of energy efficiency measures in the U.S iron and steel sector. A number of early-stage efficient technologies (i.e., emerging or demonstration technologies) are technically feasible and have the potential to make a significant contribution to energy saving and CO 2 emissions reduction, but fall short economically to be included. However, they may also have the cost effective potential for significant cost reduction and/or performance improvement in themore » future under learning effects such as ‘learning-by-doing’. The investigation is carried out using ISEEM, a technology oriented, linear optimization model. We investigated how steel demand is balanced with/without the availability learning curve, compared to a Reference scenario. The retrofit (or investment in some cases) costs of energy efficient technologies decline in the scenario where learning curve is applied. The analysis also addresses market penetration of energy efficient technologies, energy saving, and CO 2 emissions in the U.S. iron and steel sector with/without learning impact. Accordingly, the study helps those who use energy models better manage the price barriers preventing unrealistic diffusion of energy-efficiency technologies, better understand the market and learning system involved, predict future achievable learning rates more accurately, and project future savings via energy-efficiency technologies with presence of learning. We conclude from our analysis that, most of the existing energy efficiency technologies that are currently used in the U.S. iron and steel sector are cost effective. Penetration levels increases through the years, even though there is no price reduction. However, demonstration technologies are not economically feasible in the U.S. iron and steel sector with the current cost structure. In contrast, some of the demonstration technologies are adapted in the mid-term and their penetration levels increase as the prices go down with learning curve. We also observe large penetration of 225kg pulverized coal injection with the presence of learning.« less

Evaluation of the learning curve of laparoscopic choledochal cyst excision and Roux-en-Y hepaticojejunostomy in children: CUSUM analysis of a single surgeon's experience.

PubMed

Wen, Zhe; Liang, Huiying; Liang, Jiankun; Liang, Qifeng; Xia, Huimin

2017-02-01

Laparoscopic cyst excision and Roux-en-Y hepaticojejunostomy is gaining popularity as a treatment for choledochal cyst (CDC) in children. However, the learning curve for this challenging laparoscopic procedure has not been addressed. The aim of this study is to determine the characteristics of the learning curve of this procedure. This may guide the training in institutions currently not using this technique. A prospectively collected database comprising all medical records of the first 104 consecutive patients undergoing laparoscopic CDC excision and Roux-en-Y hepaticojejunostomy performed by one surgeon was studied. Multifactorial linear/logistic regression analysis was performed to identify patient-, surgeon-, and procedure-related factors associated with operating times, rates of adverse event, and length of postoperative stay. Cumulative sum analysis demonstrated a learning curve for laparoscopic choledochal cyst excision of 37 cases. Comparing the early with the late experiences (37 vs. 67 cases), the surgeon-specific outcomes significantly improved in terms of operating times (352 vs. 240 min; P < 0.001), postoperative complication rate (13.5 vs. 1.5 %; P = 0.02), and the length of hospital stay (9.4 vs. 7.8 days; P = 0.01). After multivariate analyses, independent predictors of operating times included the completion of the learning curve (CLC) (OR 0.68, 95 % CI 0.63-0.73) and adhesion score (OR middle 1.25, 95 % CI 1.08-1.45; OR high 1.40, 95 % CI 1.20-1.62; compared with the low score); significant predictors of perioperative adverse outcomes were CLC (OR 0.07, 95 % CI 0.02-0.34) and comorbidities prior to the surgery (OR 30.65, 95 % CI 1.71-549.63). The independent predictors of length of postoperative stay included CLC, preoperative comorbidities, and perioperative adverse events. CLC for laparoscopic choledochal cyst excision is 37 cases. After CLC, not only the operative time is reduced, the complications, adverse results, and the length of hospital stay all decreased significantly. The learning curve can be used as the basis for performance guiding the training.

Learning curve in robotic rectal cancer surgery: current state of affairs.

PubMed

Jiménez-Rodríguez, Rosa M; Rubio-Dorado-Manzanares, Mercedes; Díaz-Pavón, José Manuel; Reyes-Díaz, M Luisa; Vazquez-Monchul, Jorge Manuel; Garcia-Cabrera, Ana M; Padillo, Javier; De la Portilla, Fernando

2016-12-01

Robotic-assisted rectal cancer surgery offers multiple advantages for surgeons, and it seems to yield the same clinical outcomes as regards the short-time follow-up of patients compared to conventional laparoscopy. This surgical approach emerges as a technique aiming at overcoming the limitations posed by rectal cancer and other surgical fields of difficult access, in order to obtain better outcomes and a shorter learning curve. A systematic review of the literature of robot-assisted rectal surgery was carried out according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The search was conducted in October 2015 in PubMed, MEDLINE and the Cochrane Central Register of Controlled Trials, for articles published in the last 10 years and pertaining the learning curve of robotic surgery for colorectal cancer. It consisted of the following key words: "rectal cancer/learning curve/robotic-assisted laparoscopic surgery". A total of 34 references were identified, but only 9 full texts specifically addressed the analysis of the learning curve in robot-assisted rectal cancer surgery, 7 were case series and 2 were non-randomised case-comparison series. Eight papers used the cumulative sum (CUSUM) method, and only one author divided the series into two groups to compare both. The mean number of cases for phase I of the learning curve was calculated to be 29.7 patients; phase II corresponds to a mean number 37.4 patients. The mean number of cases required for the surgeon to be classed as an expert in robotic surgery was calculated to be 39 patients. Robotic advantages could have an impact on learning curve for rectal cancer and lower the number of cases that are necessary for rectal resections.

Analysis of the learning curve for peroral endoscopic myotomy for esophageal achalasia: Single-center, two-operator experience.

PubMed

Lv, Houning; Zhao, Ningning; Zheng, Zhongqing; Wang, Tao; Yang, Fang; Jiang, Xihui; Lin, Lin; Sun, Chao; Wang, Bangmao

2017-05-01

Peroral endoscopic myotomy (POEM) has emerged as an advanced technique for the treatment of achalasia, and defining the learning curve is mandatory. From August 2011 to June 2014, two operators in our institution (A&B) carried out POEM on 35 and 33 consecutive patients, respectively. Moving average and cumulative sum (CUSUM) methods were used to analyze the POEM learning curve for corrected operative time (cOT), referring to duration of per centimeter myotomy. Additionally, perioperative outcomes were compared among distinct learning curve phases. Using the moving average method, cOT reached a plateau at the 29th case and at the 24th case for operators A and B, respectively. CUSUM analysis identified three phases: initial learning period (Phase 1), efficiency period (Phase 2) and mastery period (Phase 3). The relatively smooth state in the CUSUM graph occurred at the 26th case and at the 24th case for operators A and B, respectively. Mean cOT of distinct phases for operator A were 8.32, 5.20 and 3.97 min, whereas they were 5.99, 3.06 and 3.75 min for operator B, respectively. Eckardt score and lower esophageal sphincter pressure significantly decreased during the 1-year follow-up period. Data were comparable regarding patient characteristics and perioperative outcomes. This single-center study demonstrated that expert endoscopists with experience in esophageal endoscopic submucosal dissection reached a plateau in learning of POEM after approximately 25 cases. © 2016 Japan Gastroenterological Endoscopy Society.

The learning curve: Implications of a quantitative analysis

PubMed Central

Gallistel, Charles R.; Fairhurst, Stephen; Balsam, Peter

2004-01-01

The negatively accelerated, gradually increasing learning curve is an artifact of group averaging in several commonly used basic learning paradigms (pigeon autoshaping, delay- and trace-eye-blink conditioning in the rabbit and rat, autoshaped hopper entry in the rat, plus maze performance in the rat, and water maze performance in the mouse). The learning curves for individual subjects show an abrupt, often step-like increase from the untrained level of responding to the level seen in the well trained subject. The rise is at least as abrupt as that commonly seen in psychometric functions in stimulus detection experiments. It may indicate that the appearance of conditioned behavior is mediated by an evidence-based decision process, as in stimulus detection experiments. If the appearance of conditioned behavior is taken instead to reflect the increase in an underlying associative strength, then a negligible portion of the function relating associative strength to amount of experience is behaviorally visible. Consequently, rate of learning cannot be estimated from the group-average curve; the best measure is latency to the onset of responding, determined for each subject individually. PMID:15331782

The learning curve: implications of a quantitative analysis.

PubMed

Gallistel, Charles R; Fairhurst, Stephen; Balsam, Peter

2004-09-07

The negatively accelerated, gradually increasing learning curve is an artifact of group averaging in several commonly used basic learning paradigms (pigeon autoshaping, delay- and trace-eye-blink conditioning in the rabbit and rat, autoshaped hopper entry in the rat, plus maze performance in the rat, and water maze performance in the mouse). The learning curves for individual subjects show an abrupt, often step-like increase from the untrained level of responding to the level seen in the well trained subject. The rise is at least as abrupt as that commonly seen in psychometric functions in stimulus detection experiments. It may indicate that the appearance of conditioned behavior is mediated by an evidence-based decision process, as in stimulus detection experiments. If the appearance of conditioned behavior is taken instead to reflect the increase in an underlying associative strength, then a negligible portion of the function relating associative strength to amount of experience is behaviorally visible. Consequently, rate of learning cannot be estimated from the group-average curve; the best measure is latency to the onset of responding, determined for each subject individually.

An Exploratory Analysis of Personality, Attitudes, and Study Skills on the Learning Curve within a Team-based Learning Environment

PubMed Central

Henry, Teague; Campbell, Ashley

2015-01-01

Objective. To examine factors that determine the interindividual variability of learning within a team-based learning environment. Methods. Students in a pharmacokinetics course were given 4 interim, low-stakes cumulative assessments throughout the semester and a cumulative final examination. Students’ Myers-Briggs personality type was assessed, as well as their study skills, motivations, and attitudes towards team-learning. A latent curve model (LCM) was applied and various covariates were assessed to improve the regression model. Results. A quadratic LCM was applied for the first 4 assessments to predict final examination performance. None of the covariates examined significantly impacted the regression model fit except metacognitive self-regulation, which explained some of the variability in the rate of learning. There were some correlations between personality type and attitudes towards team learning, with introverts having a lower opinion of team-learning than extroverts. Conclusion. The LCM could readily describe the learning curve. Extroverted and introverted personality types had the same learning performance even though preference for team-learning was lower in introverts. Other personality traits, study skills, or practice did not significantly contribute to the learning variability in this course. PMID:25861101

An exploratory analysis of personality, attitudes, and study skills on the learning curve within a team-based learning environment.

PubMed

Persky, Adam M; Henry, Teague; Campbell, Ashley

2015-03-25

To examine factors that determine the interindividual variability of learning within a team-based learning environment. Students in a pharmacokinetics course were given 4 interim, low-stakes cumulative assessments throughout the semester and a cumulative final examination. Students' Myers-Briggs personality type was assessed, as well as their study skills, motivations, and attitudes towards team-learning. A latent curve model (LCM) was applied and various covariates were assessed to improve the regression model. A quadratic LCM was applied for the first 4 assessments to predict final examination performance. None of the covariates examined significantly impacted the regression model fit except metacognitive self-regulation, which explained some of the variability in the rate of learning. There were some correlations between personality type and attitudes towards team learning, with introverts having a lower opinion of team-learning than extroverts. The LCM could readily describe the learning curve. Extroverted and introverted personality types had the same learning performance even though preference for team-learning was lower in introverts. Other personality traits, study skills, or practice did not significantly contribute to the learning variability in this course.

Concentrated photovoltaics system costs and learning curve analysis

NASA Astrophysics Data System (ADS)

Haysom, Joan E.; Jafarieh, Omid; Anis, Hanan; Hinzer, Karin

2013-09-01

An extensive set of costs in /W for the installed costs of CPV systems has been amassed from a range of public sources, including both individual company prices and market reports. Cost reductions over time are very evident, with current prices for 2012 in the range of 3.0 ± 0.7 /W and a predicted cost of 1.5 /W for 2020. Cost data is combined with deployment volumes in a learning curve analysis, providing a fitted learning rate of either 18.5% or 22.3% depending on the methodology. This learning rate is compared to that of PV modules and PV installed systems, and the influence of soft costs is discussed. Finally, if an annual growth rate of 39% is assumed for deployed volumes, then, using the learning rate of 20%, this would predict the achievement of a cost point of 1.5 /W by 2016.

Understanding and Taking Control of Surgical Learning Curves.

PubMed

Gofton, Wade T; Papp, Steven R; Gofton, Tyson; Beaulé, Paul E

2016-01-01

As surgical techniques continue to evolve, surgeons will have to integrate new skills into their practice. A learning curve is associated with the integration of any new procedure; therefore, it is important for surgeons who are incorporating a new technique into their practice to understand what the reported learning curve might mean for them and their patients. A learning curve should not be perceived as negative because it can indicate progress; however, surgeons need to understand how to optimize the learning curve to ensure progress with minimal patient risk. It is essential for surgeons who are implementing new procedures or skills to define potential learning curves, examine how a reported learning curve may relate to an individual surgeon's in-practice learning and performance, and suggest methods in which an individual surgeon can modify his or her specific learning curve in order to optimize surgical outcomes and patient safety. A defined personal learning contract may be a practical method for surgeons to proactively manage their individual learning curve and provide evidence of their efforts to safely improve surgical practice.

The training and learning process of transseptal puncture using a modified technique.

PubMed

Yao, Yan; Ding, Ligang; Chen, Wensheng; Guo, Jun; Bao, Jingru; Shi, Rui; Huang, Wen; Zhang, Shu; Wong, Tom

2013-12-01

As the transseptal (TS) puncture has become an integral part of many types of cardiac interventional procedures, its technique that was initial reported for measurement of left atrial pressure in 1950s, continue to evolve. Our laboratory adopted a modified technique which uses only coronary sinus catheter as the landmark to accomplishing TS punctures under fluoroscopy. The aim of this study is prospectively to evaluate the training and learning process for TS puncture guided by this modified technique. Guided by the training protocol, TS puncture was performed in 120 consecutive patients by three trainees without previous personal experience in TS catheterization and one experienced trainer as a controller. We analysed the following parameters: one puncture success rate, total procedure time, fluoroscopic time, and radiation dose. The learning curve was analysed using curve-fitting methodology. The first attempt at TS crossing was successful in 74 (82%), a second attempt was successful in 11 (12%), and 5 patients failed to puncture the interatrial septal finally. The average starting process time was 4.1 ± 0.8 min, and the estimated mean learning plateau was 1.2 ± 0.2 min. The estimated mean learning rate for process time was 25 ± 3 cases. Important aspects of learning curve can be estimated by fitting inverse curves for TS puncture. The study demonstrated that this technique was a simple, safe, economic, and effective approach for learning of TS puncture. Base on the statistical analysis, approximately 29 TS punctures will be needed for trainee to pass the steepest area of learning curve.

Predicting Robust Learning with the Visual Form of the Moment-by-Moment Learning Curve

ERIC Educational Resources Information Center

Baker, Ryan S.; Hershkovitz, Arnon; Rossi, Lisa M.; Goldstein, Adam B.; Gowda, Sujith M.

2013-01-01

We present a new method for analyzing a student's learning over time for a specific skill: analysis of the graph of the student's moment-by-moment learning over time. Moment-by-moment learning is calculated using a data-mined model that assesses the probability that a student learned a skill or concept at a specific time during learning (Baker,…

A learning curve for solar thermal power

NASA Astrophysics Data System (ADS)

Platzer, Werner J.; Dinter, Frank

2016-05-01

Photovoltaics started its success story by predicting the cost degression depending on cumulated installed capacity. This so-called learning curve was published and used for predictions for PV modules first, then predictions of system cost decrease also were developed. This approach is less sensitive to political decisions and changing market situations than predictions on the time axis. Cost degression due to innovation, use of scaling effects, improved project management, standardised procedures including the search for better sites and optimization of project size are learning effects which can only be utilised when projects are developed. Therefore a presentation of CAPEX versus cumulated installed capacity is proposed in order to show the possible future advancement of the technology to politics and market. However from a wide range of publications on cost for CSP it is difficult to derive a learning curve. A logical cost structure for direct and indirect capital expenditure is needed as the basis for further analysis. Using derived reference cost for typical power plant configurations predictions of future cost have been derived. Only on the basis of that cost structure and the learning curve levelised cost of electricity for solar thermal power plants should be calculated for individual projects with different capacity factors in various locations.

Guidelines for application of learning/cost improvement curves

NASA Technical Reports Server (NTRS)

Delionback, L. M.

1975-01-01

The differences between the terms learning curve and improvement curve are noted, as well as the differences between the Wright system and the Crawford system. Learning curve computational techniques were reviewed along with a method to arrive at a composite learning curve for a system given detail curves either by the functional techniques classification or simply categorized by subsystem. Techniques are discussed for determination of the theoretical first unit (TFU) cost using several of the currently accepted methods. Sometimes TFU cost is referred to as simply number one cost. A tabular presentation of the various learning curve slope values is given. A discussion of the various trends in the application of learning/improvement curves and an outlook for the future are presented.

Learning Curve and Clinical Outcomes of Performing Surgery with the InterTan Intramedullary Nail in Treating Femoral Intertrochanteric Fractures

PubMed Central

2017-01-01

Purpose. The purpose of this study is to evaluate the learning curve of performing surgery with the InterTan intramedullary nail in treating femoral intertrochanteric fractures, to provide valuable information and experience for surgeons who decide to learn a new procedure. Methods. We retrospectively analyzed data from 53 patients who underwent surgery using an InterTan intramedullary nail at our hospital between July 2012 and September 2015. The negative exponential curve-fit regression analysis was used to evaluate the learning curve. According to 90% learning milestone, patients were divided into two group, and the outcomes were compared. Results. The mean operative time was 69.28 (95% CI 64.57 to 74.00) minutes; with the accumulation of surgical experience, the operation time was gradually decreased. 90% of the potential improvement was expected after 18 cases. In terms of operative time, intraoperative blood loss, hospital stay, and Harris hip score significant differences were found between two groups (p = 0.009, p = 0.000, p = 0.030, and p = 0.002, resp.). Partial weight bearing time, fracture union time, tip apex distance, and the number of blood transfusions and complications were similar between two groups (p > 0.5). Conclusion. This study demonstrated that the learning curve of performing surgery with the InterTan intramedullary nail is acceptable and 90% of the expert's proficiency level is achieved at around 18 cases. PMID:28503572

Establishing the Learning Curve of Robotic Sacral Colpopexy in a Start-up Robotics Program.

PubMed

Sharma, Shefali; Calixte, Rose; Finamore, Peter S

2016-01-01

To determine the learning curve of the following segments of a robotic sacral colpopexy: preoperative setup, operative time, postoperative transition, and room turnover. A retrospective cohort study to determine the number of cases needed to reach points of efficiency in the various segments of a robotic sacral colpopexy (Canadian Task Force II-2). A university-affiliated community hospital. Women who underwent robotic sacral colpopexy at our institution from 2009 to 2013 comprise the study population. Patient characteristics and operative reports were extracted from a patient database that has been maintained since the inception of the robotics program at Winthrop University Hospital and electronic medical records. Based on additional procedures performed, 4 groups of patients were created (A-D). Learning curves for each of the segment times of interest were created using penalized basis spline (B-spline) regression. Operative time was further analyzed using an inverse curve and sequential grouping. A total of 176 patients were eligible. Nonparametric tests detected no difference in procedure times between the 4 groups (A-D) of patients. The preoperative and postoperative points of efficiency were 108 and 118 cases, respectively. The operative points of proficiency and efficiency were 25 and 36 cases, respectively. Operative time was further analyzed using an inverse curve that revealed that after 11 cases the surgeon had reached 90% of the learning plateau. Sequential grouping revealed no significant improvement in operative time after 60 cases. Turnover time could not be assessed because of incomplete data. There is a difference in the operative time learning curve for robotic sacral colpopexy depending on the statistical analysis used. The learning curve of the operative segment showed an improvement in operative time between 25 and 36 cases when using B-spline regression. When the data for operative time was fit to an inverse curve, a learning rate of 11 cases was appreciated. Using sequential grouping to describe the data, no improvement in operative time was seen after 60 cases. Ultimately, we believe that efficiency in operative time is attained after 30 to 60 cases when performing robotic sacral colpopexy. The learning curve for preoperative setup and postoperative transition, which is reflective of anesthesia and nursing staff, was approximately 110 cases. Copyright © 2016 AAGL. Published by Elsevier Inc. All rights reserved.

Robotic partial nephrectomy - Evaluation of the impact of case mix on the procedural learning curve.

PubMed

Roman, A; Ahmed, K; Challacombe, B

2016-05-01

Although Robotic partial nephrectomy (RPN) is an emerging technique for the management of small renal masses, this approach is technically demanding. To date, there is limited data on the nature and progression of the learning curve in RPN. To analyse the impact of case mix on the RPN LC and to model the learning curve. The records of the first 100 RPN performed, were analysed at our institution that were carried out by a single surgeon (B.C) (June 2010-December 2013). Cases were split based on their Preoperative Aspects and Dimensions Used for an Anatomical (PADUA) score into the following groups: 6-7, 8-9 and >10. Using a split group (20 patients in each group) and incremental analysis, the mean, the curve of best fit and R(2) values were calculated for each group. Of 100 patients (F:28, M:72), the mean age was 56.4 ± 11.9 years. The number of patients in each PADUA score groups: 6-7, 8-9 and >10 were 61, 32 and 7 respectively. An increase in incidence of more complex cases throughout the cohort was evident within the 8-9 group (2010: 1 case, 2013: 16 cases). The learning process did not significantly affect the proxies used to assess surgical proficiency in this study (operative time and warm ischaemia time). Case difficulty is an important parameter that should be considered when evaluating procedural learning curves. There is not one well fitting model that can be used to model the learning curve. With increasing experience, clinicians tend to operate on more difficult cases. Copyright © 2016 IJS Publishing Group Ltd. Published by Elsevier Ltd. All rights reserved.

Colorectal endoscopic submucosal dissection (ESD) in the West - when can satisfactory results be obtained? A single-operator learning curve analysis.

PubMed

Spychalski, Michał; Skulimowski, Aleksander; Dziki, Adam; Saito, Yutaka

2017-12-01

Up to date we lack a detailed description of the colorectal endoscopic submucosal dissection (ESD) learning curve, that would represent the experience of the Western center. The aim of this study was to define the critical points of the learning curve and to draw up lesions qualification guidelines tailored to the endoscopists experience. We have carried out a single center prospective study. Between June 2013 and December 2016, 228 primary colorectal lesions were managed by ESD procedure. In order to create a learning curve model and to carry out the analysis the cases were divided into six periods, each consisting of 38 cases. The overall en bloc resection rate was 79.39%. The lowest en bloc resection rate (52.36%) was observed in the first period. After completing 76 procedures, the resection rate surged to 86% and it was accompanied by the significant increase in the mean procedure speed of ≥9 cm 2 /h. Lesions localization and diameter had a signification impact on the outcomes. After 76 procedures, en bloc resection rate of 90.9 and 90.67% were achieved for the left side of colon and rectum, respectively. In the right side of colon statistically significant lower resection rate of 67.57% was observed. We have proved that in the setting of the Western center, colorectal ESD can yield excellent results. It seems that the key to the success during the learning period is 'tailoring' lesions qualification guidelines to the experience of the endoscopist, as lesions diameter and localization highly influence the outcomes.

A Growth Curve Analysis of Novel Word Learning by Sequential Bilingual Preschool Children

ERIC Educational Resources Information Center

Kan, Pui Fong; Kohnert, Kathryn

2012-01-01

Longitudinal word learning studies which control for experience can advance understanding of language learning and potential intra- and inter-language relationships in developing bilinguals. We examined novel word learning in both the first (L1) and the second (L2) languages of bilingual children. The rate and shape of change as well as the role…

[Individual learning curve for radical robot-assisted prostatectomy based on the example of three professionals working in one clinic].

PubMed

Rasner, P I; Pushkar', D Iu; Kolontarev, K B; Kotenkov, D V

2014-01-01

The appearance of new surgical technique always requires evaluation of its effectiveness and ease of acquisition. A comparative study of the results of the first three series of successive robot-assisted radical prostatectomy (RARP) performed on at time by three surgeons, was conducted. The series consisted of 40 procedures, and were divided into 4 groups of 10 operations for the analysis. When comparing data, statistically significant improvement of intra- and postoperative performance in each series was revealed, with increase in the number of operations performed, and in each subsequent series compared with the preceding one. We recommend to perform the planned conversion at the first operation. In our study, previous laparoscopic experience did not provide any significant advantages in the acquisition of robot-assisted technology. To characterize the individual learning curve, we recommend the use of the number of operations that the surgeon looked in the life-surgery regimen and/or in which he participated as an assistant before his own surgical activity, as well as the indicator "technical defect". In addition to the term "individual learning curve", we propose to introduce the terms "surgeon's individual training phase", and "clinic's learning curve".

The learning curve to achieve satisfactory completion rates in upper GI endoscopy: an analysis of a national training database.

PubMed

Ward, S T; Hancox, A; Mohammed, M A; Ismail, T; Griffiths, E A; Valori, R; Dunckley, P

2017-06-01

The aim of this study was to determine the number of OGDs (oesophago-gastro-duodenoscopies) trainees need to perform to acquire competency in terms of successful unassisted completion to the second part of the duodenum 95% of the time. OGD data were retrieved from the trainee e-portfolio developed by the Joint Advisory Group on GI Endoscopy (JAG) in the UK. All trainees were included unless they were known to have a baseline experience of >20 procedures or had submitted data for <20 procedures. The primary outcome measure was OGD completion, defined as passage of the endoscope to the second part of the duodenum without physical assistance. The number of OGDs required to achieve a 95% completion rate was calculated by the moving average method and learning curve cumulative summation (LC-Cusum) analysis. To determine which factors were independently associated with OGD completion, a mixed effects logistic regression model was constructed with OGD completion as the outcome variable. Data were analysed for 1255 trainees over 288 centres, representing 243 555 OGDs. By moving average method, trainees attained a 95% completion rate at 187 procedures. By LC-Cusum analysis, after 200 procedures, >90% trainees had attained a 95% completion rate. Total number of OGDs performed, trainee age and experience in lower GI endoscopy were factors independently associated with OGD completion. There are limited published data on the OGD learning curve. This is the largest study to date analysing the learning curve for competency acquisition. The JAG competency requirement for 200 procedures appears appropriate. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/.

Analysis of the vitreoretinal surgery learning curve.

PubMed

Martín-Avià, J; Romero-Aroca, P

2017-06-01

To describe intra- and post-operative complications, as well as the evolution of the surgical technique in first 4years of work of a novice retina surgeon, and evaluate minimal learning time required to reduce its complications, deciding which pathologies should still be referred to higher level hospitals, until further experience may be achieved. A study was conducted on patients that had undergone vitreoretinal surgery by a novice surgeon in Tarragona between 23rd October 2007 and 31st December 2011. The primary diagnosis, surgeon learning time, surgical technique, intra-operative and post-operative complications were recorded. A total of 247 surgeries were studied. The percentage of use of 20G and 23G calibres during the time, marks a change towards trans-conjunctival surgery from the ninth trimester (98 surgeries). Surgical complications decreased towards twelfth trimester (130 surgeries) with an increase in the previous months. The shift towards 23G technique around 100 surgeries is interpreted as greater comfort and safety by the surgeon. Increased surgical complications during the following months until its decline around 130 surgeries can be interpreted as an 'overconfidence'. It is arguable that the learning curve is slower than what the surgeon believes. An individual analysis of the complications and surgical outcomes is recommended to ascertain the status of the learning curve. Copyright © 2016 Sociedad Española de Oftalmología. Publicado por Elsevier España, S.L.U. All rights reserved.

Placement of central venous port catheters and peripherally inserted central catheters in the routine clinical setting of a radiology department: analysis of costs and intervention duration learning curve.

PubMed

Rotzinger, Roman; Gebauer, Bernhard; Schnapauff, Dirk; Streitparth, Florian; Wieners, Gero; Grieser, Christian; Freyhardt, Patrick; Hamm, Bernd; Maurer, Martin H

2017-12-01

Background Placement of central venous port catheters (CVPS) and peripherally inserted central catheters (PICC) is an integral component of state-of-the-art patient care. In the era of increasing cost awareness, it is desirable to have more information to comprehensively assess both procedures. Purpose To perform a retrospective analysis of interventional radiologic implantation of CVPS and PICC lines in a large patient population including a cost analysis of both methods as well as an investigation the learning curve in terms of the interventions' durations. Material and Methods All CVPS and PICC line related interventions performed in an interventional radiology department during a three-year period from January 2011 to December 2013 were examined. Documented patient data included sex, venous access site, and indication for CVPS or PICC placement. A cost analysis including intervention times was performed based on the prorated costs of equipment use, staff costs, and expenditures for disposables. The decrease in intervention duration in the course of time conformed to the learning curve. Results In total, 2987 interventions were performed by 16 radiologists: 1777 CVPS and 791 PICC lines. An average implantation took 22.5 ± 0.6 min (CVPS) and 10.1 ± 0.9 min (PICC lines). For CVPS, this average time was achieved by seven radiologists newly learning the procedures after performing 20 CVPS implantations. Total costs per implantation were €242 (CVPS) and €201 (PICC lines). Conclusion Interventional radiologic implantations of CVPS and PICC lines are well-established procedures, easy to learn by residents, and can be implanted at low costs.

Learning curve analysis of mitral valve repair using telemanipulative technology.

PubMed

Charland, Patrick J; Robbins, Tom; Rodriguez, Evilio; Nifong, Wiley L; Chitwood, Randolph W

2011-08-01

To determine if the time required to perform mitral valve repairs using telemanipulation technology decreases with experience and how that decrease is influenced by patient and procedure variables. A single-center retrospective review was conducted using perioperative and outcomes data collected contemporaneously on 458 mitral valve repair surgeries using telemanipulative technology. A regression model was constructed to assess learning with this technology and predict total robot time using multiple predictive variables. Statistical analysis was used to determine if models were significantly useful, to rule out correlation between predictor variables, and to identify terms that did not contribute to the prediction of total robot time. We found a statistically significant learning curve (P < .01). The institutional learning percentage∗ derived from total robot times† for the first 458 recorded cases of mitral valve repair using telemanipulative technology is 95% (R(2) = .40). More than one third of the variability in total robot time can be explained through our model using the following variables: type of repair (chordal procedures, ablations, and leaflet resections), band size, use of clips alone in band implantation, and the presence of a fellow at bedside (P < .01). Learning in mitral valve repair surgery using telemanipulative technology occurs at the East Carolina Heart Institute according to a logarithmic curve, with a learning percentage of 95%. From our regression output, we can make an approximate prediction of total robot time using an additive model. These metrics can be used by programs for benchmarking to manage the implementation of this new technology, as well as for capacity planning, scheduling, and capital budget analysis. Copyright © 2011 The American Association for Thoracic Surgery. All rights reserved.

Huang's three-step maneuver shortens the learning curve of laparoscopic spleen-preserving splenic hilar lymphadenectomy.

PubMed

Huang, Chang-Ming; Huang, Ze-Ning; Zheng, Chao-Hui; Li, Ping; Xie, Jian-Wei; Wang, Jia-Bin; Lin, Jian-Xian; Jun, Lu; Chen, Qi-Yue; Cao, Long-Long; Lin, Mi; Tu, Ru-Hong

2017-12-01

The goal of this study was to investigate the difference between the learning curves of different maneuvers in laparoscopic spleen-preserving splenic hilar lymphadenectomy for advanced upper gastric cancer. From January 2010 to April 2014, 53 consecutive patients who underwent laparoscopic spleen-preserving splenic hilar lymphadenectomy via the traditional-step maneuver (group A) and 53 consecutive patients via Huang's three-step maneuver (group B) were retrospectively analyzed. No significant difference in patient characteristics were found between the two groups. The learning curves of groups A and B were divided into phase 1 (1-43 cases and 1-30 cases, respectively) and phase 2 (44-53 cases and 31-53 cases, respectively). Compared with group A, the dissection time, bleeding loss and vascular injury were significantly decreased in group B. No significant differences in short-term outcomes were found between the two maneuvers. The multivariate analysis indicated that the body mass index, short gastric vessels, splenic artery type and maneuver were significantly associated with the dissection time in group B. No significant difference in the survival curve was found between the maneuvers. The learning curve of Huang's three-step maneuver was shorter than that of the traditional-step maneuver, and the former represents an ideal maneuver for laparoscopic spleen-preserving splenic hilar lymphadenectomy.To shorten the learning curve at the beginning of laparoscopic spleen-preserving splenic hilar lymphadenectomy, beginners should beneficially use Huang's three-step maneuver and select patients with advanced upper gastric cancer with a body mass index of less than 25 kg/m 2 and the concentrated type of splenic artery. Copyright © 2017. Published by Elsevier Ltd.

Overall Memory Impairment Identification with Mathematical Modeling of the CVLT-II Learning Curve in Multiple Sclerosis

PubMed Central

Stepanov, Igor I.; Abramson, Charles I.; Hoogs, Marietta; Benedict, Ralph H. B.

2012-01-01

The CVLT-II provides standardized scores for each of the List A five learning trials, so that the clinician can compare the patient's raw trials 1–5 scores with standardized ones. However, frequently, a patient's raw scores fluctuate making a proper interpretation difficult. The CVLT-II does not offer any other methods for classifying a patient's learning and memory status on the background of the learning curve. The main objective of this research is to illustrate that discriminant analysis provides an accurate assessment of the learning curve, if suitable predictor variables are selected. Normal controls were ninety-eight healthy volunteers (78 females and 20 males). A group of MS patients included 365 patients (266 females and 99 males) with clinically defined multiple sclerosis. We show that the best predictor variables are coefficients B3 and B4 of our mathematical model B3 ∗ exp(−B2  ∗  (X − 1)) + B4  ∗  (1 − exp(−B2  ∗  (X − 1))) because discriminant functions, calculated separately for B3 and B4, allow nearly 100% correct classification. These predictors allow identification of separate impairment of readiness to learn or ability to learn, or both. PMID:22745911

Overall Memory Impairment Identification with Mathematical Modeling of the CVLT-II Learning Curve in Multiple Sclerosis.

PubMed

Stepanov, Igor I; Abramson, Charles I; Hoogs, Marietta; Benedict, Ralph H B

2012-01-01

The CVLT-II provides standardized scores for each of the List A five learning trials, so that the clinician can compare the patient's raw trials 1-5 scores with standardized ones. However, frequently, a patient's raw scores fluctuate making a proper interpretation difficult. The CVLT-II does not offer any other methods for classifying a patient's learning and memory status on the background of the learning curve. The main objective of this research is to illustrate that discriminant analysis provides an accurate assessment of the learning curve, if suitable predictor variables are selected. Normal controls were ninety-eight healthy volunteers (78 females and 20 males). A group of MS patients included 365 patients (266 females and 99 males) with clinically defined multiple sclerosis. We show that the best predictor variables are coefficients B3 and B4 of our mathematical model B3 ∗ exp(-B2  ∗  (X - 1)) + B4  ∗  (1 - exp(-B2  ∗  (X - 1))) because discriminant functions, calculated separately for B3 and B4, allow nearly 100% correct classification. These predictors allow identification of separate impairment of readiness to learn or ability to learn, or both.

Implementation Learning and Forgetting Curve to Scheduling in Garment Industry

NASA Astrophysics Data System (ADS)

Muhamad Badri, Huda; Deros, Baba Md; Syahri, M.; Saleh, Chairul; Fitria, Aninda

2016-02-01

The learning curve shows the relationship between time and the cumulative number of units produced which using the mathematical description on the performance of workers in performing repetitive works. The problems of this study is level differences in the labors performance before and after the break which affects the company's production scheduling. The study was conducted in the garment industry, which the aims is to predict the company production scheduling using the learning curve and forgetting curve. By implementing the learning curve and forgetting curve, this paper contributes in improving the labors performance that is in line with the increase in maximum output 3 hours productive before the break are 15 unit product with learning curve percentage in the company is 93.24%. Meanwhile, the forgetting curve improving maximum output 3 hours productive after the break are 11 unit product with the percentage of forgetting curve in the company is 92.96%. Then, the obtained 26 units product on the productive hours one working day is used as the basic for production scheduling.

The Predictive Value of Ultrasound Learning Curves Across Simulated and Clinical Settings.

PubMed

Madsen, Mette E; Nørgaard, Lone N; Tabor, Ann; Konge, Lars; Ringsted, Charlotte; Tolsgaard, Martin G

2017-01-01

The aim of the study was to explore whether learning curves on a virtual-reality (VR) sonographic simulator can be used to predict subsequent learning curves on a physical mannequin and learning curves during clinical training. Twenty midwives completed a simulation-based training program in transvaginal sonography. The training was conducted on a VR simulator as well as on a physical mannequin. A subgroup of 6 participants underwent subsequent clinical training. During each of the 3 steps, the participants' performance was assessed using instruments with established validity evidence, and they advanced to the next level only after attaining predefined levels of performance. The number of repetitions and time needed to achieve predefined performance levels were recorded along with the performance scores in each setting. Finally, the outcomes were correlated across settings. A good correlation was found between time needed to achieve predefined performance levels on the VR simulator and the physical mannequin (Pearson correlation coefficient .78; P < .001). Performance scores on the VR simulator correlated well to the clinical performance scores (Pearson correlation coefficient .81; P = .049). No significant correlations were found between numbers of attempts needed to reach proficiency across the 3 different settings. A post hoc analysis found that the 50% fastest trainees at reaching proficiency during simulation-based training received higher clinical performance scores compared to trainees with scores placing them among the 50% slowest (P = .025). Performances during simulation-based sonography training may predict performance in related tasks and subsequent clinical learning curves. © 2016 by the American Institute of Ultrasound in Medicine.

Sex is not everything: the role of gender in early performance of a fundamental laparoscopic skill.

PubMed

Kolozsvari, Nicoleta O; Andalib, Amin; Kaneva, Pepa; Cao, Jiguo; Vassiliou, Melina C; Fried, Gerald M; Feldman, Liane S

2011-04-01

Existing literature on the acquisition of surgical skills suggests that women generally perform worse than men. This literature is limited by looking at an arbitrary number of trials and not adjusting for potential confounders. The objective of this study was to evaluate the impact of gender on the learning curve for a fundamental laparoscopic task. Thirty-two medical students performed the FLS peg transfer task and their scores were plotted to generate a learning curve. Nonlinear regression was used to estimate learning plateau and learning rate. Variables that may affect performance were assessed using a questionnaire. Innate visual-spatial abilities were evaluated using tests for spatial orientation, spatial scanning, and perceptual abilities. Score on first peg transfer attempt, learning plateau, and learning rate were compared for men and women using Student's t test. Innate abilities were correlated to simulator performance using Pearson's coefficient. Multivariate linear regression was used to investigate the effect of gender on early laparoscopic performance after adjusting for factors found significant on univariate analysis. Statistical significance was defined as P < 0.05. Nineteen men and 13 women participated in the study; 30 were right-handed, 12 reported high interest in surgery, and 26 had video game experience. There were no differences between men and women in initial peg transfer score, learning plateau, or learning rate. Initial peg transfer score and learning rate were higher in subjects who reported having a high interest in surgery (P = 0.02, P = 0.03). Initial score also correlated with perceptual ability score (P = 0.03). In multivariate analysis, only surgical interest remained a significant predictor of score on first peg transfer (P = 0.03) and learning rate (P = 0.02), while gender had no significant relationship to early performance. Gender did not affect the learning curve for a fundamental laparoscopic task, while interest in surgery and perceptual abilities did influence early performance.

Longitudinal Analysis of the Role of Perceived Self-Efficacy for Self-Regulated Learning in Academic Continuance and Achievement

ERIC Educational Resources Information Center

Caprara, Gian Vittorio; Fida, Roberta; Vecchione, Michele; Del Bove, Giannetta; Vecchio, Giovanni Maria; Barbaranelli, Claudio; Bandura, Albert

2008-01-01

The present study examined the developmental course of perceived efficacy for self-regulated learning and its contribution to academic achievement and likelihood of remaining in school in a sample of 412 Italian students (48% males and 52% females ranging in age from 12 to 22 years). Latent growth curve analysis revealed a progressive decline in…

Learning Curves of Virtual Mastoidectomy in Distributed and Massed Practice.

PubMed

Andersen, Steven Arild Wuyts; Konge, Lars; Cayé-Thomasen, Per; Sørensen, Mads Sølvsten

2015-10-01

Repeated and deliberate practice is crucial in surgical skills training, and virtual reality (VR) simulation can provide self-directed training of basic surgical skills to meet the individual needs of the trainee. Assessment of the learning curves of surgical procedures is pivotal in understanding skills acquisition and best-practice implementation and organization of training. To explore the learning curves of VR simulation training of mastoidectomy and the effects of different practice sequences with the aim of proposing the optimal organization of training. A prospective trial with a 2 × 2 design was conducted at an academic teaching hospital. Participants included 43 novice medical students. Of these, 21 students completed time-distributed practice from October 14 to November 29, 2013, and a separate group of 19 students completed massed practice on May 16, 17, or 18, 2014. Data analysis was performed from June 6, 2014, to March 3, 2015. Participants performed 12 repeated virtual mastoidectomies using a temporal bone surgical simulator in either a distributed (practice blocks spaced in time) or massed (all practice in 1 day) training program with randomization for simulator-integrated tutoring during the first 5 sessions. Performance was assessed using a modified Welling Scale for final product analysis by 2 blinded senior otologists. Compared with the 19 students in the massed practice group, the 21 students in the distributed practice group were older (mean age, 25.1 years), more often male (15 [62%]), and had slightly higher mean gaming frequency (2.3 on a 1-5 Likert scale). Learning curves were established and distributed practice was found to be superior to massed practice, reported as mean end score (95% CI) of 15.7 (14.4-17.0) in distributed practice vs. 13.0 (11.9-14.1) with massed practice (P = .002). Simulator-integrated tutoring accelerated the initial performance, with mean score for tutored sessions of 14.6 (13.9-15.2) vs. 13.4 (12.8-14.0) for corresponding nontutored sessions (P < .01) but at the cost of a drop in performance once tutoring ceased. The performance drop was less with distributed practice, suggesting a protective effect when acquired skills were consolidated over time. The mean performance of the nontutored participants in the distributed practice group plateaued on a score of 16.0 (15.3-16.7) at approximately the ninth repetition, but the individual learning curves were highly variable. Novices can acquire basic mastoidectomy competencies with self-directed VR simulation training. Training should be organized with distributed practice, and simulator-integrated tutoring can be useful to accelerate the initial learning curve. Practice should be deliberate and toward a standard set level of proficiency that remains to be defined rather than toward the mean learning curve plateau.

The Analysis of Seawater: A Laboratory-Centered Learning Project in General Chemistry.

ERIC Educational Resources Information Center

Selco, Jodye I.; Roberts, Julian L., Jr.; Wacks, Daniel B.

2003-01-01

Describes a sea-water analysis project that introduces qualitative and quantitative analysis methods and laboratory methods such as gravimetric analysis, potentiometric titration, ion-selective electrodes, and the use of calibration curves. Uses a problem-based cooperative teaching approach. (Contains 24 references.) (YDS)

Machine Learning Techniques for Stellar Light Curve Classification

NASA Astrophysics Data System (ADS)

Hinners, Trisha A.; Tat, Kevin; Thorp, Rachel

2018-07-01

We apply machine learning techniques in an attempt to predict and classify stellar properties from noisy and sparse time-series data. We preprocessed over 94 GB of Kepler light curves from the Mikulski Archive for Space Telescopes (MAST) to classify according to 10 distinct physical properties using both representation learning and feature engineering approaches. Studies using machine learning in the field have been primarily done on simulated data, making our study one of the first to use real light-curve data for machine learning approaches. We tuned our data using previous work with simulated data as a template and achieved mixed results between the two approaches. Representation learning using a long short-term memory recurrent neural network produced no successful predictions, but our work with feature engineering was successful for both classification and regression. In particular, we were able to achieve values for stellar density, stellar radius, and effective temperature with low error (∼2%–4%) and good accuracy (∼75%) for classifying the number of transits for a given star. The results show promise for improvement for both approaches upon using larger data sets with a larger minority class. This work has the potential to provide a foundation for future tools and techniques to aid in the analysis of astrophysical data.

Explicit and Implicit Processes Constitute the Fast and Slow Processes of Sensorimotor Learning.

PubMed

McDougle, Samuel D; Bond, Krista M; Taylor, Jordan A

2015-07-01

A popular model of human sensorimotor learning suggests that a fast process and a slow process work in parallel to produce the canonical learning curve (Smith et al., 2006). Recent evidence supports the subdivision of sensorimotor learning into explicit and implicit processes that simultaneously subserve task performance (Taylor et al., 2014). We set out to test whether these two accounts of learning processes are homologous. Using a recently developed method to assay explicit and implicit learning directly in a sensorimotor task, along with a computational modeling analysis, we show that the fast process closely resembles explicit learning and the slow process approximates implicit learning. In addition, we provide evidence for a subdivision of the slow/implicit process into distinct manifestations of motor memory. We conclude that the two-state model of motor learning is a close approximation of sensorimotor learning, but it is unable to describe adequately the various implicit learning operations that forge the learning curve. Our results suggest that a wider net be cast in the search for the putative psychological mechanisms and neural substrates underlying the multiplicity of processes involved in motor learning. Copyright © 2015 the authors 0270-6474/15/359568-12$15.00/0.

Explicit and Implicit Processes Constitute the Fast and Slow Processes of Sensorimotor Learning

PubMed Central

Bond, Krista M.; Taylor, Jordan A.

2015-01-01

A popular model of human sensorimotor learning suggests that a fast process and a slow process work in parallel to produce the canonical learning curve (Smith et al., 2006). Recent evidence supports the subdivision of sensorimotor learning into explicit and implicit processes that simultaneously subserve task performance (Taylor et al., 2014). We set out to test whether these two accounts of learning processes are homologous. Using a recently developed method to assay explicit and implicit learning directly in a sensorimotor task, along with a computational modeling analysis, we show that the fast process closely resembles explicit learning and the slow process approximates implicit learning. In addition, we provide evidence for a subdivision of the slow/implicit process into distinct manifestations of motor memory. We conclude that the two-state model of motor learning is a close approximation of sensorimotor learning, but it is unable to describe adequately the various implicit learning operations that forge the learning curve. Our results suggest that a wider net be cast in the search for the putative psychological mechanisms and neural substrates underlying the multiplicity of processes involved in motor learning. PMID:26134640

Learning curves in health care.

PubMed

Waldman, J Deane; Yourstone, Steven A; Smith, Howard L

2003-01-01

This article explores the uses of learning curve theory in medicine. Though effective application of learning curve theory in health care can result in higher quality and lower cost, it is seldom methodically applied in clinical practice. Fundamental changes are necessary in the corporate culture of medicine in order to capitalize maximally on the benefits of learning.

Enhancements of Bayesian Blocks; Application to Large Light Curve Databases

NASA Technical Reports Server (NTRS)

Scargle, Jeff

2015-01-01

Bayesian Blocks are optimal piecewise linear representations (step function fits) of light-curves. The simple algorithm implementing this idea, using dynamic programming, has been extended to include more data modes and fitness metrics, multivariate analysis, and data on the circle (Studies in Astronomical Time Series Analysis. VI. Bayesian Block Representations, Scargle, Norris, Jackson and Chiang 2013, ApJ, 764, 167), as well as new results on background subtraction and refinement of the procedure for precise timing of transient events in sparse data. Example demonstrations will include exploratory analysis of the Kepler light curve archive in a search for "star-tickling" signals from extraterrestrial civilizations. (The Cepheid Galactic Internet, Learned, Kudritzki, Pakvasa1, and Zee, 2008, arXiv: 0809.0339; Walkowicz et al., in progress).

The implications of renewable energy research and development: Policy scenario analysis with experience and learning effects

NASA Astrophysics Data System (ADS)

Kobos, Peter Holmes

This dissertation analyzes the current and potential future costs of renewable energy technology from an institutional perspective. The central hypothesis is that reliable technology cost forecasting can be achieved through standard and modified experience curves implemented in a dynamic simulation model. Additionally, drawing upon region-specific institutional lessons highlights the role of market, social, and political institutions throughout an economy. Socio-political influences and government policy pathways drive resource allocation decisions that may be predominately influenced by factors other than those considered in a traditional market-driven, mechanistic approach. Learning in economic systems as a research topic is an attractive complement to the notion of institutional pathways. The economic implications of learning by doing, as first outlined by Arrow (1962), highlight decreasing production costs as individuals, or more generally the firm, become more familiar with a production process. The standard approach in the literature has been to employ a common experience curve where cumulative production is the only independent variable affecting costs. This dissertation develops a two factor experience curve, adding research, development and demonstration (RD&D) expenditures as a second variable. To illustrate the concept in the context of energy planning, two factor experience curves are developed for wind energy technology and solar photovoltaic (PV) modules under different assumptions on learning rates for cumulative capacity and the knowledge stock (a function of past RD&D efforts). Additionally, a one factor experience curve and cost trajectory scenarios are developed for concentrated solar power and geothermal energy technology, respectively. Cost forecasts are then developed for all four of these technologies in a dynamic simulation model. Combining the theoretical framework of learning by doing with the fields of organizational learning and institutional economics, this dissertation argues that the current state of renewable energy technology costs is largely due to the past production efforts (learning by doing) and RD&D efforts (learning by searching) in these global industries. This cost pathway, however, may be altered through several policy process feedback mechanisms including targeted RD&D expenditures, maintenance of RD&D to promote learning effects, and financial incentive programs that support energy production from renewable energy technologies.

Can the learning curve of totally endoscopic robotic mitral valve repair be short-circuited?

PubMed

Yaffee, David W; Loulmet, Didier F; Kelly, Lauren A; Ward, Alison F; Ursomanno, Patricia A; Rabinovich, Annette E; Neuburger, Peter J; Krishnan, Sandeep; Hill, Frederick T; Grossi, Eugene A

2014-01-01

A concern with the initiation of totally endoscopic robotic mitral valve repair (TERMR) programs has been the risk for the learning curve. To minimize this risk, we initiated a TERMR program with a defined team and structured learning approach before clinical implementation. A dedicated team (two surgeons, one cardiac anesthesiologist, one perfusionist, and two nurses) was trained with clinical scenarios, simulations, wet laboratories, and "expert" observation for 3 months. This team then performed a series of TERMRs of varying complexity. Thirty-two isolated TERMRs were performed during the first programmatic year. All operations included mitral valve repair, left atrial appendage exclusion, and annuloplasty device implantation. Additional procedures included leaflet resection, neochordae insertion, atrial ablation, and papillary muscle shortening. Longer clamp times were associated with number of neochordae (P < 0.01), papillary muscle procedures (P < 0.01), and leaflet resection (P = 0.06). Sequential case number had no impact on cross-clamp time (P = 0.3). Analysis of nonclamp time demonstrated a 71.3% learning percentage (P < 0.01; ie, 28.7% reduction in nonclamp time with each doubling of case number). There were no hospital deaths or incidences of stroke, myocardial infarction, unplanned reoperation, respiratory failure, or renal failure. Median length of stay was 4 days. All patients were discharged home. Totally endoscopic robotic mitral valve repair can be safely performed after a pretraining regimen with emphasis on experts' current practice and team training. After a pretraining regimen, cross-clamp times were not subject to learning curve phenomena but were dependent on procedural complexity. Nonclamp times were associated with a short learning curve.

Learning curve of single port laparoscopic cholecystectomy determined using the non-linear ordinary least squares method based on a non-linear regression model: An analysis of 150 consecutive patients.

PubMed

Han, Hyung Joon; Choi, Sae Byeol; Park, Man Sik; Lee, Jin Suk; Kim, Wan Bae; Song, Tae Jin; Choi, Sang Yong

2011-07-01

Single port laparoscopic surgery has come to the forefront of minimally invasive surgery. For those familiar with conventional techniques, however, this type of operation demands a different type of eye/hand coordination and involves unfamiliar working instruments. Herein, the authors describe the learning curve and the clinical outcomes of single port laparoscopic cholecystectomy for 150 consecutive patients with benign gallbladder disease. All patients underwent single port laparoscopic cholecystectomy using a homemade glove port by one of five operators with different levels of experiences of laparoscopic surgery. The learning curve for each operator was fitted using the non-linear ordinary least squares method based on a non-linear regression model. Mean operating time was 77.6 ± 28.5 min. Fourteen patients (6.0%) were converted to conventional laparoscopic cholecystectomy. Complications occurred in 15 patients (10.0%), as follows: bile duct injury (n = 2), surgical site infection (n = 8), seroma (n = 2), and wound pain (n = 3). One operator achieved a learning curve plateau at 61.4 min per procedure after 8.5 cases and his time improved by 95.3 min as compared with initial operation time. Younger surgeons showed significant decreases in mean operation time and achieved stable mean operation times. In particular, younger surgeons showed significant decreases in operation times after 20 cases. Experienced laparoscopic surgeons can safely perform single port laparoscopic cholecystectomy using conventional or angled laparoscopic instruments. The present study shows that an operator can overcome the single port laparoscopic cholecystectomy learning curve in about eight cases.

Home Literacy Environment and Head Start Children's Language Development: The Role of Approaches to Learning

ERIC Educational Resources Information Center

Meng, Christine

2015-01-01

Research Findings: This study examined whether approaches to learning moderate the association between home literacy environment and English receptive vocabulary development. The Head Start Family and Child Experiences Survey (2003 cohort) was used for analysis. Latent growth curve modeling was utilized to test a quadratic model of English…

Three-Hand Endoscopic Endonasal Transsphenoidal Surgery: Experience With an Anatomy-Preserving Mononostril Approach Technique.

PubMed

Eseonu, Chikezie I; ReFaey, Karim; Pamias-Portalatin, Eva; Asensio, Javier; Garcia, Oscar; Boahene, Kofi D; Quiñones-Hinojosa, Alfredo

2018-02-01

Variations on the endoscopic transsphenoidal approach present unique surgical techniques that have unique effects on surgical outcomes, extent of resection (EOR), and anatomical complications. To analyze the learning curve and perioperative outcomes of the 3-hand endoscopic endonasal mononostril transsphenoidal technique. Prospective case series and retrospective data analysis of patients who were treated with the 3-hand transsphenoidal technique between January 2007 and May 2015 by a single neurosurgeon. Patient characteristics, preoperative presentation, tumor characteristics, operative times, learning curve, and postoperative outcomes were analyzed. Volumetric EOR was evaluated, and a logistic regression analysis was used to assess predictors of EOR. Two hundred seventy-five patients underwent an endoscopic transsphenoidal surgery using the 3-hand technique. One hundred eighteen patients in the early group had surgery between 2007 and 2010, while 157 patients in the late group had surgery between 2011 and 2015. Operative time was significantly shorter in the late group (161.6 min) compared to the early group (211.3 min, P = .001). Both cohorts had similar EOR (early group 84.6% vs late group 85.5%, P = .846) and postoperative outcomes. The learning curve showed that it took 54 cases to achieve operative proficiency with the 3-handed technique. Multivariate modeling suggested that prior resections and preoperative tumor size are important predictors for EOR. We describe a 3-hand, mononostril endoscopic transsphenoidal technique performed by a single neurosurgeon that has minimal anatomic distortion and postoperative complications. During the learning curve of this technique, operative time can significantly decrease, while EOR, postoperative outcomes, and complications are not jeopardized. Copyright © 2017 by the Congress of Neurological Surgeons

The learning curve of robot-assisted laparoscopic aortofemoral bypass grafting for aortoiliac occlusive disease.

PubMed

Novotný, Tomáš; Dvorák, Martin; Staffa, Robert

2011-02-01

Since the end of the 20th century, robot-assisted surgery has been finding its role among other minimally invasive methods. Vascular surgery seems to be another specialty in which the benefits of this technology can be expected. Our objective was to assess the learning curve of robot-assisted laparoscopic aortofemoral bypass grafting for aortoiliac occlusive disease in a group of 40 patients. Between May 2006 and January 2010, 40 patients (32 men, 8 women), who were a median age of 58 years (range, 48-75 years), underwent 40 robot-assisted laparoscopic aortofemoral reconstructions. Learning curve estimations were used for anastomosis, clamping, and operative time assessment. For conversion rate evaluation, the cumulative summation (CUSUM) technique was used. Statistical analysis comparing the first and second half of our group, and unilateral-to-bilateral reconstructions were performed. We created 21 aortofemoral and 19 aortobifemoral bypasses. The median proximal anastomosis time was 23 minutes (range, 18-50 minutes), median clamping time was 60 minutes (range, 40-95 minutes), and median operative time was 295 minutes (range, 180-475 minutes). The 30-day mortality rate was 0%, and no graft or wound infection or cardiopulmonary or hepatorenal complications were observed. During the median 18-month follow-up (range, 2-48 months), three early graft occlusions occurred (7%). After reoperations, the secondary patency of reconstructions was 100%. Data showed a typical short learning curve for robotic proximal anastomosis creation with anastomosis and clamping time reduction. The operative time learning curve was flat, confirming the procedure's complexity. There were two conversions to open surgery. CUSUM analysis confirmed that an acceptable conversion rate set at 5% was achieved. Comparing the first and second half of our group, all recorded times showed statistically significant improvements. Differences between unilateral and bilateral reconstructions were not statistically significant. Our results show that the success rate of robot-assisted laparoscopic aortofemoral bypass grafting is high and the complication rate is low. Anastomosis creation, one of the main difficulties of laparoscopic bypass grafting, has been overcome using the robotic operating system and its learning curve is short. However, the endoscopic dissection of the aortoiliac segment remains the most difficult part of the operation and should be addressed in further development of the method to reduce the operative times. Long-term results and potential benefits of this minimally invasive method have to be verified by randomized controlled clinical trials. Copyright © 2011 Society for Vascular Surgery. Published by Mosby, Inc. All rights reserved.

The learning curve of laparoscopic holecystectomy in general surgery resident training: old age of the patient may be a risk factor?

PubMed

Ferrarese, Alessia; Gentile, Valentina; Bindi, Marco; Rivelli, Matteo; Cumbo, Jacopo; Solej, Mario; Enrico, Stefano; Martino, Valter

2016-01-01

A well-designed learning curve is essential for the acquisition of laparoscopic skills: but, are there risk factors that can derail the surgical method? From a review of the current literature on the learning curve in laparoscopic surgery, we identified learning curve components in video laparoscopic cholecystectomy; we suggest a learning curve model that can be applied to assess the progress of general surgical residents as they learn and master the stages of video laparoscopic cholecystectomy regardless of type of patient. Electronic databases were interrogated to better define the terms "surgeon", "specialized surgeon", and "specialist surgeon"; we surveyed the literature on surgical residency programs outside Italy to identify learning curve components, influential factors, the importance of tutoring, and the role of reference centers in residency education in surgery. From the definition of acceptable error, self-efficacy, and error classification, we devised a learning curve model that may be applied to training surgical residents in video laparoscopic cholecystectomy. Based on the criteria culled from the literature, the three surgeon categories (general, specialized, and specialist) are distinguished by years of experience, case volume, and error rate; the patients were distinguished for years and characteristics. The training model was constructed as a series of key learning steps in video laparoscopic cholecystectomy. Potential errors were identified and the difficulty of each step was graded using operation-specific characteristics. On completion of each procedure, error checklist scores on procedure-specific performance are tallied to track the learning curve and obtain performance indices of measurement that chart the trainee's progress. The concept of the learning curve in general surgery is disputed. The use of learning steps may enable the resident surgical trainee to acquire video laparoscopic cholecystectomy skills proportional to the instructor's ability, the trainee's own skills, and the safety of the surgical environment. There were no patient characteristics that can derail the methods. With this training scheme, resident trainees may be provided the opportunity to develop their intrinsic capabilities without the loss of basic technical skills.

Locus of Control Orientations in Students with Intellectual Disability, Learning Disabilities, and No Disabilities: A Latent Growth Curve Analysis

ERIC Educational Resources Information Center

Shogren, Karrie A.; Bovaird, James A.; Palmer, Susan B.; Wehmeyer, Michael L.

2010-01-01

Previous research has suggested differences in the locus of control (LOC) orientations of students with intellectual disability, learning disabilities, and no disabilities, although this research has been characterized by methodological limitations. The purpose of this study was to examine the development of LOC orientations in students with…

Is the learning curve endless? One surgeon's experience with robotic prostatectomy

NASA Astrophysics Data System (ADS)

Patel, Vipul; Thaly, Rahul; Shah, Ketul

2007-02-01

Introduction: After performing 1,000 robotic prostatectomies we reflected back on our experience to determine what defined the learning curve and the essential elements that were the keys to surmounting it. Method: We retrospectively assessed our experience to attempt to define the learning curve(s), key elements of the procedure, technical refinements and changes in technology that facilitated our progress. Result: The initial learning curve to achieve basic competence and the ability to smoothly perform the procedure in less than 4 hours with acceptable outcomes was approximately 25 cases. A second learning curve was present between 75-100 cases as we approached more complicated patients. At 200 cases we were comfortably able to complete the procedure routinely in less than 2.5 hours with no specific step of the procedure hindering our progression. At 500 cases we had the introduction of new instrumentation (4th arm, biopolar Maryland, monopolar scissors) that changed our approach to the bladder neck and neurovascular bundle dissection. The most challenging part of the procedure was the bladder neck dissection. Conclusion: There is no single parameter that can be used to assess or define the learning curve. We used a combination of factors to make our subjective definition this included: operative time, smoothness of technical progression during the case along with clinical outcomes. The further our case experience progressed the more we expected of our outcomes, thus we continually modified our technique and hence embarked upon yet a new learning curve.

Utility of Interobserver Agreement Statistics in Establishing Radiology Resident Learning Curves During Self-directed Radiologic Anatomy Training.

PubMed

Tureli, Derya; Altas, Hilal; Cengic, Ismet; Ekinci, Gazanfer; Baltacioglu, Feyyaz

2015-10-01

The aim of the study was to ascertain the learning curves for the radiology residents when first introduced to an anatomic structure in magnetic resonance images (MRI) to which they have not been previously exposed to. The iliolumbar ligament is a good marker for testing learning curves of radiology residents because the ligament is not part of a routine lumbar MRI reporting and has high variability in detection. Four radiologists, three residents without previous training and one mentor, studied standard axial T1- and T2-weighted images of routine lumbar MRI examinations. Radiologists had to define iliolumbar ligament while blinded to each other's findings. Interobserver agreement analyses, namely Cohen and Fleiss κ statistics, were performed for groups of 20 cases to evaluate the self-learning curve of radiology residents. Mean κ values of resident-mentor pairs were 0.431, 0.608, 0.604, 0.826, and 0.963 in the analysis of successive groups (P < .001). The results indicate that the concordance between the experienced and inexperienced radiologists started as weak (κ <0.5) and gradually became very acceptable (κ >0.8). Therefore, a junior radiology resident can obtain enough experience in identifying a rather ambiguous anatomic structure in routine MRI after a brief instruction of a few minutes by a mentor and studying approximately 80 cases by oneself. Implementing this methodology will help radiology educators obtain more concrete ideas on the optimal time and effort required for supported self-directed visual learning processes in resident education. Copyright © 2015 AUR. Published by Elsevier Inc. All rights reserved.

Transforaminal Lumbar Interbody Fusion with Rigid Interspinous Process Fixation: A Learning Curve Analysis of a Surgeon Team's First 74 Cases.

PubMed

Doherty, Patrick; Welch, Arthur; Tharpe, Jason; Moore, Camille; Ferry, Chris

2017-05-30

Studies have shown that a significant learning curve may be associated with adopting minimally invasive transforaminal lumbar interbody fusion (MIS TLIF) with bilateral pedicle screw fixation (BPSF). Accordingly, several hybrid TLIF techniques have been proposed as surrogates to the accepted BPSF technique, asserting that less/fewer fixation(s) or less disruptive fixation may decrease the learning curve while still maintaining the minimally disruptive benefits. TLIF with interspinous process fixation (ISPF) is one such surrogate procedure. However, despite perceived ease of adaptability given the favorable proximity of the spinous processes, no evidence exists demonstrating whether or not the technique may possess its own inherent learning curve. The purpose of this study was to determine whether an intraoperative learning curve for one- and two-level TLIF + ISPF may exist for a single lead surgeon. Seventy-four consecutive patients who received one- or two-Level TLIF with rigid ISPF by a single lead surgeon were retrospectively reviewed. It was the first TLIF + ISPF case series for the lead surgeon. Intraoperative blood loss (EBL), hospitalization length-of-stay (LOS), fluoroscopy time, and postoperative complications were collected. EBL, LOS, and fluoroscopy time were modeled as a function of case number using multiple linear regression methods. A change point was included in each model to allow the trajectory of the outcomes to change during the duration of the case series. These change points were determined using profile likelihood methods. Models were fit using the maximum likelihood estimates for the change points. Age, sex, body mass index (BMI), and the number of treated levels were included as covariates. EBL, LOS, and fluoroscopy time did not significantly differ by age, sex, or BMI (p ≥ 0.12). Only EBL differed significantly by the number of levels (p = 0.026). The case number was not a significant predictor of EBL, LOS, or fluoroscopy time (p ≥ 0.21). At the time of data collection (mean time from surgery: 13.3 months), six patients had undergone revision due to interbody migration. No ISPF device complications were observed. Study outcomes support the ideal that TLIF + ISPF can be a readily adopted procedure without a significant intraoperative learning curve. However, the authors emphasize that further assessment of long-term healing outcomes is essential in fully characterizing both the efficacy and the indication learning curve for the TLIF + ISPF technique.

Analysis of Learning Curve Fitting Techniques.

DTIC Science & Technology

1987-09-01

1986. 15. Neter, John and others. Applied Linear Regression Models. Homewood IL: Irwin, 19-33. 16. SAS User’s Guide: Basics, Version 5 Edition. SAS... Linear Regression Techniques (15:23-52). Random errors are assumed to be normally distributed when using -# ordinary least-squares, according to Johnston...lot estimated by the improvement curve formula. For a more detailed explanation of the ordinary least-squares technique, see Neter, et. al., Applied

Playing off the curve - testing quantitative predictions of skill acquisition theories in development of chess performance.

PubMed

Gaschler, Robert; Progscha, Johanna; Smallbone, Kieran; Ram, Nilam; Bilalić, Merim

2014-01-01

Learning curves have been proposed as an adequate description of learning processes, no matter whether the processes manifest within minutes or across years. Different mechanisms underlying skill acquisition can lead to differences in the shape of learning curves. In the current study, we analyze the tournament performance data of 1383 chess players who begin competing at young age and play tournaments for at least 10 years. We analyze the performance development with the goal to test the adequacy of learning curves, and the skill acquisition theories they are based on, for describing and predicting expertise acquisition. On the one hand, we show that the skill acquisition theories implying a negative exponential learning curve do a better job in both describing early performance gains and predicting later trajectories of chess performance than those theories implying a power function learning curve. On the other hand, the learning curves of a large proportion of players show systematic qualitative deviations from the predictions of either type of skill acquisition theory. While skill acquisition theories predict larger performance gains in early years and smaller gains in later years, a substantial number of players begin to show substantial improvements with a delay of several years (and no improvement in the first years), deviations not fully accounted for by quantity of practice. The current work adds to the debate on how learning processes on a small time scale combine to large-scale changes.

The sales learning curve.

PubMed

Leslie, Mark; Holloway, Charles A

2006-01-01

When a company launches a new product into a new market, the temptation is to immediately ramp up sales force capacity to gain customers as quickly as possible. But hiring a full sales force too early just causes the firm to burn through cash and fail to meet revenue expectations. Before it can sell an innovative product efficiently, the entire organization needs to learn how customers will acquire and use it, a process the authors call the sales learning curve. The concept of a learning curve is well understood in manufacturing. Employees transfer knowledge and experience back and forth between the production line and purchasing, manufacturing, engineering, planning, and operations. The sales learning curve unfolds similarly through the give-and-take between the company--marketing, sales, product support, and product development--and its customers. As customers adopt the product, the firm modifies both the offering and the processes associated with making and selling it. Progress along the manufacturing curve is measured by tracking cost per unit: The more a firm learns about the manufacturing process, the more efficient it becomes, and the lower the unit cost goes. Progress along the sales learning curve is measured in an analogous way: The more a company learns about the sales process, the more efficient it becomes at selling, and the higher the sales yield. As the sales yield increases, the sales learning process unfolds in three distinct phases--initiation, transition, and execution. Each phase requires a different size--and kind--of sales force and represents a different stage in a company's production, marketing, and sales strategies. Adjusting those strategies as the firm progresses along the sales learning curve allows managers to plan resource allocation more accurately, set appropriate expectations, avoid disastrous cash shortfalls, and reduce both the time and money required to turn a profit.

MILS in a general surgery unit: learning curve, indications, and limitations.

PubMed

Patriti, Alberto; Marano, Luigi; Casciola, Luciano

2015-06-01

Minimally invasive liver surgery (MILS) is going to be a method with a wide diffusion even in general surgery units. Organization, learning curve effect, and the environment are crucial issues to evaluate before starting a program of minimally invasive liver resections. Analysis of a consecutive series of 70 patients has been used to define advantages and limits of starting a program of MILS in a general surgery unit. Seventeen MILS have been calculated with the cumulative sum method as the number of cases to complete the learning curve. Operative times [270 (60-480) vs. 180 (15-550) min; p 0.01] and rate of conversion (6/17 vs. 5/53; p 0.018) decrease after this number of cases. More complex cases can be managed after a proper optimization of all steps of liver resection. When a high confidence of the medical and nurse staff with MILS is reached, economical and strategic issues should be evaluated in order to establish a multidisciplinary hepatobiliary unit independent from the general surgery unit to manage more complex cases.

Teaching Learning Curves in an Undergraduate Economics or Operations Management Course

ERIC Educational Resources Information Center

Naidu, Jaideep T.; Sanford, John F.

2012-01-01

Learning Curves has its roots in economics and behavioral psychology. Learning Curves theory has several business applications and is widely used in the industry. As faculty of Operations Management courses, we cover this topic in some depth in the classroom. In this paper, we present some of our teaching methods and material that have helped us…

Learning curves for transapical transcatheter aortic valve replacement in the PARTNER-I trial: Technical performance, success, and safety.

PubMed

Suri, Rakesh M; Minha, Sa'ar; Alli, Oluseun; Waksman, Ron; Rihal, Charanjit S; Satler, Lowell P; Greason, Kevin L; Torguson, Rebecca; Pichard, Augusto D; Mack, Michael; Svensson, Lars G; Rajeswaran, Jeevanantham; Lowry, Ashley M; Ehrlinger, John; Mick, Stephanie L; Tuzcu, E Murat; Thourani, Vinod H; Makkar, Raj; Holmes, David; Leon, Martin B; Blackstone, Eugene H

2016-09-01

Introduction of hybrid techniques, such as transapical transcatheter aortic valve replacement (TA-TAVR), requires skills that a heart team must master to achieve technical efficiency: the technical performance learning curve. To date, the learning curve for TA-TAVR remains unknown. We therefore evaluated the rate at which technical performance improved, assessed change in occurrence of adverse events in relation to technical performance, and determined whether adverse events after TA-TAVR were linked to acquiring technical performance efficiency (the learning curve). From April 2007 to February 2012, 1100 patients, average age 85.0 ± 6.4 years, underwent TA-TAVR in the PARTNER-I trial. Learning curves were defined by institution-specific patient sequence number using nonlinear mixed modeling. Mean procedure time decreased from 131 to 116 minutes within 30 cases (P = .06) and device success increased to 90% by case 45 (P = .0007). Within 30 days, 354 patients experienced a major adverse event (stroke in 29, death in 96), with possibly decreased complications over time (P ∼ .08). Although longer procedure time was associated with more adverse events (P < .0001), these events were associated with change in patient risk profile, not the technical performance learning curve (P = .8). The learning curve for TA-TAVR was 30 to 45 procedures performed, and technical efficiency was achieved without compromising patient safety. Although fewer patients are now undergoing TAVR via nontransfemoral access, understanding TA-TAVR learning curves and their relationship with outcomes is important as the field moves toward next-generation devices, such as those to replace the mitral valve, delivered via the left ventricular apex. Copyright © 2016 The American Association for Thoracic Surgery. Published by Elsevier Inc. All rights reserved.

Learning micro incision surgery without the learning curve

PubMed Central

Navin, Shoba; Parikh, Rajul

2008-01-01

We describe a method of learning micro incision cataract surgery painlessly with the minimum of learning curves. A large-bore or standard anterior chamber maintainer (ACM) facilitates learning without change of machine or preferred surgical technique. Experience with the use of an ACM during phacoemulsification is desirable. PMID:18292624

Comparison of Learning Curves for Major and Minor Laparoscopic Liver Resection.

PubMed

Lee, Woohyung; Woo, Jung-Woo; Lee, Jin-Kwon; Park, Ji-Ho; Kim, Ju-Yeon; Kwag, Seung-Jin; Park, Taejin; Jeong, Sang-Ho; Ju, Young-Tae; Jeong, Eun-Jung; Lee, Young-Joon; Choi, Sang-Kyung; Hong, Soon-Chan; Jeong, Chi-Young

2016-06-01

Because laparoscopic liver resection (LLR) has a steep learning curve, analyzing experience is important for trainees. Several authors have described the learning curve of LLR, without comparing the learning curves between major and minor LLR. Perioperative data were retrieved from the medical records of 170 consecutive patients who underwent LLR by a single surgeon at a tertiary hospital. Learning curves were generated and compared between major and minor LLR using cumulative sum control charts and the moving average. Major and minor LLR was performed in 96 and 74 patients, respectively. The learning curves showed a steady state after case 50 for major LLR. Because of discordant results in minor LLR, subgroup analyses were performed, showing competency in LLR after cases 25 and 35 for left lateral sectionectomy and tumorectomy, respectively. Transfused red blood cell volume (0.6 versus 2.2 packs, P < .001) decreased after achievement of competence in major LLR. Blood loss exceeding 500 mL (odds ratio 2.395, 95% confidence interval 1.096-5.233, P = .028) was independently associated with LLR failure. The number of cases required to accomplish LLR differed according to the extent of resection. Extensive blood loss was independently associated with LLR failure.

Cumulative sum analysis score and phacoemulsification competency learning curve.

PubMed

Vedana, Gustavo; Cardoso, Filipe G; Marcon, Alexandre S; Araújo, Licio E K; Zanon, Matheus; Birriel, Daniella C; Watte, Guilherme; Jun, Albert S

2017-01-01

To use the cumulative sum analysis score (CUSUM) to construct objectively the learning curve of phacoemulsification competency. Three second-year residents and an experienced consultant were monitored for a series of 70 phacoemulsification cases each and had their series analysed by CUSUM regarding posterior capsule rupture (PCR) and best-corrected visual acuity. The acceptable rate for PCR was <5% (lower limit h) and the unacceptable rate was >10% (upper limit h). The acceptable rate for best-corrected visual acuity worse than 20/40 was <10% (lower limit h) and the unacceptable rate was >20% (upper limit h). The area between lower limit h and upper limit h is called the decision interval. There was no statistically significant difference in the mean age, sex or cataract grades between groups. The first trainee achieved PCR CUSUM competency at his 22 nd case. His best-corrected visual acuity CUSUM was in the decision interval from his third case and stayed there until the end, never reaching competency. The second trainee achieved PCR CUSUM competency at his 39 th case. He could reach best-corrected visual acuity CUSUM competency at his 22 nd case. The third trainee achieved PCR CUSUM competency at his 41 st case. He reached best-corrected visual acuity CUSUM competency at his 14 th case. The learning curve of competency in phacoemulsification is constructed by CUSUM and in average took 38 cases for each trainee to achieve it.

Simultaneous development of laparoscopy and robotics provides acceptable perioperative outcomes and shows robotics to have a faster learning curve and to be overall faster in rectal cancer surgery: analysis of novice MIS surgeon learning curves.

PubMed

Melich, George; Hong, Young Ki; Kim, Jieun; Hur, Hyuk; Baik, Seung Hyuk; Kim, Nam Kyu; Sender Liberman, A; Min, Byung Soh

2015-03-01

Laparoscopy offers some evidence of benefit compared to open rectal surgery. Robotic rectal surgery is evolving into an accepted approach. The objective was to analyze and compare laparoscopic and robotic rectal surgery learning curves with respect to operative times and perioperative outcomes for a novice minimally invasive colorectal surgeon. One hundred and six laparoscopic and 92 robotic LAR rectal surgery cases were analyzed. All surgeries were performed by a surgeon who was primarily trained in open rectal surgery. Patient characteristics and perioperative outcomes were analyzed. Operative time and CUSUM plots were used for evaluating the learning curve for laparoscopic versus robotic LAR. Laparoscopic versus robotic LAR outcomes feature initial group operative times of 308 (291-325) min versus 397 (373-420) min and last group times of 220 (212-229) min versus 204 (196-211) min-reversed in favor of robotics; major complications of 4.7 versus 6.5 % (NS), resection margin involvement of 2.8 versus 4.4 % (NS), conversion rate of 3.8 versus 1.1 (NS), lymph node harvest of 16.3 versus 17.2 (NS), and estimated blood loss of 231 versus 201 cc (NS). Due to faster learning curves for extracorporeal phase and total mesorectal excision phase, the robotic surgery was observed to be faster than laparoscopic surgery after the initial 41 cases. CUSUM plots demonstrate acceptable perioperative surgical outcomes from the beginning of the study. Initial robotic operative times improved with practice rapidly and eventually became faster than those for laparoscopy. Developing both laparoscopic and robotic skills simultaneously can provide acceptable perioperative outcomes in rectal surgery. It might be suggested that in the current milieu of clashing interests between evolving technology and economic constrains, there might be advantages in embracing both approaches.

The learning curve of laparoscopic liver resection after the Louisville statement 2008: Will it be more effective and smooth?

PubMed

Lin, Chung-Wei; Tsai, Tzu-Jung; Cheng, Tsung-Yen; Wei, Hung-Kuang; Hung, Chen-Fang; Chen, Yin-Yin; Chen, Chii-Ming

2016-07-01

Laparoscopic liver resection (LLR) has been proven to be feasible and safe. However, it is a difficult and complex procedure with a steep learning curve. The aim of this study was to evaluate the learning curve of LLR at our institutions since 2008. One hundred and twenty-six consecutive LLRs were included from May 2008 to December 2014. Patient characteristics, operative data, and surgical outcomes were collected prospectively and analyzed. The median tumor size was 25 mm (range 5-90 mm), and 96 % of the resected tumors were malignant. 41.3 % (52/126) of patients had pathologically proven liver cirrhosis. The median operation time was 216 min (range 40-602 min) with a median blood loss of 100 ml (range 20-2300 ml). The median length of hospital stay was 4 days (range 2-10 days). Six major postoperative complications occurred in this series, and there was no 90-day postoperative mortality. Regarding the incidence of major operative events including operation time longer than 300 min, perioperative blood loss above 500 ml, and major postoperative complications, the learning curve [as evaluated by the cumulative sum (CUSUM) technique] showed its first reverse after 22 cases. The indication of laparoscopic resection in this series extended after 60 cases to include tumors located in difficult locations (segments 4a, 7, 8) and major hepatectomy. CUSUM showed that the incidence of major operative events proceeded to increase again, and the second reverse was noted after an additional 40 cases of experience. Location of the tumor in a difficult area emerged as a significant predictor of major operative events. In carefully selected patients, CUSUM analysis showed 22 cases were needed to overcome the learning curve for minor LLR.

Laparoscopic varicocelectomy: virtual reality training and learning curve.

PubMed

Wang, Zheng; Ni, Yuhua; Zhang, Yinan; Jin, Xunbo; Xia, Qinghua; Wang, Hanbo

2014-01-01

To explore the role that virtual reality training might play in the learning curve of laparoscopic varicocelectomy. A total of 1326 laparoscopic varicocelectomy cases performed by 16 participants from July 2005 to June 2012 were retrospectively analyzed. The participants were divided into 2 groups: group A was trained by laparoscopic trainer boxes; group B was trained by a virtual reality training course preoperatively. The operation time curves were drafted, and the learning, improving, and platform stages were divided and statistically confirmed. The operation time and number of cases in the learning and improving stages of both groups were compared. Testicular artery sparing failure and postoperative hydroceles rate were statistically analyzed for the confirmation of the learning curve. The learning curve of laparoscopic varicocelectomy was 15 cases, and with 14 cases more, it came into the platform stage. The number of cases for the learning stages of both groups showed no statistical difference (P=.49), but the operation time of group B for the learning stage was less than that of group A (P<.00001). The number of cases of group B for the improving stage was significantly less than that of group A (P=.005), but the operation time of both groups in the improving stage showed no difference (P=.30). The difference of testicular artery sparing failure rates among these 3 stages was proved significant (P<.0001), the postoperative hydroceles rate showed no statistical difference (P=.60). The virtual reality training shortened the operation time in the learning stage and hastened the trainees' steps in the improving stage, but did not shorten the learning curve as expected to.

Mentorship, learning curves, and balance.

PubMed

Cohen, Meryl S; Jacobs, Jeffrey P; Quintessenza, James A; Chai, Paul J; Lindberg, Harald L; Dickey, Jamie; Ungerleider, Ross M

2007-09-01

Professionals working in the arena of health care face a variety of challenges as their careers evolve and develop. In this review, we analyze the role of mentorship, learning curves, and balance in overcoming challenges that all such professionals are likely to encounter. These challenges can exist both in professional and personal life. As any professional involved in health care matures, complex professional skills must be mastered, and new professional skills must be acquired. These skills are both technical and judgmental. In most circumstances, these skills must be learned. In 2007, despite the continued need for obtaining new knowledge and learning new skills, the professional and public tolerance for a "learning curve" is much less than in previous decades. Mentorship is the key to success in these endeavours. The success of mentorship is two-sided, with responsibilities for both the mentor and the mentee. The benefits of this relationship must be bidirectional. It is the responsibility of both the student and the mentor to assure this bidirectional exchange of benefit. This relationship requires time, patience, dedication, and to some degree selflessness. This mentorship will ultimately be the best tool for mastering complex professional skills and maturing through various learning curves. Professional mentorship also requires that mentors identify and explicitly teach their mentees the relational skills and abilities inherent in learning the management of the triad of self, relationships with others, and professional responsibilities.Up to two decades ago, a learning curve was tolerated, and even expected, while professionals involved in healthcare developed the techniques that allowed for the treatment of previously untreatable diseases. Outcomes have now improved to the point that this type of learning curve is no longer acceptable to the public. Still, professionals must learn to perform and develop independence and confidence. The responsibility to meet this challenge without a painful learning curve belongs to both the younger professionals, who must progress through the learning curve, and the more mature professionals who must create an appropriate environment for learning. In addition to mentorship, the detailed tracking of outcomes is an essential tool for mastering any learning curve. It is crucial to utilize a detailed database to track outcomes, to learn, and to protect both yourself and your patients. It is our professional responsibility to engage in self-evaluation, in part employing voluntary sharing of data. For cardiac surgical subspecialties, the databases now existing for The European Association for CardioThoracic Surgery and The Society of Thoracic Surgeons represent the ideal tool for monitoring outcomes. Evolving initiatives in the fields of paediatric cardiology, paediatric critical care, and paediatric cardiac anaesthesia will play similar roles.A variety of professional and personal challenges must be met by all those working in health care. The acquisition of learned skills, and the use of special tools, will facilitate the process of conquering these challenges. Choosing appropriate role models and mentors can help progression through any learning curve in a controlled and protected fashion. Professional and personal satisfaction are both necessities. Finding the satisfactory balance between work and home life is difficult, but possible with the right tools, organization skills, and support system at work and at home. The concepts of mentorship, learning curves and balance cannot be underappreciated.

Holmium laser enucleation of the prostate and retropubic prostatic adenomectomy: morbidity analysis and anesthesia considerations.

PubMed

Soto-Mesa, D; Amorín-Díaz, M; Pérez-Arviza, L; Fernández-Pello Montes, S; Martín-Huéscar, A

2015-11-01

Holmium laser enucleation of the prostate (HoLEP) is an alternative to prostatic adenomectomy for the surgical treatment of benign prostatic hypertrophy. We analyzed our learning curve for this technique, and we compared it in a secondary manner with prostatic adenomectomy. A retrospective comparative study was conducted that included the first 100 cases of HoLEP performed in our center and the latest 50 cases of retropubic adenomectomy. We collected data on the patients, the surgery, the anesthesia, the perioperative variables, the anesthesia complications and the postoperative variables, with a 6-month follow-up. We analyzed the learning curve without mentors for HoLEP and compared the characteristics of HoLEP in 2 separate phases (learning and stabilization phases) with the latest retropubic prostatic adenomectomies performed. Intradural anesthesia was the most common technique. The transfusion needs, length of stay (P<.01) and postoperative morbidity were lower for HoLEP than for adenomectomy. However, the retropubic adenomectomy group had larger initial prostate volumes (P<.001) and shorter surgical times (P<.001). Better surgical performance (P<.001) and a lower incidence of complications were observed in the HoLEP-B group (once the learning curve had been overcome) compared with the HoLEP-A group. In our center, HoLEP was introduced as a valid alternative to open retropubic adenomectomy, with excellent results in terms of morbidity and reduced hospital stay. In terms of the learning curve, we consider that approximately 50 patients (without mentor) is an appropriate cutoff. Local anesthesia is a good choice for the anesthesia technique. Copyright © 2014 AEU. Publicado por Elsevier España, S.L.U. All rights reserved.

Comparison of the Operative Outcomes and Learning Curves between Laparoscopic and Robotic Gastrectomy for Gastric Cancer

PubMed Central

Huang, Kuo-Hung; Lan, Yuan-Tzu; Fang, Wen-Liang; Chen, Jen-Hao; Lo, Su-Shun; Li, Anna Fen-Yau; Chiou, Shih-Hwa; Wu, Chew-Wun; Shyr, Yi-Ming

2014-01-01

Background Minimally invasive surgery, including laparoscopic and robotic gastrectomy, has become more popular in the treatment of gastric cancer. However, few studies have compared the learning curves between laparoscopic and robotic gastrectomy for gastric cancer. Methods Data were prospectively collected between July 2008 and Aug 2014. A total of 145 patients underwent minimally invasive gastrectomy for gastric cancer by a single surgeon, including 73 laparoscopic and 72 robotic gastrectomies. The clinicopathologic characteristics, operative outcomes and learning curves were compared between the two groups. Results Compared with the laparoscopic group, the robotic group was associated with less blood loss and longer operative time. After the surgeon learning curves were overcome for each technique, the operative outcomes became similar between the two groups except longer operative time in the robotic group. After accumulating more cases of robotic gastrectomy, the operative time in the laparoscopic group decreased dramatically. Conclusions After overcoming the learning curves, the operative outcomes became similar between laparoscopic and robotic gastrectomy. The experience of robotic gastrectomy could affect the learning process of laparoscopic gastrectomy. PMID:25360767

Preliminary experience and learning curve for laparoendoscopic single-site retroperitoneal pyeloplasty.

PubMed

Ou, Zhenyu; Qi, Lin; Yang, Jinrui; Chen, Xiang; Cao, Zhenzhen; Zu, Xiongbing; Liu, Longfei; Wang, Long

2013-09-01

To report our preliminary experience and to assess the learning curve for laparoendoscopic single-site retroperitoneal pyeloplasty (LESS-RP) for ureteropelvic junction obstruction (UPJO). From July 2010 to February 2012, LESS-RP was performed in 27 patients affected with UPJO by a single surgeon. A homemade single-access platform and both conventional and prebent instruments were applied. Patient characteristics and perioperative outcomes were analyzed. The cumulative sum (CUSUM) method was used to evaluate the learning curve. The LESS-RP was successfully accomplished in all 27 patients. The mean operative time (OT) was 175.9±22.5 minutes, and the mean estimated blood loss was 83.3±27.1 mL. We used the OT as a proxy to assess the learning curve. The CUSUM learning curve can be divided into two distinct phases: the initial 12 cases and the last 15 cases. There were significant differences in the mean OT (195.6 minutes versus 159.1 minutes, P<.001) and mean estimated blood loss (97.2 mL versus 72.2 mL, P=.014) between the two phases. The two phases did not differ in other parameters. LESS-RP is a safe and feasible procedure. The learning curve of a single surgeon suggests that the initial learning phase for LESS-RP can be completed after approximately 12 cases.

Assessment of Surgical Learning Curves in Transoral Robotic Surgery for Squamous Cell Carcinoma of the Oropharynx

PubMed Central

Albergotti, William G.; Gooding, William E.; Kubik, Mark W.; Geltzeiler, Mathew; Kim, Seungwon; Duvvuri, Umamaheswar; Ferris, Robert L.

2017-01-01

IMPORTANCE Transoral robotic surgery (TORS) is increasingly employed as a treatment option for squamous cell carcinoma of the oropharynx (OPSCC). Measures of surgical learning curves are needed particularly as clinical trials using this technology continue to evolve. OBJECTIVE To assess learning curves for the oncologic TORS surgeon and to identify the number of cases needed to identify the learning phase. DESIGN, SETTING, AND PARTICIPANTS A retrospective review of all patients who underwent TORS for OPSCC at the University of Pittsburgh Medical Center between March 2010 and March 2016. Cases were excluded for involvement of a subsite outside of the oropharynx, for nonmalignant abnormality or nonsquamous histology, unknown primary, no tumor in the main specimen, free flap reconstruction, and for an inability to define margin status. EXPOSURES Transoral robotic surgery for OPSCC. MAIN OUTCOMES AND MEASURES Primary learning measures defined by the authors include the initial and final margin status and time to resection of main surgical specimen. A cumulative sum learning curve was developed for each surgeon for each of the study variables. The inflection point of each surgeon’s curve was considered to be the point signaling the completion of the learning phase. RESULTS There were 382 transoral robotic procedures identified. Of 382 cases, 160 met our inclusion criteria: 68 for surgeon A, 37 for surgeon B, and 55 for surgeon C. Of the 160 included patients, 125 were men and 35 were women. The mean (SD) age of participants was 59.4 (9.5) years. Mean (SD) time to resection including robot set-up was 79 (36) minutes. The inflection points for the final margin status learning curves were 27 cases (surgeon A) and 25 cases (surgeon C). There was no inflection point for surgeon B for final margin status. Inflection points for mean time to resection were: 39 cases (surgeon A), 30 cases (surgeon B), and 27 cases (surgeon C). CONCLUSIONS AND RELEVANCE Using metrics of positive margin rate and time to resection of the main surgical specimen, the learning curve for TORS for OPSCC is surgeon-specific. Inflection points for most learning curves peak between 20 and 30 cases. PMID:28196200

CUSUM analysis of learning curves for the head-mounted microscope in phonomicrosurgery.

PubMed

Chen, Ting; Vamos, Andrew C; Dailey, Seth H; Jiang, Jack J

2016-10-01

To observe the learning curve of the head-mounted microscope in a phonomicrosurgery simulator using cumulative summation (CUSUM) analysis, which incorporates a magnetic phonomicrosurgery instrument tracking system (MPTS). Retrospective case series. Eight subjects (6 medical students and 2 surgeons inexperienced in phonomicrosurgery) operated on phonomicrosurgical simulation cutting tasks while using the head-mounted microscope for 400 minutes total. Two 20-minute sessions occurred each day for 10 total days, with operation quality (Qs ) and completion time (T) being recorded after each session. Cumulative summation analysis of Qs and T was performed by using subjects' performance data from trials completed using a traditional standing microscope as success criteria. The motion parameters from the head-mounted microscope were significantly better than the standing microscope (P < 0.01), but T was longer than that from the standing microscope (P < 0.01). No subject successfully adapted to the head-mounted microscope, as assessed by CUSUM analysis. Cumulative summation analysis can objectively monitor the learning process associated with a phonomicrosurgical simulator system, ultimately providing a tool to assess learning status. Also, motion parameters determined by our MPTS showed that, although the head-mounted microscope provides better motion control, worse Qs and longer T resulted. This decrease in Qs is likely a result of the relatively unstable visual environment that it provides. Overall, the inexperienced surgeons participating in this study failed to adapt to the head-mounted microscope in our simulated phonomicrosurgery environment. 4 Laryngoscope, 126:2295-2300, 2016. © 2015 The American Laryngological, Rhinological and Otological Society, Inc.

Learner Characteristic Based Learning Effort Curve Mode: The Core Mechanism on Developing Personalized Adaptive E-Learning Platform

ERIC Educational Resources Information Center

Hsu, Pi-Shan

2012-01-01

This study aims to develop the core mechanism for realizing the development of personalized adaptive e-learning platform, which is based on the previous learning effort curve research and takes into account the learner characteristics of learning style and self-efficacy. 125 university students from Taiwan are classified into 16 groups according…

Venous Grafts Procured During the Learning Curve for Endoscopic Veins Harvesting Show Compromised Vascular Remodeling

PubMed Central

Kiani, Soroosh; Desai, Pranjal H.; Thirumvalavan, Nannan; Kurian, Dinesh John; Flynn, Mary Margaret; Zhao, XiaoQing

2011-01-01

BACKGROUND Endoscopic vein harvest (EVH) is the US standard of care for CABG but recent comparisons to open harvest suggest that conduit quality and outcomes may be compromised. To test the hypothesis that problems with EVH may relate to its learning curve and conduit quality, we analyzed the quality and early function of conduits procured by technicians with varying EVH experience. METHODS EVH was performed during CABG by “experienced” (>900 cases, n=55 patients) vs. “novice” (<100 cases, n=30 patients) technicians. Afterwards, conduits were and examined for vascular injury using optical coherence tomography (OCT), with segments identified as injured further examined for gene expression using a tissue injury array. Conduit diameter was measured intra- and postoperatively (day 5 and 6 months) using OCT and Computed-Tomography angiography. RESULTS EVH performed by novice harvesters resulted in increased number of discrete graft injuries and higher expression of tissue injury genes. Regression analysis revealed an association between shear stress and early dilation (positive remodeling) (R2 =0.48, p <0.01). Injured veins showed blunted positive remodeling at 5 days and a greater degree of late lumen loss at 6 months. CONCLUSION Under normal conditions, intraluminal shear stress leads vein grafts to develop positive remodeling over the first postoperative week. Injury to conduits, a frequent sequela of the learning curve for EVH, was a predictor of early graft failure, blunted positive remodeling and greater negative remodeling. Given the ongoing annual volume of EVH cases, rigorous monitoring of the learning curve represents an important and unrecognized public health issue. PMID:21996436

Acquisition Challenge: The Importance of Incompressibility in Comparing Learning Curve Models

DTIC Science & Technology

2015-10-01

parameters for all four learning mod- els used in the study . The learning rate factor, b, is the slope of the linear regression line, which in this case is...incorporated within the DoD acquisition environment. This study tested three alternative learning models (the Stanford-B model, DeJong’s learning formula...appropriate tools to calculate accurate and reliable predictions. However, conventional learning curve methodology has been in practice since the pre

Space shuttle solid rocket booster cost-per-flight analysis technique

NASA Technical Reports Server (NTRS)

Forney, J. A.

1979-01-01

A cost per flight computer model is described which considers: traffic model, component attrition, hardware useful life, turnaround time for refurbishment, manufacturing rates, learning curves on the time to perform tasks, cost improvement curves on quantity hardware buys, inflation, spares philosophy, long lead, hardware funding requirements, and other logistics and scheduling constraints. Additional uses of the model include assessing the cost per flight impact of changing major space shuttle program parameters and searching for opportunities to make cost effective management decisions.

Efficacy of tension-free vaginal tape compared with transobturator tape in the treatment of stress urinary incontinence in women: analysis of learning curve, perioperative changes of voiding function

PubMed Central

2011-01-01

Background In this study, by comparing TVT surgery and TOT surgery for stress urinary incontinence in women, the characteristics and learning curves of both operative methods were studied. Methods A total of 83 women with stress urinary incontinence treated with tension-free vaginal tape (TVT) (n = 38) or transobturator tape (TOT) (n = 45) at Saiseikai Central Hospital between April 2004 and September 2009 were included. We compare the outcomes and learning curves between TVT surgery and TOT surgery. In statistical analysis, Student's t test, Fisher's exact test, and Mann-Whitney's U test were used. Results The surgical durations were 37.4 ± 15.7 minutes with TVT surgery and 31.0 ± 8.3 minutes with TOT surgery. A longer period of time was required for TVT surgery (p = 0.025). The residual urine at post-operative day 1 was higher in TVT surgery (25.9 ± 44.2 ml) than in TOT surgery (10.6 ± 19.2 ml) (p = 0.0452). The surgical duration of TVT surgery was shortened after the operator had performed 15 operations (p = 0.019). Conclusions In comparison of TVT surgery and TOT surgery, the surgical duration of TVT surgery was longer and the residual urine of TVT surgery was higher at post-operative day 1. Surgical experience could shorten the duration of TVT surgery. PMID:21726448

A Primer on the Statistical Modelling of Learning Curves in Health Professions Education

ERIC Educational Resources Information Center

Pusic, Martin V.; Boutis, Kathy; Pecaric, Martin R.; Savenkov, Oleksander; Beckstead, Jason W.; Jaber, Mohamad Y.

2017-01-01

Learning curves are a useful way of representing the rate of learning over time. Features include an index of baseline performance (y-intercept), the efficiency of learning over time (slope parameter) and the maximal theoretical performance achievable (upper asymptote). Each of these parameters can be statistically modelled on an individual and…

Nonlinear dynamics of motor learning.

PubMed

Mayer-Kress, Gottfried; Newell, Karl M; Liu, Yeou-Teh

2009-01-01

In this paper we review recent work from our studies of a nonlinear dynamics of motor learning that is grounded in the construct of an evolving attractor landscape. With the assumption that learning is goal-directed, we can quantify the observed performance as a score or measure of the distance to the learning goal. The structure of the dynamics of how the goal is approached has been traditionally studied through an analysis of learning curves. Recent years have seen a gradual paradigm shift from a 'universal power law of practice' to an analysis of performance dynamics that reveals multiple processes that include adaption and learning as well as changes in performance due to factors such as fatigue. Evidence has also been found for nonlinear phenomena such as bifurcations, hysteresis and even a form of self-organized criticality. Finally, we present a quantitative measure for the dual concepts of skill and difficulty that allows us to unfold a learning process in order to study universal properties of learning transitions.

How to Avoid a Learning Curve in Stapedotomy: A Standardized Surgical Technique.

PubMed

Kwok, Pingling; Gleich, Otto; Dalles, Katharina; Mayr, Elisabeth; Jacob, Peter; Strutz, Jürgen

2017-08-01

To evaluate, whether a learning curve for beginners in stapedotomy can be avoided by using a prosthesis with thermal memory-shape attachment in combination with a standardized laser-assisted surgical technique. Retrospective case review. Tertiary referral center. Fifty-eight ears were operated by three experienced surgeons and compared with a group of 12 cases operated by a beginner in stapedotomy. Stapedotomy. Difference of pure-tone audiometry thresholds measured before and after surgery. The average postoperative gain for air conduction in the frequencies below 2 kHz was 20 to 25 dB and decreased for the higher frequencies. Using the Mann-Whitney-U test for comparing mean gain between experienced and inexperienced surgeons showed no significant difference (p = 0.281 at 4 kHz and p > 0.7 for the other frequencies). A Spearman rank correlation of the postoperative gain for air- and bone-conduction thresholds was obtained at each test frequency for the first 12 patients consecutively treated with a thermal memory-shape attachment prosthesis by two experienced and one inexperienced surgeon. This analysis does not support the hypothesis of a "learning effect" that should be associated with an improved outcome for successively treated patients. It is possible to avoid a learning curve in stapes surgery by applying a thermal memory-shape prosthesis in a standardized laser-assisted surgical procedure.

Cumulative sum analysis for experiences of a single-session retrograde intrarenal stone surgery and analysis of predictors for stone-free status.

PubMed

Cho, Sung Yong; Choo, Min Soo; Jung, Jae Hyun; Jeong, Chang Wook; Oh, Sohee; Lee, Seung Bae; Son, Hwancheol; Jeong, Hyeon

2014-01-01

This study investigated the learning curve of a single-session retrograde intrarenal surgery (RIRS) in patients with mid-sized stones. Competence and trainee proficiency for RIRS was assessed using cumulative sum analysis (CUSUM). The study design and the use of patients' information stored in the hospital database were approved by the Institutional Review Board of our institution. A retrospective review was performed for 100 patients who underwent a single-session RIRS. Patients were included if the main stone had a maximal diameter between 10 and 30 mm. The presence of a residual stone was checked on postoperative day 1 and at one-month follow-up visit. Fragmentation efficacy was calculated "removed stone volume (mm(3)) divided by operative time (min)". CUSUM analysis was used for monitoring change in fragmentation efficacy, and we tested whether or not acceptable surgical outcomes were achieved. The mean age was 54.7±14.8 years. Serum creatinine level did not change significantly. Estimated GFR and hemoglobin were within normal limits postoperatively. The CUSUM curve tended to be flat until the 25th case and showed a rising pattern but declined again until the 56th case. After that point, the fragmentation efficacy reached a plateau. The acceptable level of fragmentation efficacy was 25 ml/min. Multivariate logistic regression analyses showed that stone-free rate was significantly lower for cases with multiple stones than those with a single stone (OR = 0.147, CI 0.032 - 0.674, P value  = 0.005) and for cases with higher number of sites (OR = 0.676, CI 0.517 - 0.882, P value  = 0.004). The statistical analysis of RIRS learning experience revealed that 56 cases were required for reaching a plateau in the learning curve. The number of stones and the number of sites were significant predictors for stone-free status.

Variation in Aptitude of Trainees in Endoscopic Ultrasonography, Based on Cumulative Sum Analysis

PubMed Central

Wani, Sachin; Hall, Matthew; Keswani, Rajesh N.; Aslanian, Harry R.; Casey, Brenna; Burbridge, Rebecca; Chak, Amitabh; Chen, Ann M.; Cote, Gregory; Edmundowicz, Steven A.; Faulx, Ashley L.; Hollander, Thomas G.; Lee, Linda S.; Mullady, Daniel; Murad, Faris; Muthusamy, Raman; Pfau, Patrick R.; Scheiman, James M.; Tokar, Jeffrey; Wagh, Mihir S.; Watson, Rabindra; Early, Dayna

2017-01-01

BACKGROUND & AIMS Studies have reported substantial variation in the competency of advanced endoscopy trainees, indicating a need for more supervised training in endoscopic ultrasound (EUS). We used a standardized, validated, data collection tool to evaluate learning curves and measure competency in EUS among trainees at multiple centers. METHODS In a prospective study performed at 15 centers, 17 trainees with no prior EUS experience were evaluated by experienced attending endosonographers at the 25th and then every 10th upper EUS examination, over a 12-month training period. A standardized data collection form was used (using a 5-point scoring system) to grade the EUS examination. Cumulative sum analysis was applied to produce a learning curve for each trainee; it tracked the overall performance based on median scores at different stations and also at each station. Competency was defined by a median score of 1, with acceptable and unacceptable failure rates of 10% and 20%, respectively. RESULTS Twelve trainees were included in the final analysis. Each of the trainees performed 265 to 540 EUS examinations (total, 4257 examinations). There was a large amount of variation in their learning curves: 2 trainees crossed the threshold for acceptable performance (at cases 225 and 245), 2 trainees had a trend toward acceptable performance (after 289 and 355 cases) but required continued observation, and 8 trainees needed additional training and observation. Similar results were observed at individual stations. CONCLUSIONS A specific case load does not ensure competency in EUS; 225 cases should be considered the minimum caseload for training because we found that no trainee achieved competency before this point. Ongoing training should be provided for trainees until competency is confirmed using objective measures. PMID:25460557

Variation in Aptitude of Trainees in Endoscopic Ultrasonography, Based on Cumulative Sum Analysis.

PubMed

Wani, Sachin; Hall, Matthew; Keswani, Rajesh N; Aslanian, Harry R; Casey, Brenna; Burbridge, Rebecca; Chak, Amitabh; Chen, Ann M; Cote, Gregory; Edmundowicz, Steven A; Faulx, Ashley L; Hollander, Thomas G; Lee, Linda S; Mullady, Daniel; Murad, Faris; Muthusamy, V Raman; Pfau, Patrick R; Scheiman, James M; Tokar, Jeffrey; Wagh, Mihir S; Watson, Rabindra; Early, Dayna

2015-07-01

Studies have reported substantial variation in the competency of advanced endoscopy trainees, indicating a need for more supervised training in endoscopic ultrasound (EUS). We used a standardized, validated, data collection tool to evaluate learning curves and measure competency in EUS among trainees at multiple centers. In a prospective study performed at 15 centers, 17 trainees with no prior EUS experience were evaluated by experienced attending endosonographers at the 25th and then every 10th upper EUS examination, over a 12-month training period. A standardized data collection form was used (using a 5-point scoring system) to grade the EUS examination. Cumulative sum analysis was applied to produce a learning curve for each trainee; it tracked the overall performance based on median scores at different stations and also at each station. Competency was defined by a median score of 1, with acceptable and unacceptable failure rates of 10% and 20%, respectively. Twelve trainees were included in the final analysis. Each of the trainees performed 265 to 540 EUS examinations (total, 4257 examinations). There was a large amount of variation in their learning curves: 2 trainees crossed the threshold for acceptable performance (at cases 225 and 245), 2 trainees had a trend toward acceptable performance (after 289 and 355 cases) but required continued observation, and 8 trainees needed additional training and observation. Similar results were observed at individual stations. A specific case load does not ensure competency in EUS; 225 cases should be considered the minimum caseload for training because we found that no trainee achieved competency before this point. Ongoing training should be provided for trainees until competency is confirmed using objective measures. Copyright © 2015 AGA Institute. Published by Elsevier Inc. All rights reserved.

Rapid and safe learning of robotic gastrectomy for gastric cancer: multidimensional analysis in a comparison with laparoscopic gastrectomy.

PubMed

Kim, H-I; Park, M S; Song, K J; Woo, Y; Hyung, W J

2014-10-01

The learning curve of robotic gastrectomy has not yet been evaluated in comparison with the laparoscopic approach. We compared the learning curves of robotic gastrectomy and laparoscopic gastrectomy based on operation time and surgical success. We analyzed 172 robotic and 481 laparoscopic distal gastrectomies performed by single surgeon from May 2003 to April 2009. The operation time was analyzed using a moving average and non-linear regression analysis. Surgical success was evaluated by a cumulative sum plot with a target failure rate of 10%. Surgical failure was defined as laparoscopic or open conversion, insufficient lymph node harvest for staging, resection margin involvement, postoperative morbidity, and mortality. Moving average and non-linear regression analyses indicated stable state for operation time at 95 and 121 cases in robotic gastrectomy, and 270 and 262 cases in laparoscopic gastrectomy, respectively. The cumulative sum plot identified no cut-off point for surgical success in robotic gastrectomy and 80 cases in laparoscopic gastrectomy. Excluding the initial 148 laparoscopic gastrectomies that were performed before the first robotic gastrectomy, the two groups showed similar number of cases to reach steady state in operation time, and showed no cut-off point in analysis of surgical success. The experience of laparoscopic surgery could affect the learning process of robotic gastrectomy. An experienced laparoscopic surgeon requires fewer cases of robotic gastrectomy to reach steady state. Moreover, the surgical outcomes of robotic gastrectomy were satisfactory. Copyright © 2013 Elsevier Ltd. All rights reserved.

A service-based BLAST command tool supported by cloud infrastructures.

PubMed

Carrión, Abel; Blanquer, Ignacio; Hernández, Vicente

2012-01-01

Notwithstanding the benefits of distributed-computing infrastructures for empowering bioinformatics analysis tools with the needed computing and storage capability, the actual use of these infrastructures is still low. Learning curves and deployment difficulties have reduced the impact on the wide research community. This article presents a porting strategy of BLAST based on a multiplatform client and a service that provides the same interface as sequential BLAST, thus reducing learning curve and with minimal impact on their integration on existing workflows. The porting has been done using the execution and data access components from the EC project Venus-C and the Windows Azure infrastructure provided in this project. The results obtained demonstrate a low overhead on the global execution framework and reasonable speed-up and cost-efficiency with respect to a sequential version.

Analysis of the learning curve for pre-cut corneal specimens in preparation for lamellar transplantation: a prospective, single-centre, consecutive case series prepared at the Lions New South Wales Eye Bank.

PubMed

Martin, Aifric Isabel; Devasahayam, Rajnesh; Hodge, Christopher; Cooper, Simon; Sutton, Gerard L

2017-09-01

This study is the first paper to establish a learning curve by a single technician. Preparation of pre-cut corneal endothelial grafts commenced at Lions New South Wales Eye Bank in December 2014. The primary objective of this study was to review the safety and reliability of the preparation method during the first year of production. This is a hospital-based, prospective case series. There were 234 consecutive donor corneal lenticules. Donor lenticules were prepared by a single operator using a linear cutting microkeratome. Immediately prior to cutting, central corneal thickness values were recorded. Measurements of the corneal bed were taken immediately following lenticule preparation. Outcomes were separated by blade sizes, and intended thickness was compared to actual thickness for each setting. Early specimens were compared to later ones to assess for a learning curve within the technique. The main parameter measured is the mean difference from intended lamellar cut thickness. The mean final cut thickness was 122.36 ± 20.35 μm, and the mean difference from intended cut was 30.17 ± 37.45 μm. No significant difference was found between results achieved with early specimens versus those achieved with later specimens (P = 0.425). Thin, reproducible endothelial grafts can routinely be produced by trained technicians at their respective eye banks without significant concerns for an extended learning curve. This service can reduce perioperative surgical complexity, required surgical paraphernalia and theatre times. The consistent preparation of single-pass, ultrathin pre-cut corneas may have additional advantages for surgeons seeking to introduce lamellar techniques. © 2017 Royal Australian and New Zealand College of Ophthalmologists.

Value of the cumulative sum test for the assessment of a learning curve: Application to the introduction of patient-specific instrumentation for total knee arthroplasty in an academic department.

PubMed

De Gori, Marco; Adamczewski, Benjamin; Jenny, Jean-Yves

2017-06-01

The purpose of the study was to use the cumulative summation (CUSUM) test to assess the learning curve during the introduction of a new surgical technique (patient-specific instrumentation) in total knee arthroplasty (TKA) in an academic department. The first 50TKAs operated on at an academic department using patient-specific templates (PSTs) were scheduled to enter the study. All patients had a preoperative computed tomography scan evaluation to plan bone resections. The PSTs were positioned intraoperatively according to the best-fit technique and their three-dimensional orientation was recorded by a navigation system. The position of the femur and tibia PST was compared to the planned position for four items for each component: coronal and sagittal orientation, medial and lateral height of resection. Items were summarized to obtain knee, femur and tibia PST scores, respectively. These scores were plotted according to chronological order and included in a CUSUM analysis. The tested hypothesis was that the PST process for TKA was immediately under control after its introduction. CUSUM test showed that positioning of the PST significantly differed from the target throughout the study. There was a significant difference between all scores and the maximal score. No case obtained the maximal score of eight points. The study was interrupted after 20 cases because of this negative evaluation. The CUSUM test is effective in monitoring the learning curve when introducing a new surgical procedure. Introducing PST for TKA in an academic department may be associated with a long-lasting learning curve. The study was registered on the clinical.gov website (Identifier NCT02429245). Copyright © 2017 Elsevier B.V. All rights reserved.

Learning Curve Analysis and Surgical Outcomes of Single-port Laparoscopic Myomectomy.

PubMed

Lee, Hee Jun; Kim, Ju Yeong; Kim, Seul Ki; Lee, Jung Ryeol; Suh, Chang Suk; Kim, Seok Hyun

2015-01-01

To identify learning curves for single-port laparoscopic myomectomy (SPLM) and evaluate surgical outcomes according to the sequence of operation. A retrospective study. A university-based hospital (Canadian Task Force classification II-2). The medical records from 205 patients who had undergone SPLM from October 2009 to May 2013 were reviewed. Because the myomectomy time was significantly affected by the size and number of myomas removed by SPLM, cases in which 2 or more of the myomas removed were >7 cm in diameter were excluded. Furthermore, cases involving additional operations performed simultaneously (e.g., ovarian or hysteroscopic surgery) were also excluded. A total of 161 cases of SPLM were included. None. We assessed the SPLM learning curve via a graph based on operation time versus sequence of cases. Patients were chronologically arranged according to their surgery dates and were then placed into 1 of 4 groups according to their operation sequence. SPLM was completed successfully in 160 of 161 cases (99.4%). One case was converted to multiport surgery. Basal characteristics of the patients between the 4 groups did not differ. The median operation times for the 4 groups were 112.0, 92.8, 83.7, and 90.0 minutes, respectively. Operation time decreased significantly in the second, third, and fourth groups compared with that in the first group (p < .001). Proficiency, which is the point at which the slope of the learning curve became less steep, was evident after about 45 operations. Results from the current study suggested that proficiency for SPLM was achieved after about 45 operations. Additionally, operation time decreased with experience without an increase in complication rate. Copyright © 2015 AAGL. Published by Elsevier Inc. All rights reserved.

Learning Curve for Seawater Reverse Osmosis Desalination Plants: Capital Cost Trend of the Past, Present, and Future

NASA Astrophysics Data System (ADS)

Caldera, Upeksha; Breyer, Christian

2017-12-01

Seawater reverse osmosis (SWRO) desalination is expected to play a pivotal role in helping to secure future global water supply. While the global reliance on SWRO plants for water security increases, there is no consensus on how the capital costs of SWRO plants will vary in the future. The aim of this paper is to analyze the past trends of the SWRO capital expenditures (capex) as the historic global cumulative online SWRO capacity increases, based on the learning curve concept. The SWRO capex learning curve is found based on 4,237 plants that came online from 1977 to 2015. A learning rate of 15% is determined, implying that the SWRO capex reduced by 15% when the cumulative capacity was doubled. Based on SWRO capacity annual growth rates of 10% and 20%, by 2030, the global average capex of SWRO plants is found to fall to 1,580 USD/(m3/d) and 1,340 USD/(m3/d), respectively. A learning curve for SWRO capital costs has not been presented previously. This research highlights the potential for decrease in SWRO capex with the increase in installation of SWRO plants and the value of the learning curve approach to estimate future SWRO capex.

Optimizing area under the ROC curve using semi-supervised learning

PubMed Central

Wang, Shijun; Li, Diana; Petrick, Nicholas; Sahiner, Berkman; Linguraru, Marius George; Summers, Ronald M.

2014-01-01

Receiver operating characteristic (ROC) analysis is a standard methodology to evaluate the performance of a binary classification system. The area under the ROC curve (AUC) is a performance metric that summarizes how well a classifier separates two classes. Traditional AUC optimization techniques are supervised learning methods that utilize only labeled data (i.e., the true class is known for all data) to train the classifiers. In this work, inspired by semi-supervised and transductive learning, we propose two new AUC optimization algorithms hereby referred to as semi-supervised learning receiver operating characteristic (SSLROC) algorithms, which utilize unlabeled test samples in classifier training to maximize AUC. Unlabeled samples are incorporated into the AUC optimization process, and their ranking relationships to labeled positive and negative training samples are considered as optimization constraints. The introduced test samples will cause the learned decision boundary in a multidimensional feature space to adapt not only to the distribution of labeled training data, but also to the distribution of unlabeled test data. We formulate the semi-supervised AUC optimization problem as a semi-definite programming problem based on the margin maximization theory. The proposed methods SSLROC1 (1-norm) and SSLROC2 (2-norm) were evaluated using 34 (determined by power analysis) randomly selected datasets from the University of California, Irvine machine learning repository. Wilcoxon signed rank tests showed that the proposed methods achieved significant improvement compared with state-of-the-art methods. The proposed methods were also applied to a CT colonography dataset for colonic polyp classification and showed promising results.1 PMID:25395692

Optimizing area under the ROC curve using semi-supervised learning.

PubMed

Wang, Shijun; Li, Diana; Petrick, Nicholas; Sahiner, Berkman; Linguraru, Marius George; Summers, Ronald M

2015-01-01

Receiver operating characteristic (ROC) analysis is a standard methodology to evaluate the performance of a binary classification system. The area under the ROC curve (AUC) is a performance metric that summarizes how well a classifier separates two classes. Traditional AUC optimization techniques are supervised learning methods that utilize only labeled data (i.e., the true class is known for all data) to train the classifiers. In this work, inspired by semi-supervised and transductive learning, we propose two new AUC optimization algorithms hereby referred to as semi-supervised learning receiver operating characteristic (SSLROC) algorithms, which utilize unlabeled test samples in classifier training to maximize AUC. Unlabeled samples are incorporated into the AUC optimization process, and their ranking relationships to labeled positive and negative training samples are considered as optimization constraints. The introduced test samples will cause the learned decision boundary in a multidimensional feature space to adapt not only to the distribution of labeled training data, but also to the distribution of unlabeled test data. We formulate the semi-supervised AUC optimization problem as a semi-definite programming problem based on the margin maximization theory. The proposed methods SSLROC1 (1-norm) and SSLROC2 (2-norm) were evaluated using 34 (determined by power analysis) randomly selected datasets from the University of California, Irvine machine learning repository. Wilcoxon signed rank tests showed that the proposed methods achieved significant improvement compared with state-of-the-art methods. The proposed methods were also applied to a CT colonography dataset for colonic polyp classification and showed promising results.

Analysis of Laparoscopic Sleeve Gastrectomy Learning Curve and Its Influence on Procedure Safety and Perioperative Complications.

PubMed

Major, Piotr; Wysocki, Michał; Dworak, Jadwiga; Pędziwiatr, Michał; Pisarska, Magdalena; Wierdak, Mateusz; Zub-Pokrowiecka, Anna; Natkaniec, Michał; Małczak, Piotr; Nowakowski, Michał; Budzyński, Andrzej

2018-06-01

Laparoscopic sleeve gastrectomy (LSG) has become an attractive bariatric procedure with promising treatment effects yet amount of data regarding institutional learning process is limited. Retrospective study included patients submitted to LSG at academic teaching hospital. Patients were divided into groups every 100 consecutive patients. LSG introduction was structured along with Enhanced Recovery after Surgery (ERAS) treatment protocol. Primary endpoint was determining the LSG learning curve's stabilization point, using operative time, intraoperative difficulties, intraoperative adverse events (IAE), and number of stapler firings. Secondary endpoints: influence on perioperative complications and reoperations. Five hundred patients were included (330 females, median age of 40 (33-49) years). Operative time in G1-G2 differed significantly from G3-G5. Stabilization point was the 200th procedure using operative time. Intraoperative difficulties of G1 differed significantly from G2-G5, with stabilization after the 100th procedure. IAE and number of stapler firings could not be used as predictor. Based on perioperative morbidity, the learning curve was stabilized at the 100th procedure. The morbidity rates in the groups were G1, 13%; G2, 4%; G3, 5%; G4, 5%; and G5, 2%. The reoperation rate in G1 was 3%; G2, 2%; G3, 2%; G4, 1%; and G5, 0%. The institutional learning process stabilization point for LSG in a newly established bariatric center is between the 100th and 200th operation. Initially, the morbidity rate is high, which should concern surgeons who are willing to perform bariatric surgery.

A learning curve-based method to implement multifunctional work teams in the Brazilian footwear sector.

PubMed

Guimarães, L B de M; Anzanello, M J; Renner, J S

2012-05-01

This paper presents a method for implementing multifunctional work teams in a footwear company that followed the Taylor/Ford system for decades. The suggested framework first applies a Learning Curve (LC) modeling to assess whether rotation between tasks of different complexities affects workers' learning rate and performance. Next, the Macroergonomic Work Analysis (MA) method (Guimarães, 1999, 2009) introduces multifunctional principles in work teams towards workers' training and resources improvement. When applied to a pilot line consisting of 100 workers, the intervention-reduced work related accidents in 80%, absenteeism in 45.65%, and eliminated work related musculoskeletal disorders (WMSD), medical consultations, and turnover. Further, the output rate of the multifunctional team increased average 3% compared to the production rate of the regular lines following the Taylor/Ford system (with the same shoe model being manufactured), while the rework and spoilage rates were reduced 85% and 69%, respectively. Copyright © 2011 Elsevier Ltd and The Ergonomics Society. All rights reserved.

Establishing the minimal number of virtual reality simulator training sessions necessary to develop basic laparoscopic skills competence: evaluation of the learning curve.

PubMed

Duarte, Ricardo Jordão; Cury, José; Oliveira, Luis Carlos Neves; Srougi, Miguel

2013-01-01

Medical literature is scarce on information to define a basic skills training program for laparoscopic surgery (peg and transferring, cutting, clipping). The aim of this study was to determine the minimal number of simulator sessions of basic laparoscopic tasks necessary to elaborate an optimal virtual reality training curriculum. Eleven medical students with no previous laparoscopic experience were spontaneously enrolled. They were submitted to simulator training sessions starting at level 1 (Immersion Lap VR, San Jose, CA), including sequentially camera handling, peg and transfer, clipping and cutting. Each student trained twice a week until 10 sessions were completed. The score indexes were registered and analyzed. The total of errors of the evaluation sequences (camera, peg and transfer, clipping and cutting) were computed and thereafter, they were correlated to the total of items evaluated in each step, resulting in a success percent ratio for each student for each set of each completed session. Thereafter, we computed the cumulative success rate in 10 sessions, obtaining an analysis of the learning process. By non-linear regression the learning curve was analyzed. By the non-linear regression method the learning curve was analyzed and a r2 = 0.73 (p < 0.001) was obtained, being necessary 4.26 (∼five sessions) to reach the plateau of 80% of the estimated acquired knowledge, being that 100% of the students have reached this level of skills. From the fifth session till the 10th, the gain of knowledge was not significant, although some students reached 96% of the expected improvement. This study revealed that after five simulator training sequential sessions the students' learning curve reaches a plateau. The forward sessions in the same difficult level do not promote any improvement in laparoscopic basic surgical skills, and the students should be introduced to a more difficult training tasks level.

NEUROBEHAVIORAL EVALUATIONS OF BINARY AND TERTIARY MIXTURES OF CHEMICALS: LESSIONS LEARNING.

EPA Science Inventory

The classical approach to the statistical analysis of binary chemical mixtures is to construct full dose-response curves for one compound in the presence of a range of doses of the second compound (isobolographic analyses). For interaction studies using more than two chemicals, ...

Commercialising Comparison: Pearson Puts the TLC in Soft Capitalism

ERIC Educational Resources Information Center

Hogan, Anna; Sellar, Sam; Lingard, Bob

2016-01-01

This paper provides a critical policy analysis of "The Learning Curve" (TLC) (2012), an initiative developed by the multinational edu-business, Pearson, in conjunction with the Economist Intelligence Unit. "TLC" exemplifies the commercialising of comparison and the efforts of edu-businesses to strategically position themselves…

Applying active learning to supervised word sense disambiguation in MEDLINE.

PubMed

Chen, Yukun; Cao, Hongxin; Mei, Qiaozhu; Zheng, Kai; Xu, Hua

2013-01-01

This study was to assess whether active learning strategies can be integrated with supervised word sense disambiguation (WSD) methods, thus reducing the number of annotated samples, while keeping or improving the quality of disambiguation models. We developed support vector machine (SVM) classifiers to disambiguate 197 ambiguous terms and abbreviations in the MSH WSD collection. Three different uncertainty sampling-based active learning algorithms were implemented with the SVM classifiers and were compared with a passive learner (PL) based on random sampling. For each ambiguous term and each learning algorithm, a learning curve that plots the accuracy computed from the test set as a function of the number of annotated samples used in the model was generated. The area under the learning curve (ALC) was used as the primary metric for evaluation. Our experiments demonstrated that active learners (ALs) significantly outperformed the PL, showing better performance for 177 out of 197 (89.8%) WSD tasks. Further analysis showed that to achieve an average accuracy of 90%, the PL needed 38 annotated samples, while the ALs needed only 24, a 37% reduction in annotation effort. Moreover, we analyzed cases where active learning algorithms did not achieve superior performance and identified three causes: (1) poor models in the early learning stage; (2) easy WSD cases; and (3) difficult WSD cases, which provide useful insight for future improvements. This study demonstrated that integrating active learning strategies with supervised WSD methods could effectively reduce annotation cost and improve the disambiguation models.

Applying active learning to supervised word sense disambiguation in MEDLINE

PubMed Central

Chen, Yukun; Cao, Hongxin; Mei, Qiaozhu; Zheng, Kai; Xu, Hua

2013-01-01

Objectives This study was to assess whether active learning strategies can be integrated with supervised word sense disambiguation (WSD) methods, thus reducing the number of annotated samples, while keeping or improving the quality of disambiguation models. Methods We developed support vector machine (SVM) classifiers to disambiguate 197 ambiguous terms and abbreviations in the MSH WSD collection. Three different uncertainty sampling-based active learning algorithms were implemented with the SVM classifiers and were compared with a passive learner (PL) based on random sampling. For each ambiguous term and each learning algorithm, a learning curve that plots the accuracy computed from the test set as a function of the number of annotated samples used in the model was generated. The area under the learning curve (ALC) was used as the primary metric for evaluation. Results Our experiments demonstrated that active learners (ALs) significantly outperformed the PL, showing better performance for 177 out of 197 (89.8%) WSD tasks. Further analysis showed that to achieve an average accuracy of 90%, the PL needed 38 annotated samples, while the ALs needed only 24, a 37% reduction in annotation effort. Moreover, we analyzed cases where active learning algorithms did not achieve superior performance and identified three causes: (1) poor models in the early learning stage; (2) easy WSD cases; and (3) difficult WSD cases, which provide useful insight for future improvements. Conclusions This study demonstrated that integrating active learning strategies with supervised WSD methods could effectively reduce annotation cost and improve the disambiguation models. PMID:23364851

Quantitative analysis of a spinal surgeon's learning curve for scoliosis surgery.

PubMed

Ryu, K J; Suh, S W; Kim, H W; Lee, D H; Yoon, Y; Hwang, J H

2016-05-01

The aim of this study was a quantitative analysis of a surgeon's learning curve for scoliosis surgery and the relationship between the surgeon's experience and post-operative outcomes, which has not been previously well described. We have investigated the operating time as a function of the number of patients to determine a specific pattern; we analysed factors affecting the operating time and compared intra- and post-operative outcomes. We analysed 47 consecutive patients undergoing scoliosis surgery performed by a single, non-trained scoliosis surgeon. Operating time was recorded for each of the four parts of the procedures: dissection, placement of pedicle screws, reduction of the deformity and wound closure. The median operating time was 310 minutes (interquartile range 277.5 to 432.5). The pattern showed a continuous decreasing trend in operating time until the patient number reached 23 to 25, after which it stabilised with fewer patient-dependent changes. The operating time was more affected by the patient number (r =- 0.75) than the number of levels fused (r = 0.59). Blood loss (p = 0.016) and length of stay in hospital (p = 0.012) were significantly less after the operating time stabilised. Post-operative functional outcome scores and the rate of complications showed no significant differences. We describe a detailed learning curve for scoliosis surgery based on a single surgeon's practise, providing useful information for novice scoliosis surgeons and for those responsible for training in spinal surgery. Cite this article: Bone Joint J 2016;98-B:679-85. ©2016 The British Editorial Society of Bone & Joint Surgery.

Da Vinci© Skills Simulator™: is an early selection of talented console surgeons possible?

PubMed

Meier, Mark; Horton, Kevin; John, Hubert

2016-12-01

To investigate whether the learning curve of robotic surgery simulator training depends on the probands' characteristics, such as age and prior experience, we conducted a study of six distinct proband groups, using the da Vinci Skills Simulator: experienced urological robotic surgeons, surgeons with experience as da Vinci tableside assistants, urological surgeons with laparoscopic experience, urological surgeons without laparoscopic experience, and complete novices aged 25 and younger and 40 and older. The results showed that all experienced robotic surgeons reached expert level (>90 %, as defined previously in the literature) within the first three repetitions and remained on a high level of performance. All other groups performed worse. Tableside assistants, laparoscopically experienced surgeons, and younger novices showed a better performance in all exercises than surgeons without laparoscopic experience and older novices. A linear mixed-effects model analysis demonstrated no significant difference in learning curves between proband groups in all exercises except the RW1 exercise for the younger proband group. In summary, we found that performance in robotic surgery, measured by performance scores in three virtual simulator modules using the EndoWrist techniques, was dependent on age and prior experience with robotic and laparoscopic surgery. However, and most importantly, the learning curve was not significantly affected by these factors. This suggests that the da Vinci Skills Simulator™ is a useful practice tool for everyone learning or performing robotic surgery, and that early selection of talented surgeons is neither possible nor necessary.

Broiler chickens can benefit from machine learning: support vector machine analysis of observational epidemiological data

PubMed Central

Hepworth, Philip J.; Nefedov, Alexey V.; Muchnik, Ilya B.; Morgan, Kenton L.

2012-01-01

Machine-learning algorithms pervade our daily lives. In epidemiology, supervised machine learning has the potential for classification, diagnosis and risk factor identification. Here, we report the use of support vector machine learning to identify the features associated with hock burn on commercial broiler farms, using routinely collected farm management data. These data lend themselves to analysis using machine-learning techniques. Hock burn, dermatitis of the skin over the hock, is an important indicator of broiler health and welfare. Remarkably, this classifier can predict the occurrence of high hock burn prevalence with accuracy of 0.78 on unseen data, as measured by the area under the receiver operating characteristic curve. We also compare the results with those obtained by standard multi-variable logistic regression and suggest that this technique provides new insights into the data. This novel application of a machine-learning algorithm, embedded in poultry management systems could offer significant improvements in broiler health and welfare worldwide. PMID:22319115

Broiler chickens can benefit from machine learning: support vector machine analysis of observational epidemiological data.

PubMed

Hepworth, Philip J; Nefedov, Alexey V; Muchnik, Ilya B; Morgan, Kenton L

2012-08-07

Machine-learning algorithms pervade our daily lives. In epidemiology, supervised machine learning has the potential for classification, diagnosis and risk factor identification. Here, we report the use of support vector machine learning to identify the features associated with hock burn on commercial broiler farms, using routinely collected farm management data. These data lend themselves to analysis using machine-learning techniques. Hock burn, dermatitis of the skin over the hock, is an important indicator of broiler health and welfare. Remarkably, this classifier can predict the occurrence of high hock burn prevalence with accuracy of 0.78 on unseen data, as measured by the area under the receiver operating characteristic curve. We also compare the results with those obtained by standard multi-variable logistic regression and suggest that this technique provides new insights into the data. This novel application of a machine-learning algorithm, embedded in poultry management systems could offer significant improvements in broiler health and welfare worldwide.

Video-Assisted Thoracic Surgical Lobectomy for Lung Cancer: Description of a Learning Curve.

PubMed

Yao, Fei; Wang, Jian; Yao, Ju; Hang, Fangrong; Cao, Shiqi; Cao, Yongke

2017-07-01

Video-assisted thoracic surgical (VATS) lobectomy is gaining popularity in the treatment of lung cancer. The aim of this study is to investigate the learning curve of VATS lobectomy by using multidimensional methods and to compare the learning curve groups with respect to perioperative clinical outcomes. We retrospectively reviewed a prospective database to identify 67 consecutive patients who underwent VATS lobectomy for lung cancer by a single surgeon. The learning curve was analyzed by using moving average and the cumulative sum (CUSUM) method. With the moving average and CUSUM analyses for the operation time, patients were stratified into two groups, with chronological order defining early and late experiences. Perioperative clinical outcomes were compared between the two learning curve groups. According to the moving average method, the peak point for operation time occurred at the 26th case. The CUSUM method also showed the operation time peak point at the 26th case. When results were compared between early- and late-experience periods, the operation time, duration of chest drainage, and postoperative hospital stay were significantly longer in the early-experience group (cases 1 to 26). The intraoperative estimated blood loss was significantly less in the late-experience group (cases 27 to 67). CUSUM charts showed a decreasing duration of chest drainage after the 36th case and shortening postoperative hospital stay after the 37th case. Multidimensional statistical analyses suggested that the learning curve for VATS lobectomy for lung cancer required ∼26 cases. Favorable intraoperative and postoperative care parameters for VATS lobectomy were observed in the late-experience group.

The influence of government actions on innovative activities in the development of environmental technologies to control sulfur dioxide emissions from stationary sources

NASA Astrophysics Data System (ADS)

Taylor, Margaret R.

2001-12-01

A better understanding of the influence of government actions on innovation is needed to inform future policy endeavors in areas ranging from industrial competitiveness to environmentally sustainable growth. Environmental control technology is a rich area for the study of this influence, since government has stronger incentives to promote innovation in these technologies than does the private sector. This dissertation investigated the case of sulfur dioxide (SO2) control technologies for electric power plants. In studying innovation in these technologies, it was very important to understand the details of these technologies as well as their long organizational history. These technologies have been affected by government actions ranging from government-sponsored research and technology transfer mechanisms to national regulatory events. The dissertation integrated insights from several complementary and repeatable innovation evaluation methods; this approach supported a fuller understanding of innovation while it structured the research results for potential future comparative analysis. Innovative activities were investigated through: patent activity analysis; technical content analysis and researcher co-authorship network analysis in a conference held for over twenty years; learning curve analysis for eighty-eight U.S. power plants; and a dozen expert interviews from a variety of innovative actors. Innovative outcomes were investigated through: analysis of observed improvements in newly installed technologies over time; evaluation of historic cost studies on standardized systems; and expert interviews. Several policy-relevant findings resulted from this dissertation. (1) The existence of national government regulation stimulated inventive activity more than government research support alone. (2) The existence and the anticipation of government regulation appeared to spur inventive activity, while regulatory stringency appeared to drive inventive activity and the communication process underlying knowledge transfer and diffusion. (3) The regulatory-forced adoption of SO2 control technologies led to a learning curve effect in which operating experience with the equipment resulted in significant cost improvements. This learning curve effect is comparable with findings in many other industries and is likely to be useful in predictions of the costs of future environmental technologies. (4) Performance improvements and cost reductions occurred in a quantifiable fashion as the technology became more widely adopted.

Learning curve for robotic-assisted surgery for rectal cancer: use of the cumulative sum method.

PubMed

Yamaguchi, Tomohiro; Kinugasa, Yusuke; Shiomi, Akio; Sato, Sumito; Yamakawa, Yushi; Kagawa, Hiroyasu; Tomioka, Hiroyuki; Mori, Keita

2015-07-01

Few data are available to assess the learning curve for robotic-assisted surgery for rectal cancer. The aim of the present study was to evaluate the learning curve for robotic-assisted surgery for rectal cancer by a surgeon at a single institute. From December 2011 to August 2013, a total of 80 consecutive patients who underwent robotic-assisted surgery for rectal cancer performed by the same surgeon were included in this study. The learning curve was analyzed using the cumulative sum method. This method was used for all 80 cases, taking into account operative time. Operative procedures included anterior resections in 6 patients, low anterior resections in 46 patients, intersphincteric resections in 22 patients, and abdominoperineal resections in 6 patients. Lateral lymph node dissection was performed in 28 patients. Median operative time was 280 min (range 135-683 min), and median blood loss was 17 mL (range 0-690 mL). No postoperative complications of Clavien-Dindo classification Grade III or IV were encountered. We arranged operative times and calculated cumulative sum values, allowing differentiation of three phases: phase I, Cases 1-25; phase II, Cases 26-50; and phase III, Cases 51-80. Our data suggested three phases of the learning curve in robotic-assisted surgery for rectal cancer. The first 25 cases formed the learning phase.

Facilitating students' application of the integral and the area under the curve concepts in physics problems

NASA Astrophysics Data System (ADS)

Nguyen, Dong-Hai

This research project investigates the difficulties students encounter when solving physics problems involving the integral and the area under the curve concepts and the strategies to facilitate students learning to solve those types of problems. The research contexts of this project are calculus-based physics courses covering mechanics and electromagnetism. In phase I of the project, individual teaching/learning interviews were conducted with 20 students in mechanics and 15 students from the same cohort in electromagnetism. The students were asked to solve problems on several topics of mechanics and electromagnetism. These problems involved calculating physical quantities (e.g. velocity, acceleration, work, electric field, electric resistance, electric current) by integrating or finding the area under the curve of functions of related quantities (e.g. position, velocity, force, charge density, resistivity, current density). Verbal hints were provided when students made an error or were unable to proceed. A total number of 140 one-hour interviews were conducted in this phase, which provided insights into students' difficulties when solving the problems involving the integral and the area under the curve concepts and the hints to help students overcome those difficulties. In phase II of the project, tutorials were created to facilitate students' learning to solve physics problems involving the integral and the area under the curve concepts. Each tutorial consisted of a set of exercises and a protocol that incorporated the helpful hints to target the difficulties that students expressed in phase I of the project. Focus group learning interviews were conducted to test the effectiveness of the tutorials in comparison with standard learning materials (i.e. textbook problems and solutions). Overall results indicated that students learning with our tutorials outperformed students learning with standard materials in applying the integral and the area under the curve concepts to physics problems. The results of this project provide broader and deeper insights into students' problem solving with the integral and the area under the curve concepts and suggest strategies to facilitate students' learning to apply these concepts to physics problems. This study also has significant implications for further research, curriculum development and instruction.

Learning neuroendoscopy with an exoscope system (video telescopic operating monitor): Early clinical results.

PubMed

Parihar, Vijay; Yadav, Y R; Kher, Yatin; Ratre, Shailendra; Sethi, Ashish; Sharma, Dhananjaya

2016-01-01

Steep learning curve is found initially in pure endoscopic procedures. Video telescopic operating monitor (VITOM) is an advance in rigid-lens telescope systems provides an alternative method for learning basics of neuroendoscopy with the help of the familiar principle of microneurosurgery. The aim was to evaluate the clinical utility of VITOM as a learning tool for neuroendoscopy. Video telescopic operating monitor was used 39 cranial and spinal procedures and its utility as a tool for minimally invasive neurosurgery and neuroendoscopy for initial learning curve was studied. Video telescopic operating monitor was used in 25 cranial and 14 spinal procedures. Image quality is comparable to endoscope and microscope. Surgeons comfort improved with VITOM. Frequent repositioning of scope holder and lack of stereopsis is initial limiting factor was compensated for with repeated procedures. Video telescopic operating monitor is found useful to reduce initial learning curve of neuroendoscopy.

Evaluation of the learning curve for thulium laser enucleation of the prostate with the aid of a simulator tool but without tutoring: comparison of two surgeons with different levels of endoscopic experience.

PubMed

Saredi, Giovanni; Pirola, Giacomo Maria; Pacchetti, Andrea; Lovisolo, Jon Alexander; Borroni, Giacomo; Sembenini, Federico; Marconi, Alberto Mario

2015-06-09

The aim of this study was to determine the learning curve for thulium laser enucleation of the prostate (ThuLEP) for two surgeons with different levels of urological endoscopic experience. From June 2012 to August 2013, ThuLEP was performed on 100 patients in our institution. We present the results of a prospective evaluation during which we analyzed data related to the learning curves for two surgeons of different levels of experience. The prostatic adenoma volumes ranged from 30 to 130 mL (average 61.2 mL). Surgeons A and B performed 48 and 52 operations, respectively. Six months after surgery, all patients were evaluated with the International Prostate Symptom Score questionnaire, uroflowmetry, and prostate-specific antigen test. Introduced in 2010, ThuLEP consists of blunt enucleation of the prostatic apex and lobes using the sheath of the resectoscope. This maneuver allows clearer visualization of the enucleation plane and precise identification of the prostatic capsule. These conditions permit total resection of the prostatic adenoma and coagulation of small penetrating vessels, thereby reducing the laser emission time. Most of the complications in this series were encountered during morcellation, which in some cases was performed under poor vision because of venous bleeding due to surgical perforation of the capsule during enucleation. Based on this analysis, we concluded that it is feasible for laser-naive urologists with endoscopic experience to learn to perform ThuLEP without tutoring. Those statements still require further validation in larger multicentric study cohort by several surgeon. The main novelty during the learning process was the use of a simulator that faithfully reproduced all of the surgical steps in prostates of various shapes and volumes.

Microinvasive Glaucoma Stent (MIGS) Surgery With Concomitant Phakoemulsification Cataract Extraction: Outcomes and the Learning Curve.

PubMed

Al-Mugheiry, Toby S; Cate, Heidi; Clark, Allan; Broadway, David C

2017-07-01

To evaluate learning effects with respect to outcomes of a microinvasive glaucoma stent (MIGS) inserted during cataract surgery in glaucoma patients. Single surgeon, observational cohort study of 25 consecutive Ivantis Hydrus microstent insertions, with a minimum follow-up of 12 months. A learning curve analysis was performed by assessing hypotensive effect, adverse effects, and surgical procedure duration, with respect to consecutive case number. Success was defined with respect to various intraocular pressure (IOP) targets (21, 18, 15 mm Hg) and reduction in required antiglaucoma medications. Complete success was defined as achieving target IOP without antiglaucoma therapy. No clinically significant adverse events or learning effects were identified, although surgical time reduced with consecutive case number. Mean follow-up was 16.8 months. At final follow-up the mean IOP for all eyes was reduced from 18.1 (±3.6) mm Hg [and a simulated untreated value of 25.9 (±5.2) mm Hg] to 15.3 (±2.2) mm Hg (P=0.007; <0.0001) and the mean number of topical antiglaucoma medications was reduced from 1.96 (±0.96) to 0.04 (±0.20) (P<0.0001). Complete success (IOP<21 mm Hg, no medications) was 96% at final follow-up. Complete success (IOP<18 mm Hg, no medications) was 80% at final follow-up, but only 32% with a target IOP of <15 mm Hg (no medications). No significant learning curve effects were observed for a trained surgeon with respect to MIGS microstent insertion performed at the time of cataract surgery. Adjunctive MIGS surgery was successful in lowering IOP to <18 mm Hg and reducing/abolishing the requirement for antiglaucoma medication in eyes with open-angle glaucoma, but less successful at achieving low IOP levels (<15 mm Hg).

Robotic radical prostatectomy learning curve of a fellowship-trained laparoscopic surgeon.

PubMed

Zorn, Kevin C; Orvieto, Marcelo A; Gong, Edward M; Mikhail, Albert A; Gofrit, Ofer N; Zagaja, Gregory P; Shalhav, Arieh L

2007-04-01

Several experienced practitioners of open surgery with limited or no laparoscopic background have adopted robot-assisted laparoscopic radical prostatectomy (RLRP) as an alternative to open radical prostatectomy (RRP), demonstrating outcomes comparable to those in large RRP and laparoscopic prostatectomy series. Thus, the significance of prior laparoscopic skills seems unclear. The learning curve, with respect to operative time and complications, in the hands of a devoted laparoscopic surgeon has not been critically assessed. We evaluated the learning curve of a highly experienced laparoscopic surgeon in achieving expertise with RLRP. We prospectively evaluated 150 consecutive patients undergoing RLRP by a single surgeon between March 2003 and September 2005. The first 25 cases were performed with the assistance of a surgeon experienced in open RRP. Data were compared for the first, second, and third groups of 50 cases. Demographic data were similar for the three groups. Urinary and sexual function data were evaluated subjectively and objectively using the RAND-36v2 Survey and the UCLA PCI preoperatively and at 3, 6, and 12 months postoperatively. The mean operative time, blood loss, and conversion rate decreased significantly with increasing experience. All open conversions occurred during the first 25 cases. Intraoperative and postoperative complication rates were similar among groups. Although the differences were not significant, urinary and sexual function recovery improved with experience. The RLRP learning curve for a fellowship-trained laparoscopic surgeon seems to be similar to that of laparoscopically naive yet experienced practitioners of open RRP. The RLRP is safe and reproducible and even during the learning curve can produce results similar to those reported in large RRP series. The importance of assistance by an experienced open RRP surgeon during the learning curve cannot be overemphasized.

The learning curve in robotic distal pancreatectomy.

PubMed

Napoli, Niccolò; Kauffmann, Emanuele F; Perrone, Vittorio Grazio; Miccoli, Mario; Brozzetti, Stefania; Boggi, Ugo

2015-09-01

No data are available on the learning curve in robotic distal pancreatectomy (RADP). The learning curve in RADP was assessed in 55 consecutive patients using the cumulative sum method, based on operative time. Data were extracted from a prospectively maintained database and analyzed retrospectively considering all events occurring within 90 days of surgery. No operation was converted to laparoscopic or open surgery and no patient died. Post-operative complications occurred in 34 patients (61.8%), being of Clavien-Dindo grade I-II in 32 patients (58.1%), including pancreatic fistula in 29 patients (52.7%). No grade C pancreatic fistula occurred. Four patients received blood transfusions (7.2%), three were readmitted (5.4%) and one required repeat surgery (1.8%). Based on the reduction of operative times (421.1 ± 20.5 vs 248.9 ± 9.3 min; p < 0.0001), completion of the learning curve was achieved after ten operations. Operative time of the first 10 operations was associated with a positive slope (0.47 + 1.78* case number; R (2) 0.97; p < 0.0001*), while that of the following 45 procedures showed a negative slope (23.52 - 0.39* case number; R (2) 0.97; p < 0.0001*). After completion of the learning curve, more patients had a malignant histology (0 vs 35.6%; p = 0.002), accounting for both higher lymph node yields (11.1 ± 12.2 vs 20.9 ± 18.5) (p = 0.04) and lower rate of spleen preservation (90 vs 55.6%) (p = 0.04). RADP was safely feasible in selected patients and the learning curve was completed after ten operations. Improvement in clinical outcome was not demonstrated, probably because of the limited occurrence of outcome comparators.

The Development of English and Mathematics Self-Efficacy: A Latent Growth Curve Analysis

ERIC Educational Resources Information Center

Phan, Huy P.

2012-01-01

Empirical research has provided evidence supporting the validation and prediction of 4 major sources of self-efficacy: enactive performance accomplishments, vicarious experiences, verbal persuasion, and emotional states. Other research studies have also attested to the importance and potency of self-efficacy in academic learning and achievement.…

Analysis of Computer Algebra System Tutorials Using Cognitive Load Theory

ERIC Educational Resources Information Center

May, Patricia

2004-01-01

Most research in the area of Computer Algebra Systems (CAS) has been designed to compare the effectiveness of instructional technology to traditional lecture-based formats. While results are promising, research also indicates evidence of the steep learning curve imposed by the technology. Yet no studies have been conducted to investigate this…

First Year Specialist Anaesthesia Training in Ireland: A Logbook Analysis

ERIC Educational Resources Information Center

O'Shaughnessy, S. M.; Skerritt, C. J.; Fitzgerald, C. W.; Irwin, R.; Walsh, F.

2017-01-01

Objective: Acquisition of a new range of skills occurs during first year anaesthesia training. The first twelve months of specialist anaesthesia training represent the steepest part of the learning curve, and thus large differences in performance should be apparent between the first and last quarters of this period. At present, no published…

Estimating the cost of production stoppage

NASA Technical Reports Server (NTRS)

Delionback, L. M.

1979-01-01

Estimation model considers learning curve quantities, and time of break to forecast losses due to break in production schedule. Major parameters capable of predicting costs are number of units made prior to production sequence, length of production break, and slope of learning curve produced prior to break.

Experience and learning curve of retroperitoneal laparoscopic ureterolithotomy for upper ureteral calculi.

PubMed

Fan, Tianyong; Xian, Peng; Yang, Lu; Liu, Yong; Wei, Qiang; Li, Hong

2009-11-01

To summarize our experience and evaluate the learning curve of retroperitoneal laparoscopic ureterolithotomy of the upper ureter. Between May 2004 and May 2007, 40 patients underwent retroperitoneal laparoscopic ureterolithotomy of the upper ureter. We divided the first and last 20 patients into group I and group II. There was no statistical difference in stone size between groups. Operative time and complications were measured as a basis for the assessment of the learning curve. In group I, the complication rate was 15% (3/20), including two patients whose procedure was converted to open surgery because of intraoperative bleeding, and one patient who experienced urine leakage because of a displaced Double-J ureteral stent. In group II, no postoperative complications occurred, while the mean operative time was significantly shorter compared with the earlier operations (65 vs 120 min). Retroperitoneal laparoscopic ureterolithotomy is safe and effective for large or impacted stones of the upper ureter. It is associated with a short learning curve in the setting of an active laparoscopic practice for selected patients.

[Pancreatoduodenectomy: learning curve within single multi-field center].

PubMed

Kaprin, A D; Kostin, A A; Nikiforov, P V; Egorov, V I; Grishin, N A; Lozhkin, M V; Petrov, L O; Bykasov, S A; Sidorov, D V

2018-01-01

To analyze learning curve by using of immediate results of pancreatoduodenectomy at multi-field oncology institute. For the period 2010-2016 at Abdominal Oncology Department of Herzen Moscow Oncology Research Institute 120 pancreatoduodenal resections were consistently performed. All patients were divided into two groups: the first 60 procedures (group A) and subsequent 60 operations (group B). Herewith, first 60 operations were performed within the first 4.5 years of study period, the next 60 operations - within remaining 2.5 years. Learning curves showed significantly variable intraoperative blood loss (1100 ml and 725 ml), surgery time (589 min and 513 min) and postoperative hospital-stay (15 days and 13 days) in group A followed by gradual improvement of these values in group B. Incidence of negative resection margin (R0) was also significantly improved in the last 60 operations (70 and 92%, respectively). Despite pancreatoduodenectomy is one of the most difficult surgical interventions in abdominal surgery learning curve will differ from one surgeon to another.

Soft tissue balancing in total knee arthroplasty using sensor-guided assessment: is there a learning curve?

PubMed

Gharaibeh, Monther A; Chen, Darren B; MacDessi, Samuel J

2018-05-01

Sensor-guided assessment for soft tissue balance in total knee arthroplasty (TKA) has been reported to improve patient satisfaction and self-reported outcome scores. As more surgeons adopt this technology in TKA, we performed this study to identify if there is a learning curve with its use. Analysis of a total of 90 consecutive cases was performed in this study. Initial and final intercompartmental pressure differences were recorded before and after knee ligament balancing. The first 45 patients (group 1) were compared to the last 45 patients (group 2) in terms of operative time and the final state of knee balance. A balanced knee was defined as pressure difference between medial and lateral compartments of ≤15 pounds. Group 1 had 10 unbalanced knees in the final pressure difference assessment, while all cases in group 2 were balanced (P < 0.001). There was no statistically significant difference in mean operative time between the two groups. A scatter plot of intercompartmental pressure difference identified that after 30 cases, the capacity to achieve knee ligament balance improved. This study suggests that there is a learning curve with the use of sensor-guided assessment in TKA in achieving knee balance; however, the differences noted between initial and final groups were small and may not be of clinical significance. © 2018 Royal Australasian College of Surgeons.

feets: feATURE eXTRACTOR for tIME sERIES

NASA Astrophysics Data System (ADS)

Cabral, Juan; Sanchez, Bruno; Ramos, Felipe; Gurovich, Sebastián; Granitto, Pablo; VanderPlas, Jake

2018-06-01

feets characterizes and analyzes light-curves from astronomical photometric databases for modelling, classification, data cleaning, outlier detection and data analysis. It uses machine learning algorithms to determine the numerical descriptors that characterize and distinguish the different variability classes of light-curves; these range from basic statistical measures such as the mean or standard deviation to complex time-series characteristics such as the autocorrelation function. The library is not restricted to the astronomical field and could also be applied to any kind of time series. This project is a derivative work of FATS (ascl:1711.017).

A Data-driven Study of RR Lyrae Near-IR Light Curves: Principal Component Analysis, Robust Fits, and Metallicity Estimates

NASA Astrophysics Data System (ADS)

Hajdu, Gergely; Dékány, István; Catelan, Márcio; Grebel, Eva K.; Jurcsik, Johanna

2018-04-01

RR Lyrae variables are widely used tracers of Galactic halo structure and kinematics, but they can also serve to constrain the distribution of the old stellar population in the Galactic bulge. With the aim of improving their near-infrared photometric characterization, we investigate their near-infrared light curves, as well as the empirical relationships between their light curve and metallicities using machine learning methods. We introduce a new, robust method for the estimation of the light-curve shapes, hence the average magnitudes of RR Lyrae variables in the K S band, by utilizing the first few principal components (PCs) as basis vectors, obtained from the PC analysis of a training set of light curves. Furthermore, we use the amplitudes of these PCs to predict the light-curve shape of each star in the J-band, allowing us to precisely determine their average magnitudes (hence colors), even in cases where only one J measurement is available. Finally, we demonstrate that the K S-band light-curve parameters of RR Lyrae variables, together with the period, allow the estimation of the metallicity of individual stars with an accuracy of ∼0.2–0.25 dex, providing valuable chemical information about old stellar populations bearing RR Lyrae variables. The methods presented here can be straightforwardly adopted for other classes of variable stars, bands, or for the estimation of other physical quantities.

Robotic thyroidectomy learning curve for beginning surgeons with little or no experience of endoscopic surgery.

PubMed

Park, Jae Hyun; Lee, Jandee; Hakim, Nor Azham; Kim, Ha Yan; Kang, Sang-Wook; Jeong, Jong Ju; Nam, Kee-Hyun; Bae, Keum-Seok; Kang, Seong Joon; Chung, Woong Youn

2015-12-01

This study assessed the results of robotic thyroidectomy by fellowship-trained surgeons in their initial independent practice, and whether standard fellowship training for robotic surgery shortens the learning curve. This prospective cohort study evaluated outcomes in 125 patients who underwent robotic thyroidectomy using gasless transaxillary single-incision technique by 2 recently graduated fellowship-trained surgeons. Learning curves were analyzed by operation time, with proficiency defined as the point at which the slope of the time curve became less steep. Of the 125 patients, 113 underwent robotic less-than-total thyroidectomy, 9 underwent robotic total thyroidectomy and 3 underwent robotic total thyroidectomy with modified radical neck dissection. Mean total times for these 3 operations were 100.8 ± 20.6 minutes, 134.2 ± 38.7 minutes, and 284.7 ± 60.4 minutes, respectively. For both surgeons, the operation times gradually decreased, reaching a plateau after 20 robotic less-than-total thyroidectomies. The surgical learning curve for robotic thyroidectomy performed by recently graduated fellowship-trained surgeons with little or no experience in endoscopic surgery showed excellent results compared with those in a large series of more experienced surgeons. © 2014 Wiley Periodicals, Inc.

Automated Blazar Light Curves Using Machine Learning

DOE Office of Scientific and Technical Information (OSTI.GOV)

Johnson, Spencer James

2017-07-27

This presentation describes a problem and methodology pertaining to automated blazar light curves. Namely, optical variability patterns for blazars require the construction of light curves and in order to generate the light curves, data must be filtered before processing to ensure quality.

2D Affine and Projective Shape Analysis.

PubMed

Bryner, Darshan; Klassen, Eric; Huiling Le; Srivastava, Anuj

2014-05-01

Current techniques for shape analysis tend to seek invariance to similarity transformations (rotation, translation, and scale), but certain imaging situations require invariance to larger groups, such as affine or projective groups. Here we present a general Riemannian framework for shape analysis of planar objects where metrics and related quantities are invariant to affine and projective groups. Highlighting two possibilities for representing object boundaries-ordered points (or landmarks) and parameterized curves-we study different combinations of these representations (points and curves) and transformations (affine and projective). Specifically, we provide solutions to three out of four situations and develop algorithms for computing geodesics and intrinsic sample statistics, leading up to Gaussian-type statistical models, and classifying test shapes using such models learned from training data. In the case of parameterized curves, we also achieve the desired goal of invariance to re-parameterizations. The geodesics are constructed by particularizing the path-straightening algorithm to geometries of current manifolds and are used, in turn, to compute shape statistics and Gaussian-type shape models. We demonstrate these ideas using a number of examples from shape and activity recognition.

Renal mass anatomic characteristics and perioperative outcomes of laparoscopic partial nephrectomy: a critical analysis.

PubMed

Tsivian, Matvey; Ulusoy, Said; Abern, Michael; Wandel, Ayelet; Sidi, A Ami; Tsivian, Alexander

2012-10-01

Anatomic parameters determining renal mass complexity have been used in a number of proposed scoring systems despite lack of a critical analysis of their independent contributions. We sought to assess the independent contribution of anatomic parameters on perioperative outcomes of laparoscopic partial nephrectomy (LPN). Preoperative imaging studies were reviewed for 147 consecutive patients undergoing LPN for a single renal mass. Renal mass anatomy was recorded: Size, growth pattern (endo-/meso-/exophytic), centrality (central/hilar/peripheral), anterior/posterior, lateral/medial, polar location. Multivariable models were used to determine associations of anatomic parameters with warm ischemia time (WIT), operative time (OT), estimated blood loss (EBL), intra- and postoperative complications, as well as renal function. All models were adjusted for the learning curve and relevant confounders. Median (range) tumor size was 3.3 cm (1.5-11 cm); 52% were central and 14% hilar. While 44% were exophytic, 23% and 33% were mesophytic and endophytic, respectively. Anatomic parameters did not uniformly predict perioperative outcomes. WIT was associated with tumor size (P=0.068), centrality (central, P=0.016; hilar, P=0.073), and endophytic growth pattern (P=0.017). OT was only associated with tumor size (P<0.001). No anatomic parameter predicted EBL. Tumor centrality increased the odds of overall and intraoperative complications, without reaching statistical significance. Postoperative renal function was not associated with any of the anatomic parameters considered after adjustment for baseline function and WIT. Learning curve, considered as a confounder, was independently associated with reduced WIT and OT as well as reduced odds of intraoperative complications. This study provides a detailed analysis of the independent impact of renal mass anatomic parameters on perioperative outcomes. Our findings suggest diverse independent contributions of the anatomic parameters to the different measures of outcomes (WIT, OT, EBL, complications, and renal function) emphasizing the importance of the learning curve.

Assessment of quality outcomes for robotic pancreaticoduodenectomy: identification of the learning curve.

PubMed

Boone, Brian A; Zenati, Mazen; Hogg, Melissa E; Steve, Jennifer; Moser, Arthur James; Bartlett, David L; Zeh, Herbert J; Zureikat, Amer H

2015-05-01

Quality assessment is an important instrument to ensure optimal surgical outcomes, particularly during the adoption of new surgical technology. The use of the robotic platform for complex pancreatic resections, such as the pancreaticoduodenectomy, requires close monitoring of outcomes during its implementation phase to ensure patient safety is maintained and the learning curve identified. To report the results of a quality analysis and learning curve during the implementation of robotic pancreaticoduodenectomy (RPD). A retrospective review of a prospectively maintained database of 200 consecutive patients who underwent RPD in a large academic center from October 3, 2008, through March 1, 2014, was evaluated for important metrics of quality. Patients were analyzed in groups of 20 to minimize demographic differences and optimize the ability to detect statistically meaningful changes in performance. Robotic pancreaticoduodenectomy. Optimization of perioperative outcome parameters. No statistical differences in mortality rates or major morbidity were noted during the study. Statistical improvements in estimated blood loss and conversions to open surgery occurred after 20 cases (600 mL vs 250 mL [P = .002] and 35.0% vs 3.3% [P < .001], respectively), incidence of pancreatic fistula after 40 cases (27.5% vs 14.4%; P = .04), and operative time after 80 cases (581 minutes vs 417 minutes [P < .001]). Complication rates, lengths of stay, and readmission rates showed continuous improvement that did not reach statistical significance. Outcomes for the last 120 cases (representing optimized metrics beyond the learning curve) included a mean operative time of 417 minutes, median estimated blood loss of 250 mL, a conversion rate of 3.3%, 90-day mortality of 3.3%, a clinically significant (grade B/C) pancreatic fistula rate of 6.9%, and a median length of stay of 9 days. Continuous assessment of quality metrics allows for safe implementation of RPD. We identified several inflexion points corresponding to optimization of performance metrics for RPD that can be used as benchmarks for surgeons who are adopting this technology.

A comparison of hysteroscopic mechanical tissue removal with bipolar electrical resection for the management of endometrial polyps in an ambulatory care setting: preliminary results.

PubMed

Pampalona, Jennifer Rovira; Bastos, Maria Degollada; Moreno, Gemma Mancebo; Pust, Andrea Buron; Montesdeoca, Gemma Escribano; Guerra Garcia, Angel; Pruñonosa, Juan Carles Mateu; Collado, Ramon Carreras; Torras, Pere Bresco

2015-01-01

To assess and compare efficacy, pain, and the learning curve associated with diagnostic therapeutic hysteroscopy using mechanical tissue removal versus bipolar electrical resection in the management of endometrial polyps in an ambulatory care setting. A randomized controlled clinical trial (Canadian Task Force classification I). Hospital de Igulada, Barcelona, Spain. A total of 133 patients diagnosed with endometrial polyp(s) were included and randomly assigned to 1 of the 2 hysteroscopic methods. Criteria assessed were total hysteroscopy time, full polypectomy procedure time, pain experienced by patients, and learning curve of staff in training. The average time to perform total hysteroscopy using the mechanical tissue removal system (TRUCLEAR 5.0 System; Smith & Nephew Inc., Andover, MD) was 6 minutes 49 seconds versus 11 minutes 37 seconds required for the bipolar electrosurgery system (GYNECARE VERSAPOINT; Ethicon Inc, Somerville, NJ) (p < .01). Results for complete polypectomy time favored the TRUCLEAR System at 3 minutes 7 seconds over the VERSAPOINT System at 8 minutes 25 seconds (p < .01). If a successful procedure is predicated on access to cavity, visualization, and complete resection and excision of endometrial polyp, the mechanical TRUCLEAR Tissue Removal System shows a higher success rate than the VERSAPOINT Bipolar Electrosurgery System at 92% and 77%, respectively. Analysis of pain using the visual analog scale revealed no significant differences between the 2 techniques (p > .05). A study of the residents' learning curve showed a higher level of autonomy with hysteroscopy using the TRUCLEAR Tissue Removal System with which residents showed a higher level of confidence compared with hysteroscopy with the VERSAPOINT Bipolar Electrosurgery System. In hysteroscopic polypectomy, the mechanical tissue removal system was significantly faster, achieved a greater success rate for complete polypectomy, and required a shorter learning curve from staff being trained in the management of endometrial polyps when compared with bipolar electrical resection. Copyright © 2015 AAGL. Published by Elsevier Inc. All rights reserved.

Microvascular Anastomosis: Proposition of a Learning Curve.

PubMed

Mokhtari, Pooneh; Tayebi Meybodi, Ali; Benet, Arnau; Lawton, Michael T

2018-04-14

Learning to perform a microvascular anastomosis is one of the most difficult tasks in cerebrovascular surgery. Previous studies offer little regarding the optimal protocols to maximize learning efficiency. This failure stems mainly from lack of knowledge about the learning curve of this task. To delineate this learning curve and provide information about its various features including acquisition, improvement, consistency, stability, and recall. Five neurosurgeons with an average surgical experience history of 5 yr and without any experience in bypass surgery performed microscopic anastomosis on progressively smaller-caliber silastic tubes (Biomet, Palm Beach Gardens, Florida) during 24 consecutive sessions. After a 1-, 2-, and 8-wk retention interval, they performed recall test on 0.7-mm silastic tubes. The anastomoses were rated based on anastomosis patency and presence of any leaks. Improvement rate was faster during initial sessions compared to the final practice sessions. Performance decline was observed in the first session of working on a smaller-caliber tube. However, this rapidly improved during the following sessions of practice. Temporary plateaus were seen in certain segments of the curve. The retention interval between the acquisition and recall phase did not cause a regression to the prepractice performance level. Learning the fine motor task of microvascular anastomosis adapts to the basic rules of learning such as the "power law of practice." Our results also support the improvement of performance during consecutive sessions of practice. The objective evidence provided may help in developing optimized learning protocols for microvascular anastomosis.

Training, Simulation, the Learning Curve, and How to Reduce Complications in Urology.

PubMed

Brunckhorst, Oliver; Volpe, Alessandro; van der Poel, Henk; Mottrie, Alexander; Ahmed, Kamran

2016-04-01

Urology is at the forefront of minimally invasive surgery to a great extent. These procedures produce additional learning challenges and possess a steep initial learning curve. Training and assessment methods in surgical specialties such as urology are known to lack clear structure and often rely on differing operative flow experienced by individuals and institutions. This article aims to assess current urology training modalities, to identify the role of simulation within urology, to define and identify the learning curves for various urologic procedures, and to discuss ways to decrease complications in the context of training. A narrative review of the literature was conducted through December 2015 using the PubMed/Medline, Embase, and Cochrane Library databases. Evidence of the validity of training methods in urology includes observation of a procedure, mentorship and fellowship, e-learning, and simulation-based training. Learning curves for various urologic procedures have been recommended based on the available literature. The importance of structured training pathways is highlighted, with integration of modular training to ensure patient safety. Valid training pathways are available in urology. The aim in urology training should be to combine all of the available evidence to produce procedure-specific curricula that utilise the vast array of training methods available to ensure that we continue to improve patient outcomes and reduce complications. The current evidence for different training methods available in urology, including simulation-based training, was reviewed, and the learning curves for various urologic procedures were critically analysed. Based on the evidence, future pathways for urology curricula have been suggested to ensure that patient safety is improved. Copyright © 2016 European Association of Urology. Published by Elsevier B.V. All rights reserved.

The 1974 AVCR Young Scholar Paper: An Open-System Model of Learning

ERIC Educational Resources Information Center

Winn, William

1975-01-01

Rejecting the cybernetic model of the learner, the author offers an open-system model based on von Bertalanffy's equation for growth of the living organism. The model produces four learning curves, not just the logarithmic curve produced by the successive approximations of the cybernetic model. (Editor)

Differentiation of Enhancing Glioma and Primary Central Nervous System Lymphoma by Texture-Based Machine Learning.

PubMed

Alcaide-Leon, P; Dufort, P; Geraldo, A F; Alshafai, L; Maralani, P J; Spears, J; Bharatha, A

2017-06-01

Accurate preoperative differentiation of primary central nervous system lymphoma and enhancing glioma is essential to avoid unnecessary neurosurgical resection in patients with primary central nervous system lymphoma. The purpose of the study was to evaluate the diagnostic performance of a machine-learning algorithm by using texture analysis of contrast-enhanced T1-weighted images for differentiation of primary central nervous system lymphoma and enhancing glioma. Seventy-one adult patients with enhancing gliomas and 35 adult patients with primary central nervous system lymphomas were included. The tumors were manually contoured on contrast-enhanced T1WI, and the resulting volumes of interest were mined for textural features and subjected to a support vector machine-based machine-learning protocol. Three readers classified the tumors independently on contrast-enhanced T1WI. Areas under the receiver operating characteristic curves were estimated for each reader and for the support vector machine classifier. A noninferiority test for diagnostic accuracy based on paired areas under the receiver operating characteristic curve was performed with a noninferiority margin of 0.15. The mean areas under the receiver operating characteristic curve were 0.877 (95% CI, 0.798-0.955) for the support vector machine classifier; 0.878 (95% CI, 0.807-0.949) for reader 1; 0.899 (95% CI, 0.833-0.966) for reader 2; and 0.845 (95% CI, 0.757-0.933) for reader 3. The mean area under the receiver operating characteristic curve of the support vector machine classifier was significantly noninferior to the mean area under the curve of reader 1 ( P = .021), reader 2 ( P = .035), and reader 3 ( P = .007). Support vector machine classification based on textural features of contrast-enhanced T1WI is noninferior to expert human evaluation in the differentiation of primary central nervous system lymphoma and enhancing glioma. © 2017 by American Journal of Neuroradiology.

Learning Style Scales: a valid and reliable questionnaire.

PubMed

Abdollahimohammad, Abdolghani; Ja'afar, Rogayah

2014-01-01

Learning-style instruments assist students in developing their own learning strategies and outcomes, in eliminating learning barriers, and in acknowledging peer diversity. Only a few psychometrically validated learning-style instruments are available. This study aimed to develop a valid and reliable learning-style instrument for nursing students. A cross-sectional survey study was conducted in two nursing schools in two countries. A purposive sample of 156 undergraduate nursing students participated in the study. Face and content validity was obtained from an expert panel. The LSS construct was established using principal axis factoring (PAF) with oblimin rotation, a scree plot test, and parallel analysis (PA). The reliability of LSS was tested using Cronbach's α, corrected item-total correlation, and test-retest. Factor analysis revealed five components, confirmed by PA and a relatively clear curve on the scree plot. Component strength and interpretability were also confirmed. The factors were labeled as perceptive, solitary, analytic, competitive, and imaginative learning styles. Cronbach's α was >0.70 for all subscales in both study populations. The corrected item-total correlations were >0.30 for the items in each component. The LSS is a valid and reliable inventory for evaluating learning style preferences in nursing students in various multicultural environments.

An Analysis of the Navy’s Fiscal Year 2017 Shipbuilding Plan

DTIC Science & Technology

2017-02-01

Navy would build a larger fleet of about 350 ships (see Table 5). Those three alternatives were chosen for illustrative purposes because variations ...3.2 billion. 2. For more on procedures for estimating and applying learning curves, see Matthew S. Goldberg and Anduin E. Touw, Statistical Methods...guidance from Matthew Goldberg (formerly of CBO) and David Mosher. Raymond Hall of CBO’s Budget Analysis Division produced the cost estimates with

Discovery of 36 eclipsing EL CVn binaries found by the Palomar Transient Factory

NASA Astrophysics Data System (ADS)

van Roestel, J.; Kupfer, T.; Ruiz-Carmona, R.; Groot, P. J.; Prince, T. A.; Burdge, K.; Laher, R.; Shupe, D. L.; Bellm, E.

2018-04-01

We report on the discovery and analysis of 36 new eclipsing EL CVn-type binaries, consisting of a core helium-composition pre-white dwarf (pre-He-WD) and an early-type main-sequence companion. This more than doubles the known population of these systems. We have used supervised machine learning methods to search 0.8 million light curves from the Palomar Transient Factory (PTF), combined with Sloan Digital Sky Survey (SDSS), Panoramic Survey Telescope and Rapid Response System (Pan-STARRS) and Two-Micron All-Sky Survey (2MASS) colours. The new systems range in orbital periods from 0.46 to 3.8 d and in apparent brightness from ˜14 to 16 mag in the PTF R or g΄ filters. For 12 of the systems, we obtained radial velocity curves with the Intermediate Dispersion Spectrograph at the Isaac Newton Telescope. We modelled the light curves, radial velocity curves and spectral energy distributions to determine the system parameters. The radii (0.3-0.7 R⊙) and effective temperatures (8000-17 000 K) of the pre-He-WDs are consistent with stellar evolution models, but the masses (0.12-0.28 M⊙) show more variance than models have predicted. This study shows that using machine learning techniques on large synoptic survey data is a powerful way to discover substantial samples of binary systems in short-lived evolutionary stages.

Gaia eclipsing binary and multiple systems. Supervised classification and self-organizing maps

NASA Astrophysics Data System (ADS)

Süveges, M.; Barblan, F.; Lecoeur-Taïbi, I.; Prša, A.; Holl, B.; Eyer, L.; Kochoska, A.; Mowlavi, N.; Rimoldini, L.

2017-07-01

Context. Large surveys producing tera- and petabyte-scale databases require machine-learning and knowledge discovery methods to deal with the overwhelming quantity of data and the difficulties of extracting concise, meaningful information with reliable assessment of its uncertainty. This study investigates the potential of a few machine-learning methods for the automated analysis of eclipsing binaries in the data of such surveys. Aims: We aim to aid the extraction of samples of eclipsing binaries from such databases and to provide basic information about the objects. We intend to estimate class labels according to two different, well-known classification systems, one based on the light curve morphology (EA/EB/EW classes) and the other based on the physical characteristics of the binary system (system morphology classes; detached through overcontact systems). Furthermore, we explore low-dimensional surfaces along which the light curves of eclipsing binaries are concentrated, and consider their use in the characterization of the binary systems and in the exploration of biases of the full unknown Gaia data with respect to the training sets. Methods: We have explored the performance of principal component analysis (PCA), linear discriminant analysis (LDA), Random Forest classification and self-organizing maps (SOM) for the above aims. We pre-processed the photometric time series by combining a double Gaussian profile fit and a constrained smoothing spline, in order to de-noise and interpolate the observed light curves. We achieved further denoising, and selected the most important variability elements from the light curves using PCA. Supervised classification was performed using Random Forest and LDA based on the PC decomposition, while SOM gives a continuous 2-dimensional manifold of the light curves arranged by a few important features. We estimated the uncertainty of the supervised methods due to the specific finite training set using ensembles of models constructed on randomized training sets. Results: We obtain excellent results (about 5% global error rate) with classification into light curve morphology classes on the Hipparcos data. The classification into system morphology classes using the Catalog and Atlas of Eclipsing binaries (CALEB) has a higher error rate (about 10.5%), most importantly due to the (sometimes strong) similarity of the photometric light curves originating from physically different systems. When trained on CALEB and then applied to Kepler-detected eclipsing binaries subsampled according to Gaia observing times, LDA and SOM provide tractable, easy-to-visualize subspaces of the full (functional) space of light curves that summarize the most important phenomenological elements of the individual light curves. The sequence of light curves ordered by their first linear discriminant coefficient is compared to results obtained using local linear embedding. The SOM method proves able to find a 2-dimensional embedded surface in the space of the light curves which separates the system morphology classes in its different regions, and also identifies a few other phenomena, such as the asymmetry of the light curves due to spots, eccentric systems, and systems with a single eclipse. Furthermore, when data from other surveys are projected to the same SOM surface, the resulting map yields a good overview of the general biases and distortions due to differences in time sampling or population.

Integration of multimodal RNA-seq data for prediction of kidney cancer survival

PubMed Central

Schwartzi, Matt; Parkl, Martin; Phanl, John H.; Wang., May D.

2016-01-01

Kidney cancer is of prominent concern in modern medicine. Predicting patient survival is critical to patient awareness and developing a proper treatment regimens. Previous prediction models built upon molecular feature analysis are limited to just gene expression data. In this study we investigate the difference in predicting five year survival between unimodal and multimodal analysis of RNA-seq data from gene, exon, junction, and isoform modalities. Our preliminary findings report higher predictive accuracy-as measured by area under the ROC curve (AUC)-for multimodal learning when compared to unimodal learning with both support vector machine (SVM) and k-nearest neighbor (KNN) methods. The results of this study justify further research on the use of multimodal RNA-seq data to predict survival for other cancer types using a larger sample size and additional machine learning methods. PMID:27532026

Relevance Vector Machine and Support Vector Machine Classifier Analysis of Scanning Laser Polarimetry Retinal Nerve Fiber Layer Measurements

PubMed Central

Bowd, Christopher; Medeiros, Felipe A.; Zhang, Zuohua; Zangwill, Linda M.; Hao, Jiucang; Lee, Te-Won; Sejnowski, Terrence J.; Weinreb, Robert N.; Goldbaum, Michael H.

2010-01-01

Purpose To classify healthy and glaucomatous eyes using relevance vector machine (RVM) and support vector machine (SVM) learning classifiers trained on retinal nerve fiber layer (RNFL) thickness measurements obtained by scanning laser polarimetry (SLP). Methods Seventy-two eyes of 72 healthy control subjects (average age = 64.3 ± 8.8 years, visual field mean deviation =−0.71 ± 1.2 dB) and 92 eyes of 92 patients with glaucoma (average age = 66.9 ± 8.9 years, visual field mean deviation =−5.32 ± 4.0 dB) were imaged with SLP with variable corneal compensation (GDx VCC; Laser Diagnostic Technologies, San Diego, CA). RVM and SVM learning classifiers were trained and tested on SLP-determined RNFL thickness measurements from 14 standard parameters and 64 sectors (approximately 5.6° each) obtained in the circumpapillary area under the instrument-defined measurement ellipse (total 78 parameters). Tenfold cross-validation was used to train and test RVM and SVM classifiers on unique subsets of the full 164-eye data set and areas under the receiver operating characteristic (AUROC) curve for the classification of eyes in the test set were generated. AUROC curve results from RVM and SVM were compared to those for 14 SLP software-generated global and regional RNFL thickness parameters. Also reported was the AUROC curve for the GDx VCC software-generated nerve fiber indicator (NFI). Results The AUROC curves for RVM and SVM were 0.90 and 0.91, respectively, and increased to 0.93 and 0.94 when the training sets were optimized with sequential forward and backward selection (resulting in reduced dimensional data sets). AUROC curves for optimized RVM and SVM were significantly larger than those for all individual SLP parameters. The AUROC curve for the NFI was 0.87. Conclusions Results from RVM and SVM trained on SLP RNFL thickness measurements are similar and provide accurate classification of glaucomatous and healthy eyes. RVM may be preferable to SVM, because it provides a Bayesian-derived probability of glaucoma as an output. These results suggest that these machine learning classifiers show good potential for glaucoma diagnosis. PMID:15790898

Emotions, Self-Regulated Learning, and Achievement in Mathematics: A Growth Curve Analysis

ERIC Educational Resources Information Center

Ahmed, Wondimu; van der Werf, Greetje; Kuyper, Hans; Minnaert, Alexander

2013-01-01

The purpose of the current study was twofold: (a) to investigate the developmental trends of 4 academic emotions (anxiety, boredom, enjoyment, and pride) and (b) to examine whether changes in emotions are linked to the changes in students' self-regulatory strategies (shallow, deep, and meta-cognitive) and achievement in mathematics. Four hundred…

Dynamic Hebbian Cross-Correlation Learning Resolves the Spike Timing Dependent Plasticity Conundrum.

PubMed

Olde Scheper, Tjeerd V; Meredith, Rhiannon M; Mansvelder, Huibert D; van Pelt, Jaap; van Ooyen, Arjen

2017-01-01

Spike Timing-Dependent Plasticity has been found to assume many different forms. The classic STDP curve, with one potentiating and one depressing window, is only one of many possible curves that describe synaptic learning using the STDP mechanism. It has been shown experimentally that STDP curves may contain multiple LTP and LTD windows of variable width, and even inverted windows. The underlying STDP mechanism that is capable of producing such an extensive, and apparently incompatible, range of learning curves is still under investigation. In this paper, it is shown that STDP originates from a combination of two dynamic Hebbian cross-correlations of local activity at the synapse. The correlation of the presynaptic activity with the local postsynaptic activity is a robust and reliable indicator of the discrepancy between the presynaptic neuron and the postsynaptic neuron's activity. The second correlation is between the local postsynaptic activity with dendritic activity which is a good indicator of matching local synaptic and dendritic activity. We show that this simple time-independent learning rule can give rise to many forms of the STDP learning curve. The rule regulates synaptic strength without the need for spike matching or other supervisory learning mechanisms. Local differences in dendritic activity at the synapse greatly affect the cross-correlation difference which determines the relative contributions of different neural activity sources. Dendritic activity due to nearby synapses, action potentials, both forward and back-propagating, as well as inhibitory synapses will dynamically modify the local activity at the synapse, and the resulting STDP learning rule. The dynamic Hebbian learning rule ensures furthermore, that the resulting synaptic strength is dynamically stable, and that interactions between synapses do not result in local instabilities. The rule clearly demonstrates that synapses function as independent localized computational entities, each contributing to the global activity, not in a simply linear fashion, but in a manner that is appropriate to achieve local and global stability of the neuron and the entire dendritic structure.

Using Dragon Curves To Learn about Length and Area.

ERIC Educational Resources Information Center

Smith, Lyle R.

1999-01-01

Utilizes dragon curves which are made with three tiles and can be used to create fascinating patterns to help students understand the concepts of length, area, and perimeter of regions as defined by dragon curves. (ASK)

Study of the convergence behavior of the complex kernel least mean square algorithm.

PubMed

Paul, Thomas K; Ogunfunmi, Tokunbo

2013-09-01

The complex kernel least mean square (CKLMS) algorithm is recently derived and allows for online kernel adaptive learning for complex data. Kernel adaptive methods can be used in finding solutions for neural network and machine learning applications. The derivation of CKLMS involved the development of a modified Wirtinger calculus for Hilbert spaces to obtain the cost function gradient. We analyze the convergence of the CKLMS with different kernel forms for complex data. The expressions obtained enable us to generate theory-predicted mean-square error curves considering the circularity of the complex input signals and their effect on nonlinear learning. Simulations are used for verifying the analysis results.

Experience curve development and cost reduction disaggregation for fuel cell markets in Japan and the US

DOE Office of Scientific and Technical Information (OSTI.GOV)

Wei, Max; Smith, Sarah J.; Sohn, Michael D.

Technology learning rates can be dynamic quantities as a technology moves from early development to piloting and from low volume manufacturing to high volume manufacturing. This work describes a generalizable technology analysis approach for disaggregating observed technology cost reductions and presents results of this approach for one specific case study (micro-combined heat and power fuel cell systems in Japan). We build upon earlier reports that combine discussion of fuel cell experience curves and qualitative discussion of cost components by providing greater detail on the contributing mechanisms to observed cost reductions, which were not quantified in earlier reports. Greater standardization ismore » added to the analysis approach, which can be applied to other technologies. This paper thus provides a key linkage that has been missing from earlier literature on energy-related technologies by integrating the output of earlier manufacturing cost studies with observed learning rates to quantitatively estimate the different components of cost reduction including economies of scale and cost reductions due to product performance and product design improvements. This work also provides updated fuel cell technology price versus volume trends from the California Self-Generation Incentive Program, including extensive data for solid-oxide fuel cells (SOFC) reported here for the first time. The Japanese micro-CHP market is found to have a learning rate of 18% from 2005 to 2015, while larger SOFC fuel cell systems (200 kW and above) in the California market are found to have a flat (near-zero) learning rate, and these are attributed to a combination of exogenous, market, and policy factors.« less

Novel word acquisition in aphasia: Facing the word-referent ambiguity of natural language learning contexts.

PubMed

Peñaloza, Claudia; Mirman, Daniel; Tuomiranta, Leena; Benetello, Annalisa; Heikius, Ida-Maria; Järvinen, Sonja; Majos, Maria C; Cardona, Pedro; Juncadella, Montserrat; Laine, Matti; Martin, Nadine; Rodríguez-Fornells, Antoni

2016-06-01

Recent research suggests that some people with aphasia preserve some ability to learn novel words and to retain them in the long-term. However, this novel word learning ability has been studied only in the context of single word-picture pairings. We examined the ability of people with chronic aphasia to learn novel words using a paradigm that presents new word forms together with a limited set of different possible visual referents and requires the identification of the correct word-object associations on the basis of online feedback. We also studied the relationship between word learning ability and aphasia severity, word processing abilities, and verbal short-term memory (STM). We further examined the influence of gross lesion location on new word learning. The word learning task was first validated with a group of forty-five young adults. Fourteen participants with chronic aphasia were administered the task and underwent tests of immediate and long-term recognition memory at 1 week. Their performance was compared to that of a group of fourteen matched controls using growth curve analysis. The learning curve and recognition performance of the aphasia group was significantly below the matched control group, although above-chance recognition performance and case-by-case analyses indicated that some participants with aphasia had learned the correct word-referent mappings. Verbal STM but not word processing abilities predicted word learning ability after controlling for aphasia severity. Importantly, participants with lesions in the left frontal cortex performed significantly worse than participants with lesions that spared the left frontal region both during word learning and on the recognition tests. Our findings indicate that some people with aphasia can preserve the ability to learn a small novel lexicon in an ambiguous word-referent context. This learning and recognition memory ability was associated with verbal STM capacity, aphasia severity and the integrity of the left inferior frontal region. Copyright © 2016 Elsevier Ltd. All rights reserved.

Novel word acquisition in aphasia: Facing the word-referent ambiguity of natural language learning contexts

PubMed Central

Peñaloza, Claudia; Mirman, Daniel; Tuomiranta, Leena; Benetello, Annalisa; Heikius, Ida-Maria; Järvinen, Sonja; Majos, Maria C.; Cardona, Pedro; Juncadella, Montserrat; Laine, Matti; Martin, Nadine; Rodríguez-Fornells, Antoni

2017-01-01

Recent research suggests that some people with aphasia preserve some ability to learn novel words and to retain them in the long-term. However, this novel word learning ability has been studied only in the context of single word-picture pairings. We examined the ability of people with chronic aphasia to learn novel words using a paradigm that presents new word forms together with a limited set of different possible visual referents and requires the identification of the correct word-object associations on the basis of online feedback. We also studied the relationship between word learning ability and aphasia severity, word processing abilities, and verbal short-term memory (STM). We further examined the influence of gross lesion location on new word learning. The word learning task was first validated with a group of forty-five young adults. Fourteen participants with chronic aphasia were administered the task and underwent tests of immediate and long-term recognition memory at 1 week. Their performance was compared to that of a group of fourteen matched controls using growth curve analysis. The learning curve and recognition performance of the aphasia group was significantly below the matched control group, although above-chance recognition performance and case-by-case analyses indicated that some participants with aphasia had learned the correct word-referent mappings. Verbal STM but not word processing abilities predicted word learning ability after controlling for aphasia severity. Importantly, participants with lesions in the left frontal cortex performed significantly worse than participants with lesions that spared the left frontal region both during word learning and on the recognition tests. Our findings indicate that some people with aphasia can preserve the ability to learn a small novel lexicon in an ambiguous word-referent context. This learning and recognition memory ability was associated with verbal STM capacity, aphasia severity and the integrity of the left inferior frontal region. PMID:27085892

Comparison of the learning curves and frustration level in performing laparoscopic and robotic training skills by experts and novices.

PubMed

Passerotti, Carlo C; Franco, Felipe; Bissoli, Julio C C; Tiseo, Bruno; Oliveira, Caio M; Buchalla, Carlos A O; Inoue, Gustavo N C; Sencan, Arzu; Sencan, Aydin; do Pardo, Rogerio Ruscitto; Nguyen, Hiep T

2015-07-01

Robotic assistance may provide for distinct technical advantages over conventional laparoscopic technique. The goals of this study were (1) to objectively evaluate the difference in the learning curves by novice and expert surgeons in performing fundamental laparoscopic skills using conventional laparoscopic surgery (CLS) and robotic-assisted laparoscopic surgery (RALS) and (2) to evaluate the surgeons' frustration level in performing these tasks. Twelve experienced and 31 novices in laparoscopy were prospectively evaluated in performing three standardized laparoscopic tasks in five consecutive, weekly training sessions. Analysis of the learning curves was based on the magnitude, rate, and quickness in performance improvement. The participant's frustration and mood were also evaluated during and after every session. For the novice participants, RALS allowed for shorter time to task completion and greater accuracy. However, significant and rapid improvement in performance as measured by magnitude, rate, and quickness at each session was also seen with CLS. For the experienced surgeons, RALS only provided a slight improvement in performance. For all participants, the use of RALS was associated with less number of sessions in which they felt frustrated, less number of frustration episodes during a session, lower frustration score during and after the session, and higher good mood score. The advantages of RALS may be of most benefit when doing more complex tasks and by less experienced surgeons. RALS should not be used as a replacement for CLS but rather in specific situations in which it has the greatest advantages.

Minimally invasive video-assisted thyroidectomy: Ascending the learning curve

PubMed Central

Capponi, Michela Giulii; Bellotti, Carlo; Lotti, Marco; Ansaloni, Luca

2015-01-01

BACKGROUND: Minimally invasive video-assisted thyroidectomy (MIVAT) is a technically demanding procedure and requires a surgical team skilled in both endocrine and endoscopic surgery. The aim of this report is to point out some aspects of the learning curve of the video-assisted thyroid surgery, through the analysis of our preliminary series of procedures. PATIENTS AND METHODS: Over a period of 8 months, we selected 36 patients for minimally invasive video-assisted surgery of the thyroid. The patients were considered eligible if they presented with a nodule not exceeding 35 mm and total thyroid volume <20 ml; presence of biochemical and ultrasound signs of thyroiditis and pre-operative diagnosis of cancer were exclusion criteria. We analysed surgical results, conversion rate, operating time, post-operative complications, hospital stay and cosmetic outcomes of the series. RESULTS: We performed 36 total thyroidectomy and in one case we performed a consensual parathyroidectomy. The procedure was successfully carried out in 33 out of 36 cases (conversion rate 8.3%). The mean operating time was 109 min (range: 80-241 min) and reached a plateau after 29 MIVAT. Post-operative complications included three transient recurrent nerve palsies and two transient hypocalcemias; no definitive hypoparathyroidism was registered. The cosmetic result was considered excellent by most patients. CONCLUSIONS: Advances in skills and technology allow surgeons to easily reproduce the standard open total thyroidectomy with video-assistance. Although the learning curve represents a time-consuming step, training remains a crucial point in gaining a reasonable confidence with video-assisted surgical technique. PMID:25883451

Assessment of a Learning Strategy among Spine Surgeons.

PubMed

Gotfryd, Alberto Ofenhejm; Corredor, Jose Alfredo; Teixeira, William Jacobsen; Martins, Delio Eulálio; Milano, Jeronimo; Iutaka, Alexandre Sadao

2017-02-01

Pilot test, observational study. To evaluate objectively the knowledge transfer provided by theoretical and practical activities during AOSpine courses for spine surgeons. During two AOSpine principles courses, 62 participants underwent precourse assessment, which consisted of questions about their professional experience, preferences regarding adolescent idiopathic scoliosis (AIS) classification, and classifying the curves by means of the Lenke classification of two AIS clinical cases. Two learning strategies were used during the course. A postcourse questionnaire was applied to reclassify the same deformity cases. Differences in the correct answers of clinical cases between pre- and postcourse were analyzed, revealing the number of participants whose accuracy in classification improved after the course. Analysis showed a decrease in the number of participants with wrong answers in both cases after the course. In the first case, statistically significant differences were observed in both curve pattern (83.3%, p   =  0.005) and lumbar spine modifier (46.6%, p   =  0.049). No statistically significant improvement was seen in the sagittal thoracic modifier (33.3%, p   =  0.309). In the second case, statistical improvement was obtained in curve pattern (27.4%, p   =  0.018). No statistically significant improvement was seen regarding lumbar spine modifier (9.8%, p   =  0.121) and sagittal thoracic modifier (12.9%, p   =  0.081). This pilot test showed objectively that learning strategies used during AOSpine courses improved the participants' knowledge. Teaching strategies must be continually improved to ensure an optimal level of knowledge transfer.

Assessment of a Learning Strategy among Spine Surgeons

PubMed Central

Gotfryd, Alberto Ofenhejm; Teixeira, William Jacobsen; Martins, Delio Eulálio; Milano, Jeronimo; Iutaka, Alexandre Sadao

2017-01-01

Study Design Pilot test, observational study. Objective To evaluate objectively the knowledge transfer provided by theoretical and practical activities during AOSpine courses for spine surgeons. Methods During two AOSpine principles courses, 62 participants underwent precourse assessment, which consisted of questions about their professional experience, preferences regarding adolescent idiopathic scoliosis (AIS) classification, and classifying the curves by means of the Lenke classification of two AIS clinical cases. Two learning strategies were used during the course. A postcourse questionnaire was applied to reclassify the same deformity cases. Differences in the correct answers of clinical cases between pre- and postcourse were analyzed, revealing the number of participants whose accuracy in classification improved after the course. Results Analysis showed a decrease in the number of participants with wrong answers in both cases after the course. In the first case, statistically significant differences were observed in both curve pattern (83.3%, p  =  0.005) and lumbar spine modifier (46.6%, p  =  0.049). No statistically significant improvement was seen in the sagittal thoracic modifier (33.3%, p  =  0.309). In the second case, statistical improvement was obtained in curve pattern (27.4%, p  =  0.018). No statistically significant improvement was seen regarding lumbar spine modifier (9.8%, p  =  0.121) and sagittal thoracic modifier (12.9%, p  =  0.081). Conclusion This pilot test showed objectively that learning strategies used during AOSpine courses improved the participants' knowledge. Teaching strategies must be continually improved to ensure an optimal level of knowledge transfer. PMID:28451507

Laparoscopy Instructional Videos: The Effect of Preoperative Compared With Intraoperative Use on Learning Curves.

PubMed

Broekema, Theo H; Talsma, Aaldert K; Wevers, Kevin P; Pierie, Jean-Pierre E N

Previous studies have shown that the use of intraoperative instructional videos has a positive effect on learning laparoscopic procedures. This study investigated the effect of the timing of the instructional videos on learning curves in laparoscopic skills training. After completing a basic skills course on a virtual reality simulator, medical students and residents with less than 1 hour experience using laparoscopic instruments were randomized into 2 groups. Using an instructional video either preoperatively or intraoperatively, both groups then performed 4 repetitions of a standardized task on the TrEndo augmented reality. With the TrEndo, 9 motion analysis parameters (MAPs) were recorded for each session (4 MAPs for each hand and time). These were the primary outcome measurements for performance. The time spent watching the instructional video was also recorded. Improvement in performance was studied within and between groups. Medical Center Leeuwarden, a secondary care hospital located in Leeuwarden, The Netherlands. Right-hand dominant medical student and residents with more than 1 hour experience operating any kind of laparoscopic instruments were participated. A total of 23 persons entered the study, of which 21 completed the study course. In both groups, at least 5 of 9 MAPs showed significant improvements between repetition 1 and 4. When both groups were compared after completion of repetition 4, no significant differences in improvement were detected. The intraoperative group showed significant improvement in 3 MAPs of the left-nondominant-hand, compared with one MAP for the preoperative group. No significant differences in learning curves could be detected between the subjects who used intraoperative instructional videos and those who used preoperative instructional videos. Intraoperative video instruction may result in improved dexterity of the nondominant hand. Copyright © 2016 Association of Program Directors in Surgery. Published by Elsevier Inc. All rights reserved.

The learning curve associated with the introduction of the subcutaneous implantable defibrillator.

PubMed

Knops, Reinoud E; Brouwer, Tom F; Barr, Craig S; Theuns, Dominic A; Boersma, Lucas; Weiss, Raul; Neuzil, Petr; Scholten, Marcoen; Lambiase, Pier D; Leon, Angel R; Hood, Margaret; Jones, Paul W; Wold, Nicholas; Grace, Andrew A; Olde Nordkamp, Louise R A; Burke, Martin C

2016-07-01

The subcutaneous implantable cardioverter defibrillator (S-ICD) was introduced to overcome complications related to transvenous leads. Adoption of the S-ICD requires implanters to learn a new implantation technique. The aim of this study was to assess the learning curve for S-ICD implanters with respect to implant-related complications, procedure time, and inappropriate shocks (IASs). In a pooled cohort from two clinical S-ICD databases, the IDE Trial and the EFFORTLESS Registry, complications, IASs at 180 days follow-up and implant procedure duration were assessed. Patients were grouped in quartiles based on experience of the implanter and Kaplan-Meier estimates of complication and IAS rates were calculated. A total of 882 patients implanted in 61 centres by 107 implanters with a median of 4 implants (IQR 1,8) were analysed. There were a total of 59 patients with complications and 48 patients with IAS. The complication rate decreased significantly from 9.8% in Quartile 1 (least experience) to 5.4% in Quartile 4 (most experience) (P = 0.02) and non-significantly for IAS from 7.9 to 4.8% (P = 0.10). Multivariable analysis demonstrated a hazard ratio of 0.78 (P = 0.045) for complications and 1.01 (P = 0.958) for IAS. Dual-zone programming increased with experience of the individual implanter (P < 0.001), which reduced IAS significantly in the multivariable model (HR 0.44, P = 0.01). Procedure time decreased from 75 to 65 min (P < 0.001). The complication rate and procedure time stabilized after Quartile 2 (>13 implants). There is a short and significant learning curve associated with physicians adopting the S-ICD. Performance stabilizes after 13 implants. © The Author 2015. Published by Oxford University Press on behalf of the European Society of Cardiology.

Deep Learning Accurately Predicts Estrogen Receptor Status in Breast Cancer Metabolomics Data.

PubMed

Alakwaa, Fadhl M; Chaudhary, Kumardeep; Garmire, Lana X

2018-01-05

Metabolomics holds the promise as a new technology to diagnose highly heterogeneous diseases. Conventionally, metabolomics data analysis for diagnosis is done using various statistical and machine learning based classification methods. However, it remains unknown if deep neural network, a class of increasingly popular machine learning methods, is suitable to classify metabolomics data. Here we use a cohort of 271 breast cancer tissues, 204 positive estrogen receptor (ER+), and 67 negative estrogen receptor (ER-) to test the accuracies of feed-forward networks, a deep learning (DL) framework, as well as six widely used machine learning models, namely random forest (RF), support vector machines (SVM), recursive partitioning and regression trees (RPART), linear discriminant analysis (LDA), prediction analysis for microarrays (PAM), and generalized boosted models (GBM). DL framework has the highest area under the curve (AUC) of 0.93 in classifying ER+/ER- patients, compared to the other six machine learning algorithms. Furthermore, the biological interpretation of the first hidden layer reveals eight commonly enriched significant metabolomics pathways (adjusted P-value <0.05) that cannot be discovered by other machine learning methods. Among them, protein digestion and absorption and ATP-binding cassette (ABC) transporters pathways are also confirmed in integrated analysis between metabolomics and gene expression data in these samples. In summary, deep learning method shows advantages for metabolomics based breast cancer ER status classification, with both the highest prediction accuracy (AUC = 0.93) and better revelation of disease biology. We encourage the adoption of feed-forward networks based deep learning method in the metabolomics research community for classification.

Non-Constant Learning Rates in Retrospective Experience Curve Analyses and their Correlation to Deployment Programs

DOE Office of Scientific and Technical Information (OSTI.GOV)

Wei, Max; Smith, Sarah J.; Sohn, Michael D.

2015-07-16

A key challenge for policy-makers and technology market forecasters is to estimate future technology costs and in particular the rate of cost reduction versus production volume. A related, critical question is what role should state and federal governments have in advancing energy efficient and renewable energy technologies? This work provides retrospective experience curves and learning rates for several energy-related technologies, each of which have a known history of federal and state deployment programs. We derive learning rates for eight technologies including energy efficient lighting technologies, stationary fuel cell systems, and residential solar photovoltaics, and provide an overview and timeline ofmore » historical deployment programs such as state and federal standards and state and national incentive programs for each technology. Piecewise linear regimes are observed in a range of technology experience curves, and public investments or deployment programs are found to be strongly correlated to an increase in learning rate across multiple technologies. A downward bend in the experience curve is found in 5 out of the 8 energy-related technologies presented here (electronic ballasts, magnetic ballasts, compact fluorescent lighting, general service fluorescent lighting, and the installed cost of solar PV). In each of the five downward-bending experience curves, we believe that an increase in the learning rate can be linked to deployment programs to some degree. This work sheds light on the endogenous versus exogenous contributions to technological innovation and highlights the impact of exogenous government sponsored deployment programs. This work can inform future policy investment direction and can shed light on market transformation and technology learning behavior.« less

Modeling Patterns of Activities using Activity Curves

PubMed Central

Dawadi, Prafulla N.; Cook, Diane J.; Schmitter-Edgecombe, Maureen

2016-01-01

Pervasive computing offers an unprecedented opportunity to unobtrusively monitor behavior and use the large amount of collected data to perform analysis of activity-based behavioral patterns. In this paper, we introduce the notion of an activity curve, which represents an abstraction of an individual’s normal daily routine based on automatically-recognized activities. We propose methods to detect changes in behavioral routines by comparing activity curves and use these changes to analyze the possibility of changes in cognitive or physical health. We demonstrate our model and evaluate our change detection approach using a longitudinal smart home sensor dataset collected from 18 smart homes with older adult residents. Finally, we demonstrate how big data-based pervasive analytics such as activity curve-based change detection can be used to perform functional health assessment. Our evaluation indicates that correlations do exist between behavior and health changes and that these changes can be automatically detected using smart homes, machine learning, and big data-based pervasive analytics. PMID:27346990

Modeling Patterns of Activities using Activity Curves.

PubMed

Dawadi, Prafulla N; Cook, Diane J; Schmitter-Edgecombe, Maureen

2016-06-01

Pervasive computing offers an unprecedented opportunity to unobtrusively monitor behavior and use the large amount of collected data to perform analysis of activity-based behavioral patterns. In this paper, we introduce the notion of an activity curve , which represents an abstraction of an individual's normal daily routine based on automatically-recognized activities. We propose methods to detect changes in behavioral routines by comparing activity curves and use these changes to analyze the possibility of changes in cognitive or physical health. We demonstrate our model and evaluate our change detection approach using a longitudinal smart home sensor dataset collected from 18 smart homes with older adult residents. Finally, we demonstrate how big data-based pervasive analytics such as activity curve-based change detection can be used to perform functional health assessment. Our evaluation indicates that correlations do exist between behavior and health changes and that these changes can be automatically detected using smart homes, machine learning, and big data-based pervasive analytics.

Using Technology in Teacher Preparation: Two Mature Teacher Educators Negotiate the Steep Learning Curve

ERIC Educational Resources Information Center

Monroe, Eula; Tolman, Marvin

2004-01-01

This paper chronicles the ventures of two mature faculty members who continue to negotiate their own steep learning curves in helping teacher education students use current technology. It describes the scaffolding provided within the university setting for the faculty members' growth. Included are elements supported by a PT3 grant that have…

The Pedagogy of Teaching Educational Vision: A Vision Coach's Field Notes about Leaders as Learners

ERIC Educational Resources Information Center

Schein, Jeffrey

2009-01-01

The emerging field of educational visioning is full of challenges and phenomena worthy of careful analysis and documentation. A relatively neglected phenomenon is the learning curve of the leaders (often lay leaders) involved in the visioning process. This article documents a range of experiences of the author serving as a vision coach to five…

A Comparison of Narrative and Expository Text Comprehension for Students at Varying Levels of SES: A Latent Growth Curve Analysis

ERIC Educational Resources Information Center

Briggs, Laura Clark

2017-01-01

Research on secondary student reading comprehension performance is scant, yet demands for improved literacy at college and career levels indicate that an understanding of trends and growth patterns is necessary to better inform teaching and learning for high school students. To improve understanding of reading performance at the secondary level,…

Autoshaped Head Poking in the Mouse: A Quantitative Analysis of the Learning Curve

ERIC Educational Resources Information Center

Papachristos, Efstathios B.; Gallistel, C. R.

2006-01-01

In autoshaping experiments, we quantified the acquisition of anticipatory head poking in individual mice, using an algorithm that finds changes in the slope of a cumulative record. In most mice, upward changes in the amount of anticipatory poking per trial were abrupt, and tended to occur at session boundaries, suggesting that the session is as…

Search strategy selection in the Morris water maze indicates allocentric map formation during learning that underpins spatial memory formation.

PubMed

Rogers, Jake; Churilov, Leonid; Hannan, Anthony J; Renoir, Thibault

2017-03-01

Using a Matlab classification algorithm, we demonstrate that a highly salient distal cue array is required for significantly increased likelihoods of spatial search strategy selection during Morris water maze spatial learning. We hypothesized that increased spatial search strategy selection during spatial learning would be the key measure demonstrating the formation of an allocentric map to the escape location. Spatial memory, as indicated by quadrant preference for the area of the pool formally containing the hidden platform, was assessed as the main measure that this allocentric map had formed during spatial learning. Our C57BL/6J wild-type (WT) mice exhibit quadrant preference in the highly salient cue paradigm but not the low, corresponding with a 120% increase in the odds of a spatial search strategy selection during learning. In contrast, quadrant preference remains absent in serotonin 1A receptor (5-HT 1A R) knockout (KO) mice, who exhibit impaired search strategy selection during spatial learning. Additionally, we also aimed to assess the impact of the quality of the distal cue array on the spatial learning curves of both latency to platform and path length using mixed-effect regression models and found no significant associations or interactions. In contrast, we demonstrated that the spatial learning curve for search strategy selection was absent during training in the low saliency paradigm. Therefore, we propose that allocentric search strategy selection during spatial learning is the learning parameter in mice that robustly indicates the formation of a cognitive map for the escape goal location. These results also suggest that both latency to platform and path length spatial learning curves do not discriminate between allocentric and egocentric spatial learning and do not reliably predict spatial memory formation. We also show that spatial memory, as indicated by the absolute time in the quadrant formerly containing the hidden platform alone (without reference to the other areas of the pool), was not sensitive to cue saliency or impaired in 5-HT 1A R KO mice. Importantly, in the absence of a search strategy analysis, this suggests that to establish that the Morris water maze has worked (i.e. control mice have formed an allocentric map to the escape goal location), a measure of quadrant preference needs to be reported to establish spatial memory formation. This has implications for studies that claim hippocampal functioning is impaired using latency to platform or path length differences within the existing Morris water maze literature. Copyright © 2016 Elsevier Inc. All rights reserved.

Evaluating the Bias of Alternative Cost Progress Models: Tests Using Aerospace Industry Acquisition Programs

DTIC Science & Technology

1992-12-01

suspect :mat, -n2 extent predict:.on cas jas ccsiziveiv crrei:=e amonc e v:arious models, :he fandom *.;aik, learn ha r ur e, i;<ea- variable and Bemis...Functions, Production Rate Adjustment Model, Learning Curve Model. Random Walk Model. Bemis Model. Evaluating Model Bias, Cost Prediction Bias. Cost...of four cost progress models--a random walk model, the tradiuonai learning curve model, a production rate model Ifixed-variable model). and a model

Using SERC for creating and publishing student generated hydrology instruction materials

NASA Astrophysics Data System (ADS)

Merwade, V.; Rajib, A.; Ruddell, B.; Fox, S.

2016-12-01

Hydrology instruction typically involves teaching of the hydrologic cycle and the processes associated with it such as precipitation, evapotranspiration, infiltration, runoff generation and hydrograph analysis. With the availability of observed and remotely sensed data in public domain, there is an opportunity to incorporate place-based learning in hydrology classrooms. However, it is not always easy and possible for an instructor to complement an existing hydrology course with new material that requires both time and technical expertise, which the instructor may not have. The work presented here describes an effort where students created the data and modeling driven instruction materials as part of their class assignment for a hydrology course at Purdue University. Students in the class were divided into groups, and each group was assigned a topic such as precipitation, evapotranspiration, streamflow, flow duration curve and flood frequency analysis. Each of the student groups was then instructed to produce an instruction material showing ways to extract/process relevant data and perform some analysis for an area with specific land use characteristic. The student contributions were then organized into learning units such that someone can do a flow duration curve analysis or flood frequency analysis and see how it changes for rural area versus urban area. Science Education Resource Center (SERC) is used as a platform to publish and share these instruction materials so it can be used as-is or through modification by any instructor or student in relevant coursework anywhere in the world.

Progress in Energy Storage Technologies: Models and Methods for Policy Analysis

NASA Astrophysics Data System (ADS)

Matteson, Schuyler W.

Climate change and other sustainability challenges have led to the development of new technologies that increase energy efficiency and reduce the utilization of finite resources. To promote the adoption of technologies with social benefits, governments often enact policies that provide financial incentives at the point of purchase. In their current form, these subsidies have the potential to increase the diffusion of emerging technologies; however, accounting for technological progress can improve program success while decreasing net public investment. This research develops novel methods using experience curves for the development of more efficient subsidy policies. By providing case studies in the field of automotive energy storage technologies, this dissertation also applies the methods to show the impacts of incorporating technological progress into energy policies. Specific findings include learning-dependent tapering subsidies for electric vehicles based on the lithium-ion battery experience curve, the effects of residual learning rates in lead-acid batteries on emerging technology cost competitiveness, and a cascading diffusion assessment of plug-in hybrid electric vehicle subsidy programs. Notably, the results show that considering learning rates in policy development can save billions of dollars in public funds, while also lending insight into the decision of whether or not to subsidize a given technology.

Early predictors of helpless thoughts and behaviors in children: developmental precursors to depressive cognitions.

PubMed

Cole, David A; Warren, Dana E; Dallaire, Danielle H; Lagrange, Beth; Travis, Rebekah; Ciesla, Jeffrey A

2007-04-01

Learned helplessness behavior and cognitions were assessed in 95 kindergarten-age children during a series of impossible puzzle trials followed by a solvable puzzle trial. Latent growth curve analysis revealed reliable individual differences in the trajectories of children's affect, motivation, and self-cognitions over time. Parents' reports of negative life events, harsh/negative parenting, and warm/positive parenting were associated with their children's learned helplessness behavioral trajectories and outcomes over the course of the puzzle trials. Results support speculations about the developmental origins of depressive explanatory or attributional style in children.

An object location memory paradigm for older adults with and without mild cognitive impairment.

PubMed

Külzow, Nadine; Kerti, Lucia; Witte, Veronica A; Kopp, Ute; Breitenstein, Caterina; Flöel, Agnes

2014-11-30

Object-location memory is critical in every-day life and known to deteriorate early in the course of neurodegenerative disease. We adapted the previously established learning paradigm "LOCATO" for use in healthy older adults and patients with mild cognitive impairment (MCI). Pictures of real-life buildings were associated with positions on a two-dimensional street map by repetitions of "correct" object-location pairings over the course of five training blocks, followed by a recall task. Correct/incorrect associations were indicated by button presses. The original two 45-item sets were reduced to 15 item-sets, and tested in healthy older adults and MCI for learning curve, recall, and re-test effects. The two 15-item versions showed comparable learning curves and recall scores within each group. While learning curves increased linearly in both groups, MCI patients performed significantly worse on learning and recall compared to healthy controls. Re-testing after 6 month showed small practice effects only. LOCATO is a simple standardized task that overcomes several limitation of previously employed visuospatial task by using real-life stimuli, minimizing verbal encoding, avoiding fine motor responses, combining explicit and implicit statistical learning, and allowing to assess learning curve in addition to recall. Results show that the shortened version of LOCATO meets the requirements for a robust and ecologically meaningful assessment of object-location memory in older adults with and without MCI. It can now be used to systematically assess acquisition of object-location memory and its modulation through adjuvant therapies like pharmacological or non-invasive brain stimulation. Copyright © 2014 Elsevier B.V. All rights reserved.

Cognitive Machine-Learning Algorithm for Cardiac Imaging: A Pilot Study for Differentiating Constrictive Pericarditis From Restrictive Cardiomyopathy.

PubMed

Sengupta, Partho P; Huang, Yen-Min; Bansal, Manish; Ashrafi, Ali; Fisher, Matt; Shameer, Khader; Gall, Walt; Dudley, Joel T

2016-06-01

Associating a patient's profile with the memories of prototypical patients built through previous repeat clinical experience is a key process in clinical judgment. We hypothesized that a similar process using a cognitive computing tool would be well suited for learning and recalling multidimensional attributes of speckle tracking echocardiography data sets derived from patients with known constrictive pericarditis and restrictive cardiomyopathy. Clinical and echocardiographic data of 50 patients with constrictive pericarditis and 44 with restrictive cardiomyopathy were used for developing an associative memory classifier-based machine-learning algorithm. The speckle tracking echocardiography data were normalized in reference to 47 controls with no structural heart disease, and the diagnostic area under the receiver operating characteristic curve of the associative memory classifier was evaluated for differentiating constrictive pericarditis from restrictive cardiomyopathy. Using only speckle tracking echocardiography variables, associative memory classifier achieved a diagnostic area under the curve of 89.2%, which improved to 96.2% with addition of 4 echocardiographic variables. In comparison, the area under the curve of early diastolic mitral annular velocity and left ventricular longitudinal strain were 82.1% and 63.7%, respectively. Furthermore, the associative memory classifier demonstrated greater accuracy and shorter learning curves than other machine-learning approaches, with accuracy asymptotically approaching 90% after a training fraction of 0.3 and remaining flat at higher training fractions. This study demonstrates feasibility of a cognitive machine-learning approach for learning and recalling patterns observed during echocardiographic evaluations. Incorporation of machine-learning algorithms in cardiac imaging may aid standardized assessments and support the quality of interpretations, particularly for novice readers with limited experience. © 2016 American Heart Association, Inc.

Learning curve for new technology?: a nationwide register-based study of 46,363 total knee arthroplasties.

PubMed

Peltola, Mikko; Malmivaara, Antti; Paavola, Mika

2013-12-04

The risk of early revision is increased for the first patients operatively treated with a newly introduced knee prosthesis. In this study, we explored the learning curves associated with ten knee implant models to determine their effect on early revision risk. We studied register data from all seventy-five surgical units that performed knee arthroplasty in Finland from 1998 to 2007. Of 54,925 patients (66,098 knees), 39,528 patients (46,363 knees) underwent arthroplasty for osteoarthritis of the knee with the ten most common total knee implants and were followed with complete data until December 31, 2010, or the time of death. We used a Cox proportional-hazards regression model for calculating the hazard ratios for early revision for the first fifteen arthroplasties and subsequent increments of numbers of arthroplasties. We found large differences among knee implants at the introduction with regard to the risk of early revision, as well as for the overall risk of early revision. A learning curve was found for four implant models, while six models did not show a learning effect on the risk of early revision. The survivorship of the studied prostheses showed substantial differences. Knee implants have model-specific learning curves and early revision risks. Some models are more difficult to implement than others. The manufacturers should consider the learning effect when designing implants and instrumentation. The surgeons should thoroughly familiarize themselves with the new knee implants before use.

Advanced Online Survival Analysis Tool for Predictive Modelling in Clinical Data Science.

PubMed

Montes-Torres, Julio; Subirats, José Luis; Ribelles, Nuria; Urda, Daniel; Franco, Leonardo; Alba, Emilio; Jerez, José Manuel

2016-01-01

One of the prevailing applications of machine learning is the use of predictive modelling in clinical survival analysis. In this work, we present our view of the current situation of computer tools for survival analysis, stressing the need of transferring the latest results in the field of machine learning to biomedical researchers. We propose a web based software for survival analysis called OSA (Online Survival Analysis), which has been developed as an open access and user friendly option to obtain discrete time, predictive survival models at individual level using machine learning techniques, and to perform standard survival analysis. OSA employs an Artificial Neural Network (ANN) based method to produce the predictive survival models. Additionally, the software can easily generate survival and hazard curves with multiple options to personalise the plots, obtain contingency tables from the uploaded data to perform different tests, and fit a Cox regression model from a number of predictor variables. In the Materials and Methods section, we depict the general architecture of the application and introduce the mathematical background of each of the implemented methods. The study concludes with examples of use showing the results obtained with public datasets.

Advanced Online Survival Analysis Tool for Predictive Modelling in Clinical Data Science

PubMed Central

Montes-Torres, Julio; Subirats, José Luis; Ribelles, Nuria; Urda, Daniel; Franco, Leonardo; Alba, Emilio; Jerez, José Manuel

2016-01-01

One of the prevailing applications of machine learning is the use of predictive modelling in clinical survival analysis. In this work, we present our view of the current situation of computer tools for survival analysis, stressing the need of transferring the latest results in the field of machine learning to biomedical researchers. We propose a web based software for survival analysis called OSA (Online Survival Analysis), which has been developed as an open access and user friendly option to obtain discrete time, predictive survival models at individual level using machine learning techniques, and to perform standard survival analysis. OSA employs an Artificial Neural Network (ANN) based method to produce the predictive survival models. Additionally, the software can easily generate survival and hazard curves with multiple options to personalise the plots, obtain contingency tables from the uploaded data to perform different tests, and fit a Cox regression model from a number of predictor variables. In the Materials and Methods section, we depict the general architecture of the application and introduce the mathematical background of each of the implemented methods. The study concludes with examples of use showing the results obtained with public datasets. PMID:27532883

The Parallel Episodic Processing (PEP) model 2.0: A single computational model of stimulus-response binding, contingency learning, power curves, and mixing costs.

PubMed

Schmidt, James R; De Houwer, Jan; Rothermund, Klaus

2016-12-01

The current paper presents an extension of the Parallel Episodic Processing model. The model is developed for simulating behaviour in performance (i.e., speeded response time) tasks and learns to anticipate both how and when to respond based on retrieval of memories of previous trials. With one fixed parameter set, the model is shown to successfully simulate a wide range of different findings. These include: practice curves in the Stroop paradigm, contingency learning effects, learning acquisition curves, stimulus-response binding effects, mixing costs, and various findings from the attentional control domain. The results demonstrate several important points. First, the same retrieval mechanism parsimoniously explains stimulus-response binding, contingency learning, and practice effects. Second, as performance improves with practice, any effects will shrink with it. Third, a model of simple learning processes is sufficient to explain phenomena that are typically (but perhaps incorrectly) interpreted in terms of higher-order control processes. More generally, we argue that computational models with a fixed parameter set and wider breadth should be preferred over those that are restricted to a narrow set of phenomena. Copyright © 2016 Elsevier Inc. All rights reserved.

SEGMENTATION OF MITOCHONDRIA IN ELECTRON MICROSCOPY IMAGES USING ALGEBRAIC CURVES.

PubMed

Seyedhosseini, Mojtaba; Ellisman, Mark H; Tasdizen, Tolga

2013-01-01

High-resolution microscopy techniques have been used to generate large volumes of data with enough details for understanding the complex structure of the nervous system. However, automatic techniques are required to segment cells and intracellular structures in these multi-terabyte datasets and make anatomical analysis possible on a large scale. We propose a fully automated method that exploits both shape information and regional statistics to segment irregularly shaped intracellular structures such as mitochondria in electron microscopy (EM) images. The main idea is to use algebraic curves to extract shape features together with texture features from image patches. Then, these powerful features are used to learn a random forest classifier, which can predict mitochondria locations precisely. Finally, the algebraic curves together with regional information are used to segment the mitochondria at the predicted locations. We demonstrate that our method outperforms the state-of-the-art algorithms in segmentation of mitochondria in EM images.

Mediators, Moderators, and Predictors of 1-Year Outcomes Among Children Treated for Early-Onset Conduct Problems: A Latent Growth Curve Analysis

ERIC Educational Resources Information Center

Beauchaine, Theodore P.; Webster-Stratton, Carolyn; Reid, M. Jamila

2006-01-01

Several child conduct problem interventions have been classified as either efficacious or well established. Nevertheless, much remains to be learned about predictors of treatment response and mechanisms of behavioral change. In this study, the authors combine data from 6 randomized clinical trials and 514 children, ages 3.0-8.5 years, to evaluate…

Analysis of the Effects of Fixed Costs on Learning Curve Calculations

DTIC Science & Technology

1994-09-01

Gansler, Jacques S . The Defense Industry. Cambridge MA: MIT Press, 1980. 11. Horngren , Charles T. and George Foster. Cost Accounting : A Managerial...Incorrect Total Cost Estimates and Comparison to Correct/Correct Total C o st E stim a te s ...7 1 12. Incorrect/Correct Total Cost Estimates and Comparison to Correct/Correct Total C o st E stim a te s

Capturing the Cumulative Effects of School Reform: An 11-Year Study of the Impacts of America's Choice on Student Achievement

ERIC Educational Resources Information Center

May, Henry; Supovitz, Jonathan A.

2006-01-01

This article presents the results of an 11-year longitudinal study of the impact of America's Choice comprehensive school reform (CSR) design on student learning gains in Rochester, New York. A quasi-experimental interrupted time-series approach using Bayesian hierarchical growth curve analysis with crossed random effects is used to compare the…

Clinical and Radiographic Outcomes With Assessment of the Learning Curve in Arthroscopically Assisted Latissimus Dorsi Tendon Transfer for Irreparable Posterosuperior Rotator Cuff Tears.

PubMed

Yamakado, Kotaro

2017-12-01

To evaluate the clinical results of an arthroscopy-assisted latissimus dorsi tendon transfer (aLD) for irreparable posterosuperior cuff tears as a primary surgery. The secondary aim of this study was to quantify the learning curve using the log-linear model. We hypothesized that aLD significantly improved shoulder function and that there was consistent reduction of the operative time in support of a learning-curve effect. After the arthroscopic partial repair was completed, the latissimus dorsi tendon was harvested via axillary mini-open incision and fixed with a knotless anchor arthroscopically. All patients were evaluated preoperatively and postoperatively using a modified University of California Los Angeles (UCLA) scoring system, active range of motion, and the visual analog scale (VAS) for pain. The operative time was recorded to quantify the learning curve using a log-linear model. Thirty patients with a mean age of 67.4 years who underwent aLD were included. At a mean of 34 months after an aLD, the mean UCLA score increased from 15.7 preoperatively to 28.8 postoperatively (P < .001). The mean active forward elevation increased from 105° preoperatively to 149° postoperatively (P < .001). The mean active external rotation increased from 22° preoperatively to 32° postoperatively (P < .001). The VAS improved from 58 mm to 18 mm (P < .001). In all but 2 cases (93%), the preoperative osteoarthritis grade was maintained. The mean operative time was 145 minutes. A significant linear correlation was observed between the operative time and cumulative volume of cases after performing a logarithmic transformation. The learning rate was calculated as 84%. Arthroscopy-assisted latissimus dorsi tendon transfer is a technically demanding procedure; however, it can lead to significant improvements in overall shoulder pain and function. This study also confirmed a learning-curve effect for the aLD. The learning rate was 84%, indicating the existence of a long learning period. Level IV, therapeutic case series. Copyright © 2017 Arthroscopy Association of North America. Published by Elsevier Inc. All rights reserved.

Deep learning in breast cancer risk assessment: evaluation of convolutional neural networks on a clinical dataset of full-field digital mammograms.

PubMed

Li, Hui; Giger, Maryellen L; Huynh, Benjamin Q; Antropova, Natalia O

2017-10-01

To evaluate deep learning in the assessment of breast cancer risk in which convolutional neural networks (CNNs) with transfer learning are used to extract parenchymal characteristics directly from full-field digital mammographic (FFDM) images instead of using computerized radiographic texture analysis (RTA), 456 clinical FFDM cases were included: a "high-risk" BRCA1/2 gene-mutation carriers dataset (53 cases), a "high-risk" unilateral cancer patients dataset (75 cases), and a "low-risk dataset" (328 cases). Deep learning was compared to the use of features from RTA, as well as to a combination of both in the task of distinguishing between high- and low-risk subjects. Similar classification performances were obtained using CNN [area under the curve [Formula: see text]; standard error [Formula: see text

Using Machine Learning To Predict Which Light Curves Will Yield Stellar Rotation Periods

NASA Astrophysics Data System (ADS)

Agüeros, Marcel; Teachey, Alexander

2018-01-01

Using time-domain photometry to reliably measure a solar-type star's rotation period requires that its light curve have a number of favorable characteristics. The probability of recovering a period will be a non-linear function of these light curve features, which are either astrophysical in nature or set by the observations. We employ standard machine learning algorithms (artificial neural networks and random forests) to predict whether a given light curve will produce a robust rotation period measurement from its Lomb-Scargle periodogram. The algorithms are trained and validated using salient statistics extracted from both simulated light curves and their corresponding periodograms, and we apply these classifiers to the most recent Intermediate Palomar Transient Factory (iPTF) data release. With this pipeline, we anticipate measuring rotation periods for a significant fraction of the ∼4x108 stars in the iPTF footprint.

Why the Kantian ideal survives medical learning curves, and why it matters.

PubMed

Brecher, B

2006-09-01

The "Kantian ideal" is often misunderstood as invoking individual autonomy rather than rational self legislation. Le Morvan and Stock's otherwise insightful discussion of "Medical learning curves and the Kantian ideal"--for example--draws the mistaken inference that that ideal is inconsistent with the realities of medical practice. But it is not. Rationally to be a patient entails accepting its necessary conditions.

A Method for Writing Open-Ended Curved Arrow Notation Questions for Multiple-Choice Exams and Electronic-Response Systems

ERIC Educational Resources Information Center

Ruder, Suzanne M.; Straumanis, Andrei R.

2009-01-01

A critical stage in the process of developing a conceptual understanding of organic chemistry is learning to use curved arrow notation. From this stems the ability to predict reaction products and mechanisms beyond the realm of memorization. Since evaluation (i.e., testing) is known to be a key driver of student learning, it follows that a new…

DeepInfer: open-source deep learning deployment toolkit for image-guided therapy

NASA Astrophysics Data System (ADS)

Mehrtash, Alireza; Pesteie, Mehran; Hetherington, Jorden; Behringer, Peter A.; Kapur, Tina; Wells, William M.; Rohling, Robert; Fedorov, Andriy; Abolmaesumi, Purang

2017-03-01

Deep learning models have outperformed some of the previous state-of-the-art approaches in medical image analysis. Instead of using hand-engineered features, deep models attempt to automatically extract hierarchical representations at multiple levels of abstraction from the data. Therefore, deep models are usually considered to be more flexible and robust solutions for image analysis problems compared to conventional computer vision models. They have demonstrated significant improvements in computer-aided diagnosis and automatic medical image analysis applied to such tasks as image segmentation, classification and registration. However, deploying deep learning models often has a steep learning curve and requires detailed knowledge of various software packages. Thus, many deep models have not been integrated into the clinical research work ows causing a gap between the state-of-the-art machine learning in medical applications and evaluation in clinical research procedures. In this paper, we propose "DeepInfer" - an open-source toolkit for developing and deploying deep learning models within the 3D Slicer medical image analysis platform. Utilizing a repository of task-specific models, DeepInfer allows clinical researchers and biomedical engineers to deploy a trained model selected from the public registry, and apply it to new data without the need for software development or configuration. As two practical use cases, we demonstrate the application of DeepInfer in prostate segmentation for targeted MRI-guided biopsy and identification of the target plane in 3D ultrasound for spinal injections.

DeepInfer: Open-Source Deep Learning Deployment Toolkit for Image-Guided Therapy.

PubMed

Mehrtash, Alireza; Pesteie, Mehran; Hetherington, Jorden; Behringer, Peter A; Kapur, Tina; Wells, William M; Rohling, Robert; Fedorov, Andriy; Abolmaesumi, Purang

2017-02-11

Deep learning models have outperformed some of the previous state-of-the-art approaches in medical image analysis. Instead of using hand-engineered features, deep models attempt to automatically extract hierarchical representations at multiple levels of abstraction from the data. Therefore, deep models are usually considered to be more flexible and robust solutions for image analysis problems compared to conventional computer vision models. They have demonstrated significant improvements in computer-aided diagnosis and automatic medical image analysis applied to such tasks as image segmentation, classification and registration. However, deploying deep learning models often has a steep learning curve and requires detailed knowledge of various software packages. Thus, many deep models have not been integrated into the clinical research workflows causing a gap between the state-of-the-art machine learning in medical applications and evaluation in clinical research procedures. In this paper, we propose "DeepInfer" - an open-source toolkit for developing and deploying deep learning models within the 3D Slicer medical image analysis platform. Utilizing a repository of task-specific models, DeepInfer allows clinical researchers and biomedical engineers to deploy a trained model selected from the public registry, and apply it to new data without the need for software development or configuration. As two practical use cases, we demonstrate the application of DeepInfer in prostate segmentation for targeted MRI-guided biopsy and identification of the target plane in 3D ultrasound for spinal injections.

DeepInfer: Open-Source Deep Learning Deployment Toolkit for Image-Guided Therapy

PubMed Central

Mehrtash, Alireza; Pesteie, Mehran; Hetherington, Jorden; Behringer, Peter A.; Kapur, Tina; Wells, William M.; Rohling, Robert; Fedorov, Andriy; Abolmaesumi, Purang

2017-01-01

Deep learning models have outperformed some of the previous state-of-the-art approaches in medical image analysis. Instead of using hand-engineered features, deep models attempt to automatically extract hierarchical representations at multiple levels of abstraction from the data. Therefore, deep models are usually considered to be more flexible and robust solutions for image analysis problems compared to conventional computer vision models. They have demonstrated significant improvements in computer-aided diagnosis and automatic medical image analysis applied to such tasks as image segmentation, classification and registration. However, deploying deep learning models often has a steep learning curve and requires detailed knowledge of various software packages. Thus, many deep models have not been integrated into the clinical research workflows causing a gap between the state-of-the-art machine learning in medical applications and evaluation in clinical research procedures. In this paper, we propose “DeepInfer” – an open-source toolkit for developing and deploying deep learning models within the 3D Slicer medical image analysis platform. Utilizing a repository of task-specific models, DeepInfer allows clinical researchers and biomedical engineers to deploy a trained model selected from the public registry, and apply it to new data without the need for software development or configuration. As two practical use cases, we demonstrate the application of DeepInfer in prostate segmentation for targeted MRI-guided biopsy and identification of the target plane in 3D ultrasound for spinal injections. PMID:28615794

Safety, efficiency and learning curves in robotic surgery: a human factors analysis.

PubMed

Catchpole, Ken; Perkins, Colby; Bresee, Catherine; Solnik, M Jonathon; Sherman, Benjamin; Fritch, John; Gross, Bruno; Jagannathan, Samantha; Hakami-Majd, Niv; Avenido, Raymund; Anger, Jennifer T

2016-09-01

Expense, efficiency of use, learning curves, workflow integration and an increased prevalence of serious incidents can all be barriers to adoption. We explored an observational approach and initial diagnostics to enhance total system performance in robotic surgery. Eighty-nine robotic surgical cases were observed in multiple operating rooms using two different surgical robots (the S and Si), across several specialties (Urology, Gynecology, and Cardiac Surgery). The main measures were operative duration and rate of flow disruptions-described as 'deviations from the natural progression of an operation thereby potentially compromising safety or efficiency.' Contextual parameters collected were surgeon experience level and training, type of surgery, the model of robot and patient factors. Observations were conducted across four operative phases (operating room pre-incision; robot docking; main surgical intervention; post-console). A mean of 9.62 flow disruptions per hour (95 % CI 8.78-10.46) were predominantly caused by coordination, communication, equipment and training problems. Operative duration and flow disruption rate varied with surgeon experience (p = 0.039; p < 0.001, respectively), training cases (p = 0.012; p = 0.007) and surgical type (both p < 0.001). Flow disruption rates in some phases were also sensitive to the robot model and patient characteristics. Flow disruption rate is sensitive to system context and generates improvement diagnostics. Complex surgical robotic equipment increases opportunities for technological failures, increases communication requirements for the whole team, and can reduce the ability to maintain vision in the operative field. These data suggest specific opportunities to reduce the training costs and the learning curve.

Evaluating the learning curve for robot-assisted laparoscopic radical cystectomy.

PubMed

Pruthi, Raj S; Smith, Angela; Wallen, Eric M

2008-11-01

We seek to describe the learning curve of robot-assisted laparoscopic radical cystectomy by evaluating some of the surgical, oncologic, and clinical outcomes in our initial experience with 50 consecutive patients undergoing this novel procedure. Fifty consecutive patients (representing our initial experience with robot-assisted cystectomy) underwent radical cystectomy and urinary diversion from January 2006 to December 2007. Several different metrics were used to evaluate the learning curve of this procedure, including estimated blood loss (EBL), operative (OR) time, pathologic outcomes, and complication rate. We evaluated patients as a continuous variable, divided into five distinct time periods (quintiles), and stratified by first half and second half of robotic experience. EBL was not significantly lower until the third quintile (patients 21-30), after which further significant reductions were not observed. Mean OR time declined between each quintile for the first 30 patients (1-10 v 11-20 v 21-30). No significant declines occurred after the third quintile (21-30). When evaluated as a continuous variable, the statistical cut point at which no further significant reductions were observed was after patient 20 for OR time. No differences were observed with regard to time to flatus, bowel movement, or hospital discharge. Furthermore, complications were not different between the initial 25 patients and the most recent patients. There has been no case of a positive margin, and there was only one inadvertent bladder entry. Lymph node yield has also not significantly changed over time. This report helps to define the learning curve associated with robot-assisted laparoscopic radical cystectomy for bladder cancer. Despite the higher OR times and blood loss that is observed early in the learning curve, no such compromises are observed with regard to these oncologic parameters even early in the experience.

Robotic Mitral Valve Repair: The Learning Curve.

PubMed

Goodman, Avi; Koprivanac, Marijan; Kelava, Marta; Mick, Stephanie L; Gillinov, A Marc; Rajeswaran, Jeevanantham; Brzezinski, Anna; Blackstone, Eugene H; Mihaljevic, Tomislav

Adoption of robotic mitral valve surgery has been slow, likely in part because of its perceived technical complexity and a poorly understood learning curve. We sought to correlate changes in technical performance and outcome with surgeon experience in the "learning curve" part of our series. From 2006 to 2011, two surgeons undertook robotically assisted mitral valve repair in 458 patients (intent-to-treat); 404 procedures were completed entirely robotically (as-treated). Learning curves were constructed by modeling surgical sequence number semiparametrically with flexible penalized spline smoothing best-fit curves. Operative efficiency, reflecting technical performance, improved for (1) operating room time for case 1 to cases 200 (early experience) and 400 (later experience), from 414 to 364 to 321 minutes (12% and 22% decrease, respectively), (2) cardiopulmonary bypass time, from 148 to 102 to 91 minutes (31% and 39% decrease), and (3) myocardial ischemic time, from 119 to 75 to 68 minutes (37% and 43% decrease). Composite postoperative complications, reflecting safety, decreased from 17% to 6% to 2% (63% and 85% decrease). Intensive care unit stay decreased from 32 to 28 to 24 hours (13% and 25% decrease). Postoperative stay fell from 5.2 to 4.5 to 3.8 days (13% and 27% decrease). There were no in-hospital deaths. Predischarge mitral regurgitation of less than 2+, reflecting effectiveness, was achieved in 395 (97.8%), without correlation to experience; return-to-work times did not change substantially with experience. Technical efficiency of robotic mitral valve repair improves with experience and permits its safe and effective conduct.

A Machine-Learning Algorithm Toward Color Analysis for Chronic Liver Disease Classification, Employing Ultrasound Shear Wave Elastography.

PubMed

Gatos, Ilias; Tsantis, Stavros; Spiliopoulos, Stavros; Karnabatidis, Dimitris; Theotokas, Ioannis; Zoumpoulis, Pavlos; Loupas, Thanasis; Hazle, John D; Kagadis, George C

2017-09-01

The purpose of the present study was to employ a computer-aided diagnosis system that classifies chronic liver disease (CLD) using ultrasound shear wave elastography (SWE) imaging, with a stiffness value-clustering and machine-learning algorithm. A clinical data set of 126 patients (56 healthy controls, 70 with CLD) was analyzed. First, an RGB-to-stiffness inverse mapping technique was employed. A five-cluster segmentation was then performed associating corresponding different-color regions with certain stiffness value ranges acquired from the SWE manufacturer-provided color bar. Subsequently, 35 features (7 for each cluster), indicative of physical characteristics existing within the SWE image, were extracted. A stepwise regression analysis toward feature reduction was used to derive a reduced feature subset that was fed into the support vector machine classification algorithm to classify CLD from healthy cases. The highest accuracy in classification of healthy to CLD subject discrimination from the support vector machine model was 87.3% with sensitivity and specificity values of 93.5% and 81.2%, respectively. Receiver operating characteristic curve analysis gave an area under the curve value of 0.87 (confidence interval: 0.77-0.92). A machine-learning algorithm that quantifies color information in terms of stiffness values from SWE images and discriminates CLD from healthy cases is introduced. New objective parameters and criteria for CLD diagnosis employing SWE images provided by the present study can be considered an important step toward color-based interpretation, and could assist radiologists' diagnostic performance on a daily basis after being installed in a PC and employed retrospectively, immediately after the examination. Copyright © 2017 World Federation for Ultrasound in Medicine & Biology. Published by Elsevier Inc. All rights reserved.

A specific implicit sequence learning deficit as an underlying cause of dyslexia? Investigating the role of attention in implicit learning tasks.

PubMed

Staels, Eva; Van den Broeck, Wim

2017-05-01

Recently, a general implicit sequence learning deficit was proposed as an underlying cause of dyslexia. This new hypothesis was investigated in the present study by including a number of methodological improvements, for example, the inclusion of appropriate control conditions. The second goal of the study was to explore the role of attentional functioning in implicit and explicit learning tasks. In a 2 × 2 within-subjects design 4 tasks were administered in 30 dyslexic and 38 control children: an implicit and explicit serial reaction time (RT) task and an implicit and explicit contextual cueing task. Attentional functioning was also administered. The entire learning curves of all tasks were analyzed using latent growth curve modeling in order to compare performances between groups and to examine the role of attentional functioning on the learning curves. The amount of implicit learning was similar for both groups. However, the dyslexic group showed slower RTs throughout the entire task. This group difference reduced and became nonsignificant after controlling for attentional functioning. Both implicit learning tasks, but none of the explicit learning tasks, were significantly affected by attentional functioning. Dyslexic children do not suffer from a specific implicit sequence learning deficit. The slower RTs of the dyslexic children throughout the entire implicit sequence learning process are caused by their comorbid attention problems and overall slowness. A key finding of the present study is that, in contrast to what was assumed for a long time, implicit learning relies on attentional resources, perhaps even more than explicit learning does. (PsycINFO Database Record (c) 2017 APA, all rights reserved).

Genotype-environment interaction in passive avoidance learning of the paradise fish (Macropodus opercularis).

PubMed

Csányi, V; Gervai, J

1985-01-01

Passive dark avoidance conditioning and effects of the presence and absence of a fish-like dummy on the training process were studied in four inbred strains of paradise fish. Strain differences were found in the shuttle activity during habituation trials, and in the sensitivity to the mild electric shock punishment. The presence or absence of the dummy in the punished dark side of the shuttle box had a genotype-dependent effect on the measures taken during the conditioning process. The statistical analysis of the learning curves revealed differences in the way the strains varied in the different environments, i.e. genotype--environment interaction components of variances were identified. The results are discussed in the light of previous investigations and their implication in further genetic analysis.

Comparison of the learning curves of digital examination and transabdominal sonography for the determination of fetal head position during labor.

PubMed

Rozenberg, P; Porcher, R; Salomon, L J; Boirot, F; Morin, C; Ville, Y

2008-03-01

To evaluate the learning curve of transabdominal sonography for the determination of fetal head position in labor and to compare it with that of digital vaginal examination. A student midwife who had never performed digital vaginal examination or ultrasound examination was recruited for this study. Instructions on how to perform digital vaginal examination and ultrasound examination were given before and after completing the first vaginal and ultrasound examinations, and repeated for each subsequent examination for as long as necessary. Digital and ultrasound diagnoses of the fetal head position were always performed first by the student midwife, and repeated by an experienced midwife or physician. The learning curve for identification of the fetal head position by either one of the two methods was analyzed using the cumulative sums (CUSUM) method for measurement errors. One hundred patients underwent digital vaginal examination and 99 had transabdominal sonography for the determination of fetal head position. An error rate of around 50% for vaginal examination was nearly constant during the first 50 examinations. It decreased subsequently, to stabilize at a low level from the 82(nd) patient. Errors of +/- 180 degrees were the most frequent. The learning curve for ultrasound imaging stabilized earlier than that of vaginal examination, after the 32(nd) patient. The most frequent errors with ultrasound examination were the inability to conclude on a diagnosis, particularly at the beginning of training, followed by errors of +/- 45 degrees. Based on our findings for the student tested, learning and accuracy of the determination of fetal head position in labor were easier and higher, respectively, with transabdominal sonography than with digital examination. This should encourage physicians to introduce clinical ultrasound examination into their practice. CUSUM charts provide a reliable representation of the learning curve, by accumulating evidence of performance. Copyright (c) 2008 ISUOG. Published by John Wiley & Sons, Ltd.

Time-Decayed User Profile for Second Language Vocabulary Learning System

ERIC Educational Resources Information Center

Li, Li; Wei, Xiao

2014-01-01

Vocabulary learning is the foundation of second language learning. Many E-learning systems have been developed to help learners to learn vocabulary efficiently. Most of these systems employ Ebbinghaus Forgetting Curve to make the review schedule for learners. However, learners are different in learning ability and the review schedule based on…

Why the Kantian ideal survives medical learning curves, and why it matters

PubMed Central

Brecher, B

2006-01-01

The “Kantian ideal” is often misunderstood as invoking individual autonomy rather than rational self legislation. Le Morvan and Stock's otherwise insightful discussion of “Medical learning curves and the Kantian ideal”—for example—draws the mistaken inference that that ideal is inconsistent with the realities of medical practice. But it is not. Rationally to be a patient entails accepting its necessary conditions. PMID:16943330

The learning curve of laparoendoscopic single-Site (LESS) fundoplication: definable, short, and safe.

PubMed

Ross, Sharona B; Choung, Edward; Teta, Anthony F; Colibao, Lotiffa; Luberice, Kenneth; Paul, Harold; Rosemurgy, Alexander S

2013-01-01

This study of laparoendoscopic single-site (LESS) fundoplication for gastroesophageal reflux disease was undertaken to determine the "learning curve" for implementing LESS fundoplication. One hundred patients, 38% men, with a median age of 61 years and median body mass index of 26 kg/m(2) , underwent LESS fundoplications. The operative times, placement of additional trocars, conversions to "open" operations, and complications were compared among patient quartiles to establish a learning curve. Median data are reported. The median operative times and complications did not differ among 25-patient cohorts. Additional trocars were placed in 27% of patients, 67% of whom were in the first 25-patient cohort. Patients undergoing LESS fundoplication had a dramatic relief in the frequency and severity of all symptoms of reflux across all cohorts equally (P < .05), particularly for heartburn and regurgitation, without causing dysphagia. LESS fundoplication ameliorates symptoms of gastroesophageal reflux disease without apparent scarring. Notably, few operations required additional trocars after the first 25-patient cohort. Patient selection became more inclusive (eg, more "redo" fundoplications) with increasing experience, whereas operative times and complications remained relatively unchanged. The learning curve of LESS fundoplication is definable, short, and safe. We believe that patients will seek LESS fundoplication because of the efficacy and superior cosmetic outcomes; surgeons will need to meet this demand.

Prognostication of Learning Curve on Surgical Management of Vasculobiliary Injuries after Cholecystectomy

PubMed Central

Dar, Faisal Saud; Zia, Haseeb; Rafique, Muhammad Salman; Khan, Nusrat Yar; Salih, Mohammad; Hassan Shah, Najmul

2016-01-01

Background. Concomitant vascular injury might adversely impact outcomes after iatrogenic bile duct injury (IBDI). Whether a new HPB center should embark upon repair of complex biliary injuries with associated vascular injuries during learning curve is unknown. The objective of this study was to determine outcome of surgical management of IBDI with and without vascular injuries in a new HPB center during its learning curve. Methods. We retrospectively reviewed patients who underwent surgical management of IBDI at our center. A total of 39 patients were included. Patients without (Group 1) and with vascular injuries (Group 2) were compared. Outcome was defined as 90-day morbidity and mortality. Results. Median age was 39 (20–80) years. There were 10 (25.6%) vascular injuries. E2 injuries were associated significantly with high frequency of vascular injuries (66% versus 15.1%) (P = 0.01). Right hepatectomy was performed in three patients. Out of these, two had a right hepatic duct stricture and one patient had combined right arterial and portal venous injury. The number of patients who developed postoperative complications was not significantly different between the two groups (11.1% versus 23.4%) (P = 0.6). Conclusion. Learning curve is not a negative prognostic variable in the surgical management of iatrogenic vasculobiliary injuries after cholecystectomy. PMID:27525124

Insights from Hydrogen Refueling Station Manufacturing Competitiveness Analysis

DOE Office of Scientific and Technical Information (OSTI.GOV)

Mayyas, Ahmad

2015-12-18

In work for the Clean Energy Manufacturing Analysis Center (CEMAC), NREL is currently collaborating with Great Lakes Wind Network in conducting a comprehensive hydrogen refueling stations manufacturing competitiveness and supply chain analyses. In this project, CEMAC will be looking at several metrics that will facilitate understanding of the interactions between and within the HRS supply chain, such metrics include innovation potential, intellectual properties, learning curves, related industries and clustering, existing supply chains, ease of doing business, and regulations and safety. This presentation to Fuel Cell Seminar and Energy Exposition 2015 highlights initial findings from CEMAC's analysis.

Rapid learning dynamics in individual honeybees during classical conditioning.

PubMed

Pamir, Evren; Szyszka, Paul; Scheiner, Ricarda; Nawrot, Martin P

2014-01-01

Associative learning in insects has been studied extensively by a multitude of classical conditioning protocols. However, so far little emphasis has been put on the dynamics of learning in individuals. The honeybee is a well-established animal model for learning and memory. We here studied associative learning as expressed in individual behavior based on a large collection of data on olfactory classical conditioning (25 datasets, 3298 animals). We show that the group-averaged learning curve and memory retention score confound three attributes of individual learning: the ability or inability to learn a given task, the generally fast acquisition of a conditioned response (CR) in learners, and the high stability of the CR during consecutive training and memory retention trials. We reassessed the prevailing view that more training results in better memory performance and found that 24 h memory retention can be indistinguishable after single-trial and multiple-trial conditioning in individuals. We explain how inter-individual differences in learning can be accommodated within the Rescorla-Wagner theory of associative learning. In both data-analysis and modeling we demonstrate how the conflict between population-level and single-animal perspectives on learning and memory can be disentangled.

Rapid learning dynamics in individual honeybees during classical conditioning

PubMed Central

Pamir, Evren; Szyszka, Paul; Scheiner, Ricarda; Nawrot, Martin P.

2014-01-01

Associative learning in insects has been studied extensively by a multitude of classical conditioning protocols. However, so far little emphasis has been put on the dynamics of learning in individuals. The honeybee is a well-established animal model for learning and memory. We here studied associative learning as expressed in individual behavior based on a large collection of data on olfactory classical conditioning (25 datasets, 3298 animals). We show that the group-averaged learning curve and memory retention score confound three attributes of individual learning: the ability or inability to learn a given task, the generally fast acquisition of a conditioned response (CR) in learners, and the high stability of the CR during consecutive training and memory retention trials. We reassessed the prevailing view that more training results in better memory performance and found that 24 h memory retention can be indistinguishable after single-trial and multiple-trial conditioning in individuals. We explain how inter-individual differences in learning can be accommodated within the Rescorla–Wagner theory of associative learning. In both data-analysis and modeling we demonstrate how the conflict between population-level and single-animal perspectives on learning and memory can be disentangled. PMID:25309366

The Analysis of Riboflavin in Urine Using Fluorescence

NASA Astrophysics Data System (ADS)

Henderleiter, Julie A.; Hyslop, Richard M.

1996-06-01

To become functional as scientists, chemistry students must integrate concepts learned in their classes and apply them to novel, "real life" situations. The laboratory provides an important place for the students to practice integrating concepts. This laboratory experiment, designed for undergraduate biochemistry students, requires each student to determine the amount of riboflavin excreted by his/her body following oral administration of riboflavin contained in a multi-vitamin tablet. The experimental procedure describes a protocol for the analysis of riboflavin concentration in urine using a fluorometric assay. The students must draw upon their knowledge of solution preparation, construction of a standard curve, and back-calculation procedures to determine the concentration of riboflavin in their urine. Students need to combine knowledge from general and analytical chemistry with that learned in biochemistry to complete this analysis, thus providing an opportunity to integrate knowledge while answering a novel question.

Co-registration of cone beam CT and preoperative MRI for improved accuracy of electrode localization following cochlear implantation.

PubMed

Dragovic, A S; Stringer, A K; Campbell, L; Shaul, C; O'Leary, S J; Briggs, R J

2018-05-01

To investigate the clinical usefulness and practicality of co-registration of Cone Beam CT (CBCT) with preoperative Magnetic Resonance Imaging (MRI) for intracochlear localization of electrodes after cochlear implantation. Images of 20 adult patients who underwent CBCT after implantation were co-registered with preoperative MRI scans. Time taken for co-registration was recorded. The images were analysed by clinicians of varying levels of expertise to determine electrode position and ease of interpretation. After a short learning curve, the average co-registration time was 10.78 minutes (StdDev 2.37). All clinicians found the co-registered images easier to interpret than CBCT alone. The mean concordance of CBCT vs. co-registered image analysis between consultant otologists was 60% (17-100%) and 86% (60-100%), respectively. The sensitivity and specificity for CBCT to identify Scala Vestibuli insertion or translocation was 100 and 75%, respectively. The negative predictive value was 100%. CBCT should be performed following adult cochlear implantation for audit and quality control of surgical technique. If SV insertion or translocation is suspected, co-registration with preoperative MRI should be performed to enable easier analysis. There will be a learning curve for this process in terms of both the co-registration and the interpretation of images by clinicians.

Learning curve for robotic-assisted laparoscopic rectal cancer surgery.

PubMed

Jiménez-Rodríguez, Rosa M; Díaz-Pavón, José Manuel; de la Portilla de Juan, Fernando; Prendes-Sillero, Emilio; Dussort, Hisnard Cadet; Padillo, Javier

2013-06-01

One of the main uses of robotic assisted abdominal surgery is the mesorectal excision in patients with rectal cancer. The aim of the present study is to analyse the learning curve for robotic assisted laparoscopic resection of rectal cancer. We included in our study 43 consecutive rectal cancer resections (16 females and 27 males) performed from January 2008 through December 2010. Mean age of patients was 66 ± 9.0 years. Surgical procedures included both abdomino-perineal and anterior resections. We analysed the following parameters: demographic data of the patients included in the study, intra- and postoperative data, time taking to set up the robot for operations (set-up or docking time), operative time, intra- and postoperative complications, conversion rates and pathological specimen features. The learning curve was analysed using cumulative sum (CUSUM) methodology. The procedures understudied included seven abdomino-perineal resections and 36 anterior resections. In our series of patients, mean robotic set-up time was 62.9 ± 24.6 min, and the mean operative time was 197.4 ± 44.3 min. Once we applied CUSUM methodology, we obtained two graphs for CUSUM values (operating time and success), both of them showing three well-differentiated phases: phase 1 (the initial 9-11 cases), phase 2 (the middle 12 cases) and phase 3 (the remaining 20-22 cases). Phase 1 represents initial learning; phase 2 plateau represents increased competence in the use of the robotic system, and finally, phase 3 represents the period of highest skill or mastery with a reduction in docking time (p = 0.000), but a slight increase in operative time (p = 0.007). The CUSUM curve shows three phases in the learning and use of robotic assisted rectal cancer surgery which correspond to the phases of initial learning of the technique, consolidation and higher expertise or mastery. The data obtained suggest that the estimated learning curve for robotic assisted rectal cancer surgery is achieved after 21-23 cases.

Does my high blood pressure improve your survival? Overall and subgroup learning curves in health.

PubMed

Van Gestel, Raf; Müller, Tobias; Bosmans, Johan

2017-09-01

Learning curves in health are of interest for a wide range of medical disciplines, healthcare providers, and policy makers. In this paper, we distinguish between three types of learning when identifying overall learning curves: economies of scale, learning from cumulative experience, and human capital depreciation. In addition, we approach the question of how treating more patients with specific characteristics predicts provider performance. To soften collinearity problems, we explore the use of least absolute shrinkage and selection operator regression as a variable selection method and Theil-Goldberger mixed estimation to augment the available information. We use data from the Belgian Transcatheter Aorta Valve Implantation (TAVI) registry, containing information on the first 860 TAVI procedures in Belgium. We find that treating an additional TAVI patient is associated with an increase in the probability of 2-year survival by about 0.16%-points. For adverse events like renal failure and stroke, we find that an extra day between procedures is associated with an increase in the probability for these events by 0.12%-points and 0.07%-points, respectively. Furthermore, we find evidence for positive learning effects from physicians' experience with defibrillation, treating patients with hypertension, and the use of certain types of replacement valves during the TAVI procedure. Copyright © 2017 John Wiley & Sons, Ltd.

Effects of antiepileptic drugs on learning as assessed by a repeated acquisition of response sequences task in rats.

PubMed

Shannon, Harlan E; Love, Patrick L

2007-02-01

Patients with epilepsy can have impaired cognitive abilities. Antiepileptic drugs (AEDs) may contribute to the cognitive deficits observed in patients with epilepsy, and have been shown to induce cognitive impairments in healthy individuals. However, there are few systematic data on the effects of AEDs on specific cognitive domains. We have previously demonstrated that a number of AEDs can impair working memory and attention. The purpose of the present study was to evaluate the effects of AEDs on learning as measured by a repeated acquisition of response sequences task in nonepileptic rats. The GABA-related AEDs phenobarbital and chlordiazepoxide significantly disrupted performance by shifting the learning curve to the right and increasing errors, whereas tiagabine and valproate did not. The sodium channel blockers carbamazepine and phenytoin suppressed responding at higher doses, whereas lamotrigine shifted the learning curve to the right and increased errors, and topiramate was without significant effect. Levetiracetam also shifted the learning curve to the right and increased errors. The disruptions produced by triazolam, chlordiazepoxide, lamotrigine, and levetiracetam were qualitatively similar to the effects of the muscarinic cholinergic receptor antagonist scopolamine. The present results indicate that AEDs can impair learning, but there are differences among AEDs in the magnitude of the disruption in nonepileptic rats, with drugs that enhance GABA receptor function and some that block sodium channels producing the most consistent impairment of learning.

Machine learning algorithms for mode-of-action classification in toxicity assessment.

PubMed

Zhang, Yile; Wong, Yau Shu; Deng, Jian; Anton, Cristina; Gabos, Stephan; Zhang, Weiping; Huang, Dorothy Yu; Jin, Can

2016-01-01

Real Time Cell Analysis (RTCA) technology is used to monitor cellular changes continuously over the entire exposure period. Combining with different testing concentrations, the profiles have potential in probing the mode of action (MOA) of the testing substances. In this paper, we present machine learning approaches for MOA assessment. Computational tools based on artificial neural network (ANN) and support vector machine (SVM) are developed to analyze the time-concentration response curves (TCRCs) of human cell lines responding to tested chemicals. The techniques are capable of learning data from given TCRCs with known MOA information and then making MOA classification for the unknown toxicity. A novel data processing step based on wavelet transform is introduced to extract important features from the original TCRC data. From the dose response curves, time interval leading to higher classification success rate can be selected as input to enhance the performance of the machine learning algorithm. This is particularly helpful when handling cases with limited and imbalanced data. The validation of the proposed method is demonstrated by the supervised learning algorithm applied to the exposure data of HepG2 cell line to 63 chemicals with 11 concentrations in each test case. Classification success rate in the range of 85 to 95 % are obtained using SVM for MOA classification with two clusters to cases up to four clusters. Wavelet transform is capable of capturing important features of TCRCs for MOA classification. The proposed SVM scheme incorporated with wavelet transform has a great potential for large scale MOA classification and high-through output chemical screening.

Descemet Membrane Endothelial Keratoplasty Learning Curve for Graft Preparation in an Eye Bank Using 645 Donor Corneas.

PubMed

Parekh, Mohit; Ruzza, Alessandro; Romano, Vito; Favaro, Elisa; Baruzzo, Mattia; Salvalaio, Gianni; Grassetto, Andrea; Ferrari, Stefano; Ponzin, Diego

2018-06-01

To investigate the learning curve of Descemet membrane endothelial keratoplasty (DMEK) graft preparation in an eye bank. Four operators prepared 645 DMEK grafts using the stripping technique between 2014 and 2017 at the Veneto Eye Bank Foundation, Italy. Endothelial cell loss (ECL) and tissue wastage were recorded retrospectively after DMEK preparation and correlated with the number of tissues prepared each year by each operator. On average, our operators performed 1 donor preparation a week over the course of this study. Only donors older than 60 years were used in this study, and approximately 10% of donors had diabetes. The Wilcoxon test for paired data and 1-way ANOVA were used for checking statistical significance with the Tukey test as post hoc analysis. P < 0.05 was considered statistically significant. ECL did not change significantly over time from operator 1. Significant ECL drop was noted from operator 2 between years 2014-2016 (P = 0.0049) and 2017 (P = 0.0094); from operator 3 between years 2015-2016 (P = 0.0288) and 2017 (P = 0.0097); and from operator 4 between 2015-2016 (P = 0.0469) and 2017 (P = 0.0331). Operators 1 and 3 did not show a significant difference, considering every 50 grafts prepared by each operator. Operator 2 showed significant ECL drop between 1 to 50 and 51 to 100 (P = 0.0002) and 1 to 50 and 101 to 150 (P = 0.0001) grafts. Operator 4 showed significant ECL drop between 1 to 50 and 101 to 150 (P = 0.002) and 51 to 100 and 101 to 141 (P = 0.0207) grafts. No intraoperator difference was observed per 50 grafts (P > 0.05). There is a learning curve for DMEK graft preparation. ECL and tissue wastage can be reduced with practice and skills. However, each operator may be limited to his or her own learning capability.

Choosing the Optimal Number of B-spline Control Points (Part 1: Methodology and Approximation of Curves)

NASA Astrophysics Data System (ADS)

Harmening, Corinna; Neuner, Hans

2016-09-01

Due to the establishment of terrestrial laser scanner, the analysis strategies in engineering geodesy change from pointwise approaches to areal ones. These areal analysis strategies are commonly built on the modelling of the acquired point clouds. Freeform curves and surfaces like B-spline curves/surfaces are one possible approach to obtain space continuous information. A variety of parameters determines the B-spline's appearance; the B-spline's complexity is mostly determined by the number of control points. Usually, this number of control points is chosen quite arbitrarily by intuitive trial-and-error-procedures. In this paper, the Akaike Information Criterion and the Bayesian Information Criterion are investigated with regard to a justified and reproducible choice of the optimal number of control points of B-spline curves. Additionally, we develop a method which is based on the structural risk minimization of the statistical learning theory. Unlike the Akaike and the Bayesian Information Criteria this method doesn't use the number of parameters as complexity measure of the approximating functions but their Vapnik-Chervonenkis-dimension. Furthermore, it is also valid for non-linear models. Thus, the three methods differ in their target function to be minimized and consequently in their definition of optimality. The present paper will be continued by a second paper dealing with the choice of the optimal number of control points of B-spline surfaces.

Assessing the learning curve for the acquisition of laparoscopic skills on a virtual reality simulator.

PubMed

Sherman, V; Feldman, L S; Stanbridge, D; Kazmi, R; Fried, G M

2005-05-01

The aim of this study was to develop summary metrics and assess the construct validity for a virtual reality laparoscopic simulator (LapSim) by comparing the learning curves of three groups with different levels of laparoscopic expertise. Three groups of subjects ('expert', 'junior', and 'naïve') underwent repeated trials on three LapSim tasks. Formulas were developed to calculate scores for efficiency ('time-error') and economy of 'motion' ('motion') using metrics generated by the software after each drill. Data (mean +/- SD) were evaluated by analysis of variance (ANOVA). Significance was set at p < 0.05. All three groups improved significantly from baseline to final for both 'time-error' and 'motion' scores. There were significant differences between groups in time error performances at baseline and final, due to higher scores in the 'expert' group. A significant difference in 'motion' scores was seen only at baseline. We have developed summary metrics for the LapSim that differentiate among levels of laparoscopic experience. This study also provides evidence of construct validity for the LapSim.

Stimulating investment in energy materials and technologies to combat climate change: an overview of learning curve analysis and niche market support.

PubMed

Foxon, Timothy J

2010-07-28

This paper addresses the probable levels of investment needed in new technologies for energy conversion and storage that are essential to address climate change, drawing on past evidence on the rate of cost improvements in energy technologies. A range of energy materials and technologies with lower carbon emissions over their life cycle are being developed, including fuel cells (FCs), hydrogen storage, batteries, supercapacitors, solar energy and nuclear power, and it is probable that most, if not all, of these technologies will be needed to mitigate climate change. High rates of innovation and deployment will be needed to meet targets such as the UK's goal of reducing its greenhouse gas emissions by 80 per cent by 2050, which will require significant levels of investment. Learning curves observed for reductions in unit costs of energy technologies, such as photovoltaics and FCs, can provide evidence on the probable future levels of investment needed. The paper concludes by making recommendations for policy measures to promote such investment from both the public and private sectors.

Exploring the spatio-temporal neural basis of face learning

PubMed Central

Yang, Ying; Xu, Yang; Jew, Carol A.; Pyles, John A.; Kass, Robert E.; Tarr, Michael J.

2017-01-01

Humans are experts at face individuation. Although previous work has identified a network of face-sensitive regions and some of the temporal signatures of face processing, as yet, we do not have a clear understanding of how such face-sensitive regions support learning at different time points. To study the joint spatio-temporal neural basis of face learning, we trained subjects to categorize two groups of novel faces and recorded their neural responses using magnetoencephalography (MEG) throughout learning. A regression analysis of neural responses in face-sensitive regions against behavioral learning curves revealed significant correlations with learning in the majority of the face-sensitive regions in the face network, mostly between 150–250 ms, but also after 300 ms. However, the effect was smaller in nonventral regions (within the superior temporal areas and prefrontal cortex) than that in the ventral regions (within the inferior occipital gyri (IOG), midfusiform gyri (mFUS) and anterior temporal lobes). A multivariate discriminant analysis also revealed that IOG and mFUS, which showed strong correlation effects with learning, exhibited significant discriminability between the two face categories at different time points both between 150–250 ms and after 300 ms. In contrast, the nonventral face-sensitive regions, where correlation effects with learning were smaller, did exhibit some significant discriminability, but mainly after 300 ms. In sum, our findings indicate that early and recurring temporal components arising from ventral face-sensitive regions are critically involved in learning new faces. PMID:28570739

Exploring the spatio-temporal neural basis of face learning.

PubMed

Yang, Ying; Xu, Yang; Jew, Carol A; Pyles, John A; Kass, Robert E; Tarr, Michael J

2017-06-01

Humans are experts at face individuation. Although previous work has identified a network of face-sensitive regions and some of the temporal signatures of face processing, as yet, we do not have a clear understanding of how such face-sensitive regions support learning at different time points. To study the joint spatio-temporal neural basis of face learning, we trained subjects to categorize two groups of novel faces and recorded their neural responses using magnetoencephalography (MEG) throughout learning. A regression analysis of neural responses in face-sensitive regions against behavioral learning curves revealed significant correlations with learning in the majority of the face-sensitive regions in the face network, mostly between 150-250 ms, but also after 300 ms. However, the effect was smaller in nonventral regions (within the superior temporal areas and prefrontal cortex) than that in the ventral regions (within the inferior occipital gyri (IOG), midfusiform gyri (mFUS) and anterior temporal lobes). A multivariate discriminant analysis also revealed that IOG and mFUS, which showed strong correlation effects with learning, exhibited significant discriminability between the two face categories at different time points both between 150-250 ms and after 300 ms. In contrast, the nonventral face-sensitive regions, where correlation effects with learning were smaller, did exhibit some significant discriminability, but mainly after 300 ms. In sum, our findings indicate that early and recurring temporal components arising from ventral face-sensitive regions are critically involved in learning new faces.

Clinical Outcomes and Complications during the Learning Curve for Reverse Total Shoulder Arthroplasty: An Analysis of the First 40 Cases

PubMed Central

Song, Kwang-Soon; Koo, Tae-Won

2017-01-01

Background The purpose of this study was to investigate the results and complications during the learning curve of reverse total shoulder arthroplasty (RTSA) for rotator cuff deficiency. Methods We retrospectively reviewed the first 40 cases of RTSA performed by a single surgeon. The mean age of patients was 72.7 years (range, 63 to 81 years) and mean follow-up period was 26.7 months (range, 9 to 57 months). Clinical outcomes were evaluated using a visual analog scale (VAS) for pain, the University of California at Los Angeles (UCLA) shoulder score, American Shoulder and Elbow Surgeon (ASES) score, subjective shoulder value (SSV), and active range of motion (ROM). Intraoperative and postoperative complications were also evaluated. Results The average VAS pain score, UCLA score, ASES score, and SSV improved from 6.9%, 12.8%, 29.0%, and 29.0% before surgery to 1.6%, 27.0%, 73.3%, and 71.5% after surgery, respectively (p < 0.001). The mean forward flexion, abduction, and external rotation improved from 68.0°, 56.9°, and 28.0° before surgery to 131.0°, 112.3°, and 38.8° after surgery, respectively (p < 0.001, p < 0.001, and p = 0.021). However, the mean internal rotation did not improve after surgery (p = 0.889). Scapular notching was observed in 33 patients (51.5%). Eight shoulders (20%) had complications, including 2 major (1 deep infection and 1 glenoid fixation failure) and 6 minor complications (3 brachial plexus injuries, 2 acromial fractures, and 1 intraoperative periprosthetic fracture). Conclusions The first 40 cases of RTSA performed by a single surgeon during the learning curve period showed satisfactory short-term follow-up results with an acceptable complication rate. PMID:28567225

Multilevel linear modelling of the response-contingent learning of young children with significant developmental delays.

PubMed

Raab, Melinda; Dunst, Carl J; Hamby, Deborah W

2018-02-27

The purpose of the study was to isolate the sources of variations in the rates of response-contingent learning among young children with multiple disabilities and significant developmental delays randomly assigned to contrasting types of early childhood intervention. Multilevel, hierarchical linear growth curve modelling was used to analyze four different measures of child response-contingent learning where repeated child learning measures were nested within individual children (Level-1), children were nested within practitioners (Level-2), and practitioners were nested within the contrasting types of intervention (Level-3). Findings showed that sources of variations in rates of child response-contingent learning were associated almost entirely with type of intervention after the variance associated with differences in practitioners nested within groups were accounted for. Rates of child learning were greater among children whose existing behaviour were used as the building blocks for promoting child competence (asset-based practices) compared to children for whom the focus of intervention was promoting child acquisition of missing skills (needs-based practices). The methods of analysis illustrate a practical approach to clustered data analysis and the presentation of results in ways that highlight sources of variations in the rates of response-contingent learning among young children with multiple developmental disabilities and significant developmental delays. Copyright © 2018 The Author(s). Published by Elsevier Ltd.. All rights reserved.

The Role of Learning in Health Technology Assessments: An Empirical Assessment of Endovascular Aneurysm Repairs in German Hospitals.

PubMed

Varabyova, Yauheniya; Blankart, Carl Rudolf; Schreyögg, Jonas

2017-02-01

Changes in performance due to learning may dynamically influence the results of a technology evaluation through the change in effectiveness and costs. In this study, we estimate the effect of learning using the example of two minimally invasive treatments of abdominal aortic aneurysms: endovascular aneurysm repair (EVAR) and fenestrated EVAR (fEVAR). The analysis is based on the administrative data of over 40,000 patients admitted with unruptured abdominal aortic aneurysm to more than 500 different hospitals over the years 2006 to 2013. We examine two patient outcomes, namely, in-hospital mortality and length of stay using hierarchical regression models with random effects at the hospital level. The estimated models control for patient and hospital characteristics and take learning interdependency between EVAR and fEVAR into account. In case of EVAR, we observe a significant decrease both in the in-hospital mortality and length of stay with experience accumulated at the hospital level; however, the learning curve for fEVAR in both outcomes is effectively flat. To foster the consideration of learning in health technology assessments of medical devices, a general framework for estimating learning effects is derived from the analysis. © 2017 The Authors. Health Economics published by John Wiley & Sons, Ltd. © 2017 The Authors. Health Economics published by John Wiley & Sons, Ltd.

"A steep learning curve": junior doctor perspectives on the transition from medical student to the health-care workplace.

PubMed

Sturman, Nancy; Tan, Zachary; Turner, Jane

2017-05-26

The transition from medical student to hospital-based first year junior doctor (termed "intern" in Australia) is known to be challenging, and recent changes in clinical learning environments may reduce graduate preparedness for the intern workplace. Although manageable challenges and transitions are a stimulus to learning, levels of burnout in junior medical colleagues are concerning. In order to prepare and support medical graduates, educators need to understand contemporary junior doctor perspectives on this transition. Final-year University of Queensland medical students recruited junior doctors working in diverse hospital settings, and videorecorded individual semi-structured interviews about their transition from medical student to working as a junior doctor. Two clinical academics (NS and JT) and an intern (ZT) independently conducted a descriptive analysis of interview transcripts, and identified preliminary emerging concepts and themes, before reaching agreement by consensus on the major overarching themes. Three key themes emerged from the analysis of 15 interviews: internship as a "steep learning curve"; relationships and team; and seeking help. Participants described the intern transition as physically, mentally and emotionally exhausting. They learned to manage long days, administrative and clinical tasks, frequent interruptions and time pressures; identify priorities; deal with criticism without compromising key relationships; communicate succinctly; understand team roles (including their own status within hospital hierarchies); and negotiate conflict. Participants reported a drop in self-confidence, and difficulty maintaining self-care and social relationships. Although participants emphasised the importance of escalating concerns and seeking help to manage patients, they appeared more reluctant to seek help for personal issues and reported a number of barriers to doing so. Findings may assist educators in refining their intern preparation and intern training curricula, and ensuring that medical school and intern preparation priorities are not seen as competing. Insights from non-medical disciplines into the organisational and relational challenges facing junior doctors and their health-care teams may enhance inter-professional learning opportunities. Workplace support and teaching, especially from junior colleagues, is highly valued during the demanding intern transition.

Undergraduate Student Self-Efficacy and Perceptions of Virtual World Learning Experience

ERIC Educational Resources Information Center

Stanton, Lorraine May

2017-01-01

Virtual worlds are innovative teaching and learning methods that can provide immersive and engaging learning experiences (Lu, 2010). Though they have potential benefits, students sometimes experience a steep learning curve and discomfort with the technology (Warburton, 2009). This study explored how students in two American Studies classes using…

Individual Differences in a Positional Learning Task across the Adult Lifespan

ERIC Educational Resources Information Center

Rast, Philippe; Zimprich, Daniel

2010-01-01

This study aimed at modeling individual and average non-linear trajectories of positional learning using a structured latent growth curve approach. The model is based on an exponential function which encompasses three parameters: Initial performance, learning rate, and asymptotic performance. These learning parameters were compared in a positional…

The learning curve of laparoendoscopic single-site (LESS) cholecystectomy: definable, short, and safe.

PubMed

Hernandez, Jonathan; Ross, Sharona; Morton, Connor; McFarlin, Kellie; Dahal, Sujat; Golkar, Farhaad; Albrink, Michael; Rosemurgy, Alexander

2010-11-01

The applications of laparoendoscopic single-site (LESS) surgery, including cholecystectomy, are occurring quickly, although little is generally known about issues associated with the learning curve of this new technique including operative time, conversion rates, and safety. We prospectively followed all patients undergoing LESS cholecystectomy, and compared operations undertaken at our institutions in cohorts of 25 patients with respect to operative times, conversion rates, and complications. One-hundred fifty patients of mean age 46 years underwent LESS cholecystectomy. No significant differences in operative times were demonstrable between any of the 25-patient cohorts operated on at our institution. A significant reduction in operative times (p < 0.001) after completion of 75 LESS procedures was, however, identified with the experience of a single surgeon. No significant reduction in the number of procedures requiring an additional trocar(s) or conversion to open operations was observed after completion of 25 LESS cholecystectomies. Complication rates were low, and not significantly different between any 25-patient cohorts. For surgeons proficient with multi-incision laparoscopic cholecystectomy, the learning curve for LESS cholecystectomy begins near proficiency. Operative complications and conversions were infrequent and unchanged across successive 25-patient cohorts, and were similar to those reported for multi-incision laparoscopic cholecystectomy after the learning curve. Copyright © 2010 American College of Surgeons. Published by Elsevier Inc. All rights reserved.

Detection of tuberculosis patterns in digital photographs of chest X-ray images using Deep Learning: feasibility study.

PubMed

Becker, A S; Blüthgen, C; Phi van, V D; Sekaggya-Wiltshire, C; Castelnuovo, B; Kambugu, A; Fehr, J; Frauenfelder, T

2018-03-01

To evaluate the feasibility of Deep Learning-based detection and classification of pathological patterns in a set of digital photographs of chest X-ray (CXR) images of tuberculosis (TB) patients. In this prospective, observational study, patients with previously diagnosed TB were enrolled. Photographs of their CXRs were taken using a consumer-grade digital still camera. The images were stratified by pathological patterns into classes: cavity, consolidation, effusion, interstitial changes, miliary pattern or normal examination. Image analysis was performed with commercially available Deep Learning software in two steps. Pathological areas were first localised; detected areas were then classified. Detection was assessed using receiver operating characteristics (ROC) analysis, and classification using a confusion matrix. The study cohort was 138 patients with human immunodeficiency virus (HIV) and TB co-infection (median age 34 years, IQR 28-40); 54 patients were female. Localisation of pathological areas was excellent (area under the ROC curve 0.82). The software could perfectly distinguish pleural effusions from intraparenchymal changes. The most frequent misclassifications were consolidations as cavitations, and miliary patterns as interstitial patterns (and vice versa). Deep Learning analysis of CXR photographs is a promising tool. Further efforts are needed to build larger, high-quality data sets to achieve better diagnostic performance.

Laparoscopic recurrent inguinal hernia repair during the learning curve: it can be done?

PubMed

Bracale, Umberto; Sciuto, Antonio; Andreuccetti, Jacopo; Merola, Giovanni; Pecchia, Leandro; Melillo, Paolo; Pignata, Giusto

2017-01-01

Trans-Abdominal Preperitoneal Patch (TAPP) repairs for Recurrent Hernia (RH) is a technically demanding procedure. It has to be performed only by surgeons with extensive experience in the laparoscopic approach. The purpose of this study is to evaluate the surgical safety and the efficacy of TAPP for RH performed in a tutoring program by surgeons in practice (SP). All TAPP repairs for RH performed by the same surgical team have been included in the study. We have evaluated the results of three SP during their learning curve in a tutoring program. Then these results have been compared to those of a highly experienced laparoscopic surgeon (Benchmark). A total of 530 TAPP repairs have been performed. Among these, 83 TAPP have been executed for RH, of which 43 by the Benchmark and 40 by the SP. When we have compared the outcomes of the Benchmark with those of SP, no significant difference has been observed about morbidity and recurrence while the operative time has been significantly longer for the SP. No intraoperative complications have occurred. International guidelines urge that TAPP repair for RH has to be performed only by surgeons with extensive experience in the laparoscopic approach. The results of the present study demonstrate that TAPP for RH could be performed also by surgeons in training during a learning program. We retain that an adequate tutoring program could lead a surgeon in practice to perform more complex hernia procedures without jeopardizing patient safety throughout the learning curve period. Laparoscopy, Learning Curve, Recurrent Hernia.

Reinforcement Learning of Linking and Tracing Contours in Recurrent Neural Networks

PubMed Central

Brosch, Tobias; Neumann, Heiko; Roelfsema, Pieter R.

2015-01-01

The processing of a visual stimulus can be subdivided into a number of stages. Upon stimulus presentation there is an early phase of feedforward processing where the visual information is propagated from lower to higher visual areas for the extraction of basic and complex stimulus features. This is followed by a later phase where horizontal connections within areas and feedback connections from higher areas back to lower areas come into play. In this later phase, image elements that are behaviorally relevant are grouped by Gestalt grouping rules and are labeled in the cortex with enhanced neuronal activity (object-based attention in psychology). Recent neurophysiological studies revealed that reward-based learning influences these recurrent grouping processes, but it is not well understood how rewards train recurrent circuits for perceptual organization. This paper examines the mechanisms for reward-based learning of new grouping rules. We derive a learning rule that can explain how rewards influence the information flow through feedforward, horizontal and feedback connections. We illustrate the efficiency with two tasks that have been used to study the neuronal correlates of perceptual organization in early visual cortex. The first task is called contour-integration and demands the integration of collinear contour elements into an elongated curve. We show how reward-based learning causes an enhancement of the representation of the to-be-grouped elements at early levels of a recurrent neural network, just as is observed in the visual cortex of monkeys. The second task is curve-tracing where the aim is to determine the endpoint of an elongated curve composed of connected image elements. If trained with the new learning rule, neural networks learn to propagate enhanced activity over the curve, in accordance with neurophysiological data. We close the paper with a number of model predictions that can be tested in future neurophysiological and computational studies. PMID:26496502

Yaxx: Yet another X-ray extractor

NASA Astrophysics Data System (ADS)

Aldcroft, Tom

2013-06-01

Yaxx is a Perl script that facilitates batch data processing using Perl open source software and commonly available software such as CIAO/Sherpa, S-lang, SAS, and FTOOLS. For Chandra and XMM analysis it includes automated spectral extraction, fitting, and report generation. Yaxx can be run without climbing an extensive learning curve; even so, yaxx is highly configurable and can be customized to support complex analysis. yaxx uses template files and takes full advantage of the unique Sherpa / S-lang environment to make much of the processing user configurable. Although originally developed with an emphasis on X-ray data analysis, yaxx evolved to be a general-purpose pipeline scripting package.

Assessment of competency in endoscopy: establishing and validating generalizable competency benchmarks for colonoscopy.

PubMed

Sedlack, Robert E; Coyle, Walter J

2016-03-01

The Mayo Colonoscopy Skills Assessment Tool (MCSAT) has previously been used to describe learning curves and competency benchmarks for colonoscopy; however, these data were limited to a single training center. The newer Assessment of Competency in Endoscopy (ACE) tool is a refinement of the MCSAT tool put forth by the Training Committee of the American Society for Gastrointestinal Endoscopy, intended to include additional important quality metrics. The goal of this study is to validate the changes made by updating this tool and establish more generalizable and reliable learning curves and competency benchmarks for colonoscopy by examining a larger national cohort of trainees. In a prospective, multicenter trial, gastroenterology fellows at all stages of training had their core cognitive and motor skills in colonoscopy assessed by staff. Evaluations occurred at set intervals of every 50 procedures throughout the 2013 to 2014 academic year. Skills were graded by using the ACE tool, which uses a 4-point grading scale defining the continuum from novice to competent. Average learning curves for each skill were established at each interval in training and competency benchmarks for each skill were established using the contrasting groups method. Ninety-three gastroenterology fellows at 10 U.S. academic institutions had 1061 colonoscopies assessed by using the ACE tool. Average scores of 3.5 were found to be inclusive of all minimal competency thresholds identified for each core skill. Cecal intubation times of less than 15 minutes and independent cecal intubation rates of 90% were also identified as additional competency thresholds during analysis. The average fellow achieved all cognitive and motor skill endpoints by 250 procedures, with >90% surpassing these thresholds by 300 procedures. Nationally generalizable learning curves for colonoscopy skills in gastroenterology fellows are described. Average ACE scores of 3.5, cecal intubation rates of 90%, and intubation times less than 15 minutes are recommended as minimal competency criteria. On average, it takes 250 procedures to achieve competence in colonoscopy. The thresholds found in this multicenter cohort by using the ACE tool are nearly identical to the previously established MCSAT benchmarks and are consistent with recent gastroenterology training recommendations but far higher than current training requirements in other specialties. Copyright © 2016 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.

Learning curve for gastric cancer patients with laparoscopy-assisted distal gastrectomy

PubMed Central

Zhao, Lin-Yong; Zhang, Wei-Han; Sun, Yan; Chen, Xin-Zu; Yang, Kun; Liu, Kai; Chen, Xiao-Long; Wang, Yi-Gao; Song, Xiao-Hai; Xue, Lian; Zhou, Zong-Guang; Hu, Jian-Kun

2016-01-01

Abstract Laparoscopy-assisted distal gastrectomy (LADG) is widely used for gastric cancer (GC) patients nowadays. This study aimed to investigate the time trend of outcomes so as to describe the learning curve for GC patients with LADG at a single medical institution in western China over a 6-year period. A total of 246 consecutive GC patients with LADG were divided into 5 groups (group A: 46 patients from 2006 to 2007; group B: 47 patients in 2008; group C: 49 patients in 2009; group D: 73 patients in 2010; and group E: 31 patients in 2011). All surgeries were conducted by the same surgeon. Comparative analyses were successively performed by Mann–Whitney U test or Student t test among the 5 different groups for the clinical data, including clinicopathologic characteristics, surgical parameters, postoperative course, and survival outcomes, through which the learning curve was described. There were no differences in the baseline information among the 5 groups (P > 0.05), and the proportion of advanced GC patients with LADG slightly increased from 58.7% to 77.4% during the 6 years. Besides, the proportion of D2/D2+ lymphadenectomy and the number of retrieved lymph nodes gradually grew from 60.9% to 80.6% and from 20.0 to 28.8, respectively. In addition, the operation time decreased from 299.2 to 267.8 minutes, while the estimated blood loss dropped from 175.2 to 146.8 mL. Furthermore, some surgical parameters (surgical duration and blood loss) and postoperative course (such as postoperative complications, the time to ambulation, to first flatus, and to first liquid intake as well as the length of hospital stay) were all observed to be significantly different between group A and other groups (P < 0.05), illustrating a similar downward trend and remaining stable to form a plateau after 46 cases in group A. However, no difference on overall survival was found among these 5 groups, and multivariate analysis indicated that factors, such as age, tumor differentiation, tumor size, and T stage as well as N stage, were independent prognostic factors for patients with LADG. Improvement on surgical parameters and postoperative course can be seen over the past years, and the cutoff value of the learning curve of LADG for surgeons with rich experience in open operation might be 46 cases. PMID:27631257

Technical refinement and learning curve for attenuating neurapraxia during robotic-assisted radical prostatectomy to improve sexual function.

PubMed

Alemozaffar, Mehrdad; Duclos, Antoine; Hevelone, Nathanael D; Lipsitz, Stuart R; Borza, Tudor; Yu, Hua-Yin; Kowalczyk, Keith J; Hu, Jim C

2012-06-01

While radical prostatectomy surgeon learning curves have characterized less blood loss, shorter operative times, and fewer positive margins, there is a dearth of studies characterizing learning curves for improving sexual function. Additionally, while learning curve studies often define volume thresholds for improvement, few of these studies demonstrate specific technical modifications that allow reproducibility of improved outcomes. Demonstrate and quantify the learning curve for improving sexual function outcomes based on technical refinements that reduce neurovascular bundle displacement during nerve-sparing robot-assisted radical prostatectomy (RARP). We performed a retrospective study of 400 consecutive RARPs, categorized into groups of 50, performed after elimination of continuous surgeon/assistant neurovascular bundle countertraction. Our approach to RARP has been described previously. A single-console robotic system was used for all cases. Expanded Prostate Cancer Index Composite sexual function was measured within 1 yr of RARP. Linear regression was performed to determine factors influencing the recovery of sexual function. Greater surgeon experience was associated with better 5-mo sexual function (p = 0.007) and a trend for better 12-mo sexual function (p = 0.061), with improvement plateauing after 250-300 cases. Additionally, younger patient age (both p<0.02) and better preoperative sexual function (<0.001) were associated with better 5- and 12-mo sexual function. Moreover, trainee robotic console time during nerve sparing was associated with worse 12-mo sexual function (p=0.021), while unilateral nerve sparing/non-nerve sparing was associated with worse 5-mo sexual function (p = 0.009). Limitations include the retrospective single-surgeon design. With greater surgeon experience, attenuating lateral displacement of the neurovascular bundle and resultant neurapraxia improve postoperative sexual function. However, to maximize outcomes, appropriate patient selection must be exercised when allowing trainee nerve-sparing involvement. Copyright © 2012 European Association of Urology. Published by Elsevier B.V. All rights reserved.

A study of active learning methods for named entity recognition in clinical text.

PubMed

Chen, Yukun; Lasko, Thomas A; Mei, Qiaozhu; Denny, Joshua C; Xu, Hua

2015-12-01

Named entity recognition (NER), a sequential labeling task, is one of the fundamental tasks for building clinical natural language processing (NLP) systems. Machine learning (ML) based approaches can achieve good performance, but they often require large amounts of annotated samples, which are expensive to build due to the requirement of domain experts in annotation. Active learning (AL), a sample selection approach integrated with supervised ML, aims to minimize the annotation cost while maximizing the performance of ML-based models. In this study, our goal was to develop and evaluate both existing and new AL methods for a clinical NER task to identify concepts of medical problems, treatments, and lab tests from the clinical notes. Using the annotated NER corpus from the 2010 i2b2/VA NLP challenge that contained 349 clinical documents with 20,423 unique sentences, we simulated AL experiments using a number of existing and novel algorithms in three different categories including uncertainty-based, diversity-based, and baseline sampling strategies. They were compared with the passive learning that uses random sampling. Learning curves that plot performance of the NER model against the estimated annotation cost (based on number of sentences or words in the training set) were generated to evaluate different active learning and the passive learning methods and the area under the learning curve (ALC) score was computed. Based on the learning curves of F-measure vs. number of sentences, uncertainty sampling algorithms outperformed all other methods in ALC. Most diversity-based methods also performed better than random sampling in ALC. To achieve an F-measure of 0.80, the best method based on uncertainty sampling could save 66% annotations in sentences, as compared to random sampling. For the learning curves of F-measure vs. number of words, uncertainty sampling methods again outperformed all other methods in ALC. To achieve 0.80 in F-measure, in comparison to random sampling, the best uncertainty based method saved 42% annotations in words. But the best diversity based method reduced only 7% annotation effort. In the simulated setting, AL methods, particularly uncertainty-sampling based approaches, seemed to significantly save annotation cost for the clinical NER task. The actual benefit of active learning in clinical NER should be further evaluated in a real-time setting. Copyright © 2015 Elsevier Inc. All rights reserved.

Using a high-fidelity patient simulator with first-year medical students to facilitate learning of cardiovascular function curves.

PubMed

Harris, David M; Ryan, Kathleen; Rabuck, Cynthia

2012-09-01

Students are relying on technology for learning more than ever, and educators need to adapt to facilitate student learning. High-fidelity patient simulators (HFPS) are usually reserved for the clinical years of medical education and are geared to improve clinical decision skills, teamwork, and patient safety. Finding ways to incorporate HFPS into preclinical medical education represents more of a challenge, and there is limited literature regarding its implementation. The main objective of this study was to implement a HFPS activity into a problem-based curriculum to enhance the learning of basic sciences. More specifically, the focus was to aid in student learning of cardiovascular function curves and help students develop heart failure treatment strategies based on basic cardiovascular physiology concepts. Pretests and posttests, along with student surveys, were used to determine student knowledge and perception of learning in two first-year medical school classes. There was an increase of 21% and 22% in the percentage of students achieving correct answers on a posttest compared with their pretest score. The median number of correct questions increased from pretest scores of 2 and 2.5 to posttest scores of 4 and 5 of a possible total of 6 in each respective year. Student survey data showed agreement that the activity aided in learning. This study suggests that a HFPS activity can be implemented during the preclinical years of medical education to address basic science concepts. Additionally, it suggests that student learning of cardiovascular function curves and heart failure strategies are facilitated.

Creating Data and Modeling Enabled Hydrology Instruction Using Collaborative Approach

NASA Astrophysics Data System (ADS)

Merwade, V.; Rajib, A.; Ruddell, B. L.; Fox, S.

2017-12-01

Hydrology instruction typically involves teaching of the hydrologic cycle and the processes associated with it such as precipitation, evapotranspiration, infiltration, runoff generation and hydrograph analysis. With the availability of observed and remotely sensed data related to many hydrologic fluxes, there is an opportunity to use these data for place based learning in hydrology classrooms. However, it is not always easy and possible for an instructor to complement an existing hydrology course with new material that requires both the time and technical expertise, which the instructor may not have. The work presented here describes an effort where students create the data and modeling driven instruction material as a part of their class assignment for a hydrology course at Purdue University. The data driven hydrology education project within Science Education Resources Center (SERC) is used as a platform to publish and share the instruction material so it can be used by future students in the same course or any other course anywhere in the world. Students in the class were divided into groups, and each group was assigned a topic such as precipitation, evapotranspiration, streamflow, flow duration curve and frequency analysis. Each student in the group was then asked to get data and do some analysis for an area with specific landuse characteristic such as urban, rural and agricultural. The student contribution were then organized into learning units such that someone can do a flow duration curve analysis or flood frequency analysis to see how it changes for rural area versus urban area. The hydrology education project within SERC cyberinfrastructure enables any other instructor to adopt this material as is or through modification to suit his/her place based instruction needs.

V1 orientation plasticity is explained by broadly tuned feedforward inputs and intracortical sharpening.

PubMed

Teich, Andrew F; Qian, Ning

2010-03-01

Orientation adaptation and perceptual learning change orientation tuning curves of V1 cells. Adaptation shifts tuning curve peaks away from the adapted orientation, reduces tuning curve slopes near the adapted orientation, and increases the responses on the far flank of tuning curves. Learning an orientation discrimination task increases tuning curve slopes near the trained orientation. These changes have been explained previously in a recurrent model (RM) of orientation selectivity. However, the RM generates only complex cells when they are well tuned, so that there is currently no model of orientation plasticity for simple cells. In addition, some feedforward models, such as the modified feedforward model (MFM), also contain recurrent cortical excitation, and it is unknown whether they can explain plasticity. Here, we compare plasticity in the MFM, which simulates simple cells, and a recent modification of the RM (MRM), which displays a continuum of simple-to-complex characteristics. Both pre- and postsynaptic-based modifications of the recurrent and feedforward connections in the models are investigated. The MRM can account for all the learning- and adaptation-induced plasticity, for both simple and complex cells, while the MFM cannot. The key features from the MRM required for explaining plasticity are broadly tuned feedforward inputs and sharpening by a Mexican hat intracortical interaction profile. The mere presence of recurrent cortical interactions in feedforward models like the MFM is insufficient; such models have more rigid tuning curves. We predict that the plastic properties must be absent for cells whose orientation tuning arises from a feedforward mechanism.

A strategic systems perspective of organizational learning theory: models for a case study at the Jet Propulsion Laboratory

NASA Technical Reports Server (NTRS)

Neece, O.

2000-01-01

Organizational learning is an umbrella term that covers a variety of topics including; learning curves, productivity, organizational memory, organizational forgetting, knowledge transfer, knowledge sharing and knowledge creation. This treatise will review some of these theories in concert with a model of how organizations learn.

The Role of Cognitive Abilities in Laparoscopic Simulator Training

ERIC Educational Resources Information Center

Groenier, M.; Schraagen, J. M. C.; Miedema, H. A. T.; Broeders, I. A. J. M.

2014-01-01

Learning minimally invasive surgery (MIS) differs substantially from learning open surgery and trainees differ in their ability to learn MIS. Previous studies mainly focused on the role of visuo-spatial ability (VSA) on the learning curve for MIS. In the current study, the relationship between spatial memory, perceptual speed, and general…

Adolescent Idiopathic Scoliosis Surgery by a Neurosurgeon: Learning Curve for Neurosurgeons.

PubMed

Hyun, Seung-Jae; Han, Sanghyun; Kim, Ki-Jeong; Jahng, Tae-Ahn; Kim, Yongjung J; Rhim, Seung-Chul; Kim, Hyun-Jib

2018-02-01

To determine a neurosurgeon's learning curve of surgical treatment for adolescent idiopathic scoliosis (AIS) patients. This study is a retrospective analysis. Forty-six patients were treated by a single neurosurgeon between 2011 and 2017 using posterior segmental instrumentation and fusion. According to the time period, the former and latter 23 patients were divided into group 1 and group 2, respectively. Patients' demographic data, curve magnitude, number of levels treated, amount of correction achieved, radiographic/clinical outcomes, and complications were compared between the groups. The majority were females (34 vs. 12) with average ages of 15.0 versus 15.6, respectively. The mean follow-up period was 24.6 months. The average number of fusion levels was similar with 10.3 and 11.5 vertebral bodies in groups 1 and 2, respectively. The average Cobb angle of major curvature was 59.8° and 58.5° in groups 1 and 2, respectively. There observed significant reductions of operative time (324.4 vs. 224.7 minutes, P = 0.007) and estimated blood loss (648.3 vs. 438.0 mL, P = 0.027) in group 2. The correction rate of the major structural curve was greater in group 2 (70.7% vs. 81.0%, P = 0.001). There was no case of neurologic deficit, infection, and revision for screw malposition. One patient of group 1 underwent fusion extension surgery for shoulder asymmetry. Radiographic and clinical outcomes of AIS patients treated by a neurosurgeon were acceptable. AIS surgery may be performed with an acceptable rate of complications after about 20 surgeries. With acquisition of surgical experiences, neurosurgeons could perform deformity surgery for AIS effectively and safely. Copyright © 2017 Elsevier Inc. All rights reserved.

Digital tomosynthesis for evaluating metastatic lung nodules: nodule visibility, learning curves, and reading times.

PubMed

Lee, Kyung Hee; Goo, Jin Mo; Lee, Sang Min; Park, Chang Min; Bahn, Young Eun; Kim, Hyungjin; Song, Yong Sub; Hwang, Eui Jin

2015-01-01

To evaluate nodule visibility, learning curves, and reading times for digital tomosynthesis (DT). We included 80 patients who underwent computed tomography (CT) and DT before pulmonary metastasectomy. One experienced chest radiologist annotated all visible nodules on thin-section CT scans using computer-aided detection software. Two radiologists used CT as the reference standard and retrospectively graded the visibility of nodules on DT. Nodule detection performance was evaluated in four sessions of 20 cases each by six readers. After each session, readers were unblinded to the DT images by revealing the true-positive markings and were instructed to self-analyze their own misreads. Receiver-operating-characteristic curves were determined. Among 414 nodules on CT, 53.3% (221/414) were visible on DT. The main reason for not seeing a nodule on DT was small size (93.3%, ≤ 5 mm). DT revealed a substantial number of malignant nodules (84.1%, 143/170). The proportion of malignant nodules among visible nodules on DT was significantly higher (64.7%, 143/221) than that on CT (41.1%, 170/414) (p < 0.001). Area under the curve (AUC) values at the initial session were > 0.8, and the average detection rate for malignant nodules was 85% (210/246). The inter-session analysis of the AUC showed no significant differences among the readers, and the detection rate for malignant nodules did not differ across sessions. A slight improvement in reading times was observed. Most malignant nodules > 5 mm were visible on DT. As nodule detection performance was high from the initial session, DT may be readily applicable for radiology residents and board-certified radiologists.

Machine learning methods for the classification of gliomas: Initial results using features extracted from MR spectroscopy.

PubMed

Ranjith, G; Parvathy, R; Vikas, V; Chandrasekharan, Kesavadas; Nair, Suresh

2015-04-01

With the advent of new imaging modalities, radiologists are faced with handling increasing volumes of data for diagnosis and treatment planning. The use of automated and intelligent systems is becoming essential in such a scenario. Machine learning, a branch of artificial intelligence, is increasingly being used in medical image analysis applications such as image segmentation, registration and computer-aided diagnosis and detection. Histopathological analysis is currently the gold standard for classification of brain tumors. The use of machine learning algorithms along with extraction of relevant features from magnetic resonance imaging (MRI) holds promise of replacing conventional invasive methods of tumor classification. The aim of the study is to classify gliomas into benign and malignant types using MRI data. Retrospective data from 28 patients who were diagnosed with glioma were used for the analysis. WHO Grade II (low-grade astrocytoma) was classified as benign while Grade III (anaplastic astrocytoma) and Grade IV (glioblastoma multiforme) were classified as malignant. Features were extracted from MR spectroscopy. The classification was done using four machine learning algorithms: multilayer perceptrons, support vector machine, random forest and locally weighted learning. Three of the four machine learning algorithms gave an area under ROC curve in excess of 0.80. Random forest gave the best performance in terms of AUC (0.911) while sensitivity was best for locally weighted learning (86.1%). The performance of different machine learning algorithms in the classification of gliomas is promising. An even better performance may be expected by integrating features extracted from other MR sequences. © The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav.

A randomized control trial to evaluate the importance of pre-training basic laparoscopic psychomotor skills upon the learning curve of laparoscopic intra-corporeal knot tying.

PubMed

Molinas, Carlos Roger; Binda, Maria Mercedes; Sisa, Cesar Manuel; Campo, Rudi

2017-01-01

Training of basic laparoscopic psychomotor skills improves the acquisition of more advanced laparoscopic tasks, such as laparoscopic intra-corporeal knot tying (LICK). This randomized controlled trial was designed to evaluate whether pre-training of basic skills, as laparoscopic camera navigation (LCN), hand-eye coordination (HEC), and bimanual coordination (BMC), and the combination of the three of them, has any beneficial effect upon the learning curve of LICK. The study was carried out in a private center in Asunción, Paraguay, by 80 medical students without any experience in surgery. Four laparoscopic tasks were performed in the ENCILAP model (LCN, HEC, BMC, and LICK). Participants were allocated to 5 groups (G1-G5). The study was structured in 5 phases. In phase 1, they underwent a base-line test ( T 1 ) for all tasks (1 repetition of each task in consecutive order). In phase 2, participants underwent different training programs (30 consecutive repetitions) for basic tasks according to the group they belong to (G1: none; G2: LCN; G3: HEC; G4: BMC; and G5: LCN, HEC, and BMC). In phase 3, they were tested again ( T 2 ) in the same manner than at T 1 . In phase 4, they underwent a standardized training program for LICK (30 consecutive repetitions). In phase 5, they were tested again ( T 3 ) in the same manner than at T 1 and T 2 . At each repetition, scoring was based on the time taken for task completion system. The scores were plotted and non-linear regression models were used to fit the learning curves to one- and two-phase exponential decay models for each participant (individual curves) and for each group (group curves). The LICK group learning curves fitted better to the two-phase exponential decay model. From these curves, the starting points ( Y 0), the point after HEC training/before LICK training ( Y 1), the Plateau, and the rate constants ( K ) were calculated. All groups, except for G4, started from a similar point ( Y 0). At Y 1, G5 scored already better than the others (G1 p  = .004; G2 p  = .04; G3 p  < .0001; G4 NS). Although all groups reached a similar Plateau, G5 has a quicker learning than the others, demonstrated by a higher K (G1 p  < 0.0001; G2 p  < 0.0001; G3 p  < 0.0001; and G4 p  < 0.0001). Our data confirms that training improves laparoscopic skills and demonstrates that pre-training of all basic skills (i.e., LCN, HEC, and BMC) shortens the LICK learning curve.

Selection of software for mechanical engineering undergraduates

DOE Office of Scientific and Technical Information (OSTI.GOV)

Cheah, C. T.; Yin, C. S.; Halim, T.

A major problem with the undergraduate mechanical course is the limited exposure of students to software packages coupled with the long learning curve on the existing software packages. This work proposes the use of appropriate software packages for the entire mechanical engineering curriculum to ensure students get sufficient exposure real life design problems. A variety of software packages are highlighted as being suitable for undergraduate work in mechanical engineering, e.g. simultaneous non-linear equations; uncertainty analysis; 3-D modeling software with the FEA; analysis tools for the solution of problems in thermodynamics, fluid mechanics, mechanical system design, and solid mechanics.

A systematic review and meta-analysis of randomized controlled trials comparing hysteroscopic morcellation with resectoscopy for patients with endometrial lesions.

PubMed

Li, Chunbo; Dai, Zhiyuan; Gong, Yuping; Xie, Bingying; Wang, Bei

2017-01-01

Results on the efficacy of hysteroscopic morcellation for patients with endometrial lesions remain conflicting. To compare hysteroscopic morcellation with conventional resectoscopy for removal of endometrial lesions. Electronic databases were searched for reports published up to February 1, 2016, using terms such as "morcellator," "morcellators," "morcellate," "morcellation," "morcellated," "hysteroscopy," "hysteroscopy," "uteroscope," and "transcervical." Randomized controlled trials were included if they assessed success rate, procedure speed, complications, tolerability, and/or learning curve. Data were extracted by two independent reviewers and a meta-analysis was performed. Four trials including 392 patients were analyzed. Successful removal of all endometrial lesions was more frequent with hysteroscopic morcellation than conventional resectoscopy (odds ratio 4.49, 95% confidence interval [CI] 1.94-10.41; P<0.001). Total operative time was also shorter with hysteroscopic morcellation (mean difference -4.94 minutes, 95% CI -7.20 to -2.68; P<0.001). No significant differences in complications were found. Meta-analyses were not possible for tolerability and learning curve. In one study, hysteroscopic morcellation was acceptable to more patients (P=0.009). Hysteroscopic morcellation is associated with a higher operative success rate and a shorter operative time among patients with endometrial lesions than is resectoscopy. More high-quality trials are required to validate these results. © 2016 International Federation of Gynecology and Obstetrics.

Precision Learning Assessment: An Alternative to Traditional Assessment Techniques.

ERIC Educational Resources Information Center

Caltagirone, Paul J.; Glover, Christopher E.

1985-01-01

A continuous and curriculum-based assessment method, Precision Learning Assessment (PLA), which integrates precision teaching and norm-referenced techniques, was applied to a math computation curriculum for 214 third graders. The resulting districtwide learning curves defining average annual progress through the computation curriculum provided…

Totally laparoscopic aortic surgery: comparison of the apron and retrocolic techniques in a porcine model.

PubMed

Huynh, Hai; Elkouri, Stephane; Beaudoin, Nathalie; Bruneau, Luc; Guimond, Cathie; Daniel, Véronique; Blair, Jean-François

2007-01-01

This study evaluated the learning curve for a second-year general surgery resident and compared 2 totally laparoscopic aortic surgery techniques in 10 pigs: the transretroperitoneal apron approach and the transperitoneal retrocolic approach. Five end points were compared: success rate, percentage of conversion, time required, laparoscopic anastomosis quality, and learning curve. The first 3 interventions required an open conversion. The last 7 were done without complications. Mean dissection time was significantly higher with the apron approach compared with the retrocolic approach. The total times for operation, clamping, and arteriotomy time were similar. All laparoscopic anastomoses were patent and without stenosis. The initial learning curve for laparoscopic anastomosis was relatively short for a second-year surgery resident. Both techniques resulted in satisfactory exposure of the aorta and similar mean operative and clamping time. Training on an ex vivo laparoscopic box trainer and on an animal model seems to be complementary to decrease laparoscopic anastomosis completion time.

Enhanced Night Vision Via a Combination of Poisson Interpolation and Machine Learning

DTIC Science & Technology

2006-02-01

of 0-255, they are mostly similar. The right plot shows a family of m(x, ψ) curves of ψ=2 (the most linear) through ψ=1024 (the most curved ...complicating low-light imaging. Nayar and Branzoi [04] later suggested a second variant using a DLP micromirror array to modulate the exposure, via time...255, they are mostly similar. The right plot shows a family of m(x, ψ) curves of ψ=2 (the most linear) through ψ=1024 (the most curved

Machine learning-based kinetic modeling: a robust and reproducible solution for quantitative analysis of dynamic PET data

NASA Astrophysics Data System (ADS)

Pan, Leyun; Cheng, Caixia; Haberkorn, Uwe; Dimitrakopoulou-Strauss, Antonia

2017-05-01

A variety of compartment models are used for the quantitative analysis of dynamic positron emission tomography (PET) data. Traditionally, these models use an iterative fitting (IF) method to find the least squares between the measured and calculated values over time, which may encounter some problems such as the overfitting of model parameters and a lack of reproducibility, especially when handling noisy data or error data. In this paper, a machine learning (ML) based kinetic modeling method is introduced, which can fully utilize a historical reference database to build a moderate kinetic model directly dealing with noisy data but not trying to smooth the noise in the image. Also, due to the database, the presented method is capable of automatically adjusting the models using a multi-thread grid parameter searching technique. Furthermore, a candidate competition concept is proposed to combine the advantages of the ML and IF modeling methods, which could find a balance between fitting to historical data and to the unseen target curve. The machine learning based method provides a robust and reproducible solution that is user-independent for VOI-based and pixel-wise quantitative analysis of dynamic PET data.

Machine learning-based kinetic modeling: a robust and reproducible solution for quantitative analysis of dynamic PET data.

PubMed

Pan, Leyun; Cheng, Caixia; Haberkorn, Uwe; Dimitrakopoulou-Strauss, Antonia

2017-05-07

A variety of compartment models are used for the quantitative analysis of dynamic positron emission tomography (PET) data. Traditionally, these models use an iterative fitting (IF) method to find the least squares between the measured and calculated values over time, which may encounter some problems such as the overfitting of model parameters and a lack of reproducibility, especially when handling noisy data or error data. In this paper, a machine learning (ML) based kinetic modeling method is introduced, which can fully utilize a historical reference database to build a moderate kinetic model directly dealing with noisy data but not trying to smooth the noise in the image. Also, due to the database, the presented method is capable of automatically adjusting the models using a multi-thread grid parameter searching technique. Furthermore, a candidate competition concept is proposed to combine the advantages of the ML and IF modeling methods, which could find a balance between fitting to historical data and to the unseen target curve. The machine learning based method provides a robust and reproducible solution that is user-independent for VOI-based and pixel-wise quantitative analysis of dynamic PET data.

Minimally invasive video-assisted thyroid surgery: how can we improve the learning curve?

PubMed

Castagnola, G; Giulii Cappone, M; Tierno, S M; Mezzetti, G; Centanini, F; Vetrone, I; Bellotti, C

2012-10-01

Minimally invasive video-assisted thyroidectomy (MIVAT) is a technically demanding procedure and requires a surgical team skilled in both endocrine and endoscopic surgery. A time consuming learning and training period is mandatory at the beginning of the experience. The aim of our report is to focus some aspects of the learning curve of the surgeon who practices video-assisted thyroid procedures for the first time, through the analysis of our preliminary series of 36 cases. From September 2004 to April 2005 we selected 36 patients for minimally invasive video-assisted surgery of the thyroid. The patients were considered eligible if they presented with a nodule not exceeding 35 mm in maximum diameter; total thyroid volume within normal range; absence of biochemical and echographic signs of thyroiditis. We analyzed surgical results, conversion rate, operating time, post-operative complications, hospital stay, cosmetic outcome of the series. We performed 36 total thyroidectomy. The procedure was successfully carried out in 33/36 cases. Post-operative complications included 3 transient recurrent nerve palsies and 2 transient hypocalcemias; no definitive hypoparathyroidism was registered. All patients were discharged 2 days after operation. The cosmetic result was considered excellent by most patients. Advances in skills and technology have enabled surgeons to reproduce most open surgical techniques with video-assistance or laparoscopically. Training is essential to acquire any new surgical technique and it should be organized in detail to exploit it completely.

Texture Analysis and Machine Learning for Detecting Myocardial Infarction in Noncontrast Low-Dose Computed Tomography: Unveiling the Invisible.

PubMed

Mannil, Manoj; von Spiczak, Jochen; Manka, Robert; Alkadhi, Hatem

2018-06-01

The aim of this study was to test whether texture analysis and machine learning enable the detection of myocardial infarction (MI) on non-contrast-enhanced low radiation dose cardiac computed tomography (CCT) images. In this institutional review board-approved retrospective study, we included non-contrast-enhanced electrocardiography-gated low radiation dose CCT image data (effective dose, 0.5 mSv) acquired for the purpose of calcium scoring of 27 patients with acute MI (9 female patients; mean age, 60 ± 12 years), 30 patients with chronic MI (8 female patients; mean age, 68 ± 13 years), and in 30 subjects (9 female patients; mean age, 44 ± 6 years) without cardiac abnormality, hereafter termed controls. Texture analysis of the left ventricle was performed using free-hand regions of interest, and texture features were classified twice (Model I: controls versus acute MI versus chronic MI; Model II: controls versus acute and chronic MI). For both classifications, 6 commonly used machine learning classifiers were used: decision tree C4.5 (J48), k-nearest neighbors, locally weighted learning, RandomForest, sequential minimal optimization, and an artificial neural network employing deep learning. In addition, 2 blinded, independent readers visually assessed noncontrast CCT images for the presence or absence of MI. In Model I, best classification results were obtained using the k-nearest neighbors classifier (sensitivity, 69%; specificity, 85%; false-positive rate, 0.15). In Model II, the best classification results were found with the locally weighted learning classification (sensitivity, 86%; specificity, 81%; false-positive rate, 0.19) with an area under the curve from receiver operating characteristics analysis of 0.78. In comparison, both readers were not able to identify MI in any of the noncontrast, low radiation dose CCT images. This study indicates the ability of texture analysis and machine learning in detecting MI on noncontrast low radiation dose CCT images being not visible for the radiologists' eye.

Uniportal video-assisted thoracoscopic surgery: safety, efficacy and learning curve during the first 250 cases in Quebec, Canada.

PubMed

Drevet, Gabrielle; Ugalde Figueroa, Paula

2016-03-01

Video-assisted thoracoscopic surgery (VATS) using a single incision (uniportal) may result in better pain control, earlier mobilization and shorter hospital stays. Here, we review the safety and efficiency of our initial experience with uniportal VATS and evaluate our learning curve. We conducted a retrospective review of uniportal VATS using a prospectively maintained departmental database and analyzed patients who had undergone a lung anatomic resection separately from patients who underwent other resections. To assess the learning curve, we compared the first 10 months of the study period with the second 10 months. From January 2014 to August 2015, 250 patients underwent intended uniportal VATS, including 180 lung anatomic resections (72%) and 70 other resections (28%). Lung anatomic resection was successfully completed using uniportal VATS in 153 patients (85%), which comprised all the anatomic segmentectomies (29 patients), 80% (4 of 5) of the pneumonectomies and 82% (120 of 146) of the lobectomies attempted. The majority of lung anatomic resections that required conversion to thoracotomy occurred in the first half of our study period. Seventy patients underwent other uniportal VATS resections. Wedge resections were the most common of these procedures (25 patients, 35.7%). Although 24 of the 70 patients (34%) required the placement of additional ports, none required conversion to thoracotomy. Uniportal VATS was safe and feasible for both standard and complex pulmonary resections. However, when used for pulmonary anatomic resections, uniportal VATS entails a steep learning curve.

Displays mounted on cutting blocks reduce the learning curve in navigated total knee arthroplasty.

PubMed

Schnurr, Christoph; Eysel, Peer; König, Dietmar Pierre

2011-01-01

The use of computer navigation in total knee arthroplasty (TKA) improves the implant alignment but increases the operation time. Studies have shown that the operation time is further prolonged due to the surgeon's learning curve, and longer operation times have been associated with higher morbidity risks. It has been our hypothesis that an improvement in the human-machine interface might reduce the time required during the learning curve. Accordingly, we asked whether the use of navigation devices with a display fixed on the surgical instruments would reduce the operation time in navigated TKAs performed by navigation beginners. Thirty medical students were randomized and used two navigation devices in rotation: these were the Kolibri® device with an external display and the Dash® device with a display that was fixed on the cutting blocks. The time for adjustment of the tibial and femoral cutting blocks on knee models while using these devices was measured. A significant time reduction was demonstration when the Dash® device was used: The time reduction was 21% for the tibial block (p = 0.007), 40% for the femoral block (p < 0.001), and 32% for the whole procedure (p < 0.001). The integrated display, fixed on surgical instruments in a manner similar to a spirit level, seems to be more user-friendly for navigation beginners. Hence, unproductive time losses during the learning curve may be diminished.

Learning curve of hysteroscopic placement of tubal sterilization microinserts in 15 gynecologists in the Netherlands.

PubMed

Janse, Juliënne A; Pattij, Thyrza O S; Eijkemans, Marinus J C; Broekmans, Frank J; Veersema, Sebastiaan; Schreuder, Henk W R

2013-09-01

To evaluate the learning curve of hysteroscopic placement of tubal sterilization microinserts by gynecologists in the Netherlands. Prospective multicenter study (Canadian Task Force II-2). Ten community (teaching) hospitals in the Netherlands. A total of 631 women who underwent permanent sterilization by tubal microinserts. Hysteroscopic placement of tubal sterilization microinserts performed by 15 gynecologists experienced in performing operative hysteroscopy, starting from their very first placement. Effect of increasing experience in time on procedure time, pain score, successful bilateral placement, and complications. Bilateral successful placement with confirmation of adequate positioning at follow-up evaluation was achieved in 480 (76.1%) patients at first attempt and in 44 (7.0%) at second attempt. Median procedure time was 8.0 minutes (range: 3-40), and 31 (4.9%) patients were lost to follow-up evaluation. Gravidity showed to be a confounding factor and was consequently adjusted for. A learning curve was seen in a statistically significant decrease of procedure time with increasing experience. The decrease in procedure time extended to 11 to 15 cases and was followed by a plateau phase of the subsequent 60 cases. In contrast, pain scores, successful placement, and complication rate appeared not to improve with increasing experience. A learning curve for hysteroscopic tubal sterilization was seen for procedure time, but successful placement, pain score, and complication rate were not clearly influenced by increasing experience. Copyright © 2013 American Society for Reproductive Medicine. Published by Elsevier Inc. All rights reserved.

Learning curve and early clinical outcomes for a robotic surgery novice performing robotic single site cholecystectomy.

PubMed

Angus, Andrew A; Sahi, Saad L; McIntosh, Bruce B

2014-06-01

A rapid training protocol has been developed for robotic surgery novices to learn robotic single-incision techniques. This study assesses the learning curve and early clinical results for a robotic surgery novice starting single-site cholecystectomy. A chart review was performed on the surgeon's first 55 patients to undergo this procedure. Average patient age was 46.01 ± 4.25 (range 21-86) years and BMI was 26.57 ± 4.25 (range 19.4-36.6) kg/m(2) . The mean port placement with docking time was 11.34 ± 3.74 (range 7-23) min. Mean console time was 28.74 ± 11.04 (range 15-66) min. Average total OR time was 61.84 ± 14.66 (range 40-105) min. All procedures were successfully completed without conversion or added ports. Complications included several minor procedural gall bladder perforations and miscellaneous postoperative symptomatic complaints. Robotic single site cholecystectomy can be safely performed by a robotic novice within a minimal learning curve and have early clinical results that are comparable to the published data of robotic experts. Copyright © 2013 John Wiley & Sons, Ltd.

Validation of a structured training and assessment curriculum for technical skill acquisition in minimally invasive surgery: a randomized controlled trial.

PubMed

Palter, Vanessa N; Orzech, Neil; Reznick, Richard K; Grantcharov, Teodor P

2013-02-01

: To develop and validate an ex vivo comprehensive curriculum for a basic laparoscopic procedure. : Although simulators have been well validated as tools to teach technical skills, their integration into comprehensive curricula is lacking. Moreover, neither the effect of ex vivo training on learning curves in the operating room (OR), nor the effect on nontechnical proficiency has been investigated. : This randomized single-blinded prospective trial allocated 20 surgical trainees to a structured training and assessment curriculum (STAC) group or conventional residency training. The STAC consisted of case-based learning, proficiency-based virtual reality training, laparoscopic box training, and OR participation. After completion of the intervention, all participants performed 5 sequential laparoscopic cholecystectomies in the OR. The primary outcome measure was the difference in technical performance between the 2 groups during the first laparoscopic cholecystectomy. Secondary outcome measures included differences with respect to learning curves in the OR, technical proficiency of each sequential laparoscopic cholecystectomy, and nontechnical skills. : Residents in the STAC group outperformed residents in the conventional group in the first (P = 0.004), second (P = 0.036), third (P = 0.021), and fourth (P = 0.023) laparoscopic cholecystectomies. The conventional group demonstrated a significant learning curve in the OR (P = 0.015) in contrast to the STAC group (P = 0.032). Residents in the STAC group also had significantly higher nontechnical skills (P = 0.027). : Participating in the STAC shifted the learning curve for a basic laparoscopic procedure from the operating room into the simulation laboratory. STAC-trained residents had superior technical proficiency in the OR and nontechnical skills compared with conventionally trained residents. (The study registration ID is NCT01560494.).

Photometric Supernova Classification with Machine Learning

NASA Astrophysics Data System (ADS)

Lochner, Michelle; McEwen, Jason D.; Peiris, Hiranya V.; Lahav, Ofer; Winter, Max K.

2016-08-01

Automated photometric supernova classification has become an active area of research in recent years in light of current and upcoming imaging surveys such as the Dark Energy Survey (DES) and the Large Synoptic Survey Telescope, given that spectroscopic confirmation of type for all supernovae discovered will be impossible. Here, we develop a multi-faceted classification pipeline, combining existing and new approaches. Our pipeline consists of two stages: extracting descriptive features from the light curves and classification using a machine learning algorithm. Our feature extraction methods vary from model-dependent techniques, namely SALT2 fits, to more independent techniques that fit parametric models to curves, to a completely model-independent wavelet approach. We cover a range of representative machine learning algorithms, including naive Bayes, k-nearest neighbors, support vector machines, artificial neural networks, and boosted decision trees (BDTs). We test the pipeline on simulated multi-band DES light curves from the Supernova Photometric Classification Challenge. Using the commonly used area under the curve (AUC) of the Receiver Operating Characteristic as a metric, we find that the SALT2 fits and the wavelet approach, with the BDTs algorithm, each achieve an AUC of 0.98, where 1 represents perfect classification. We find that a representative training set is essential for good classification, whatever the feature set or algorithm, with implications for spectroscopic follow-up. Importantly, we find that by using either the SALT2 or the wavelet feature sets with a BDT algorithm, accurate classification is possible purely from light curve data, without the need for any redshift information.

Endoscopic sleeve gastroplasty: the learning curve.

PubMed

Hill, Christine; El Zein, Mohamad; Agnihotri, Abhishek; Dunlap, Margo; Chang, Angela; Agrawal, Alison; Barola, Sindhu; Ngamruengphong, Saowanee; Chen, Yen-I; Kalloo, Anthony N; Khashab, Mouen A; Kumbhari, Vivek

2017-09-01

 Endoscopic sleeve gastroplasty (ESG) is gaining traction as a minimally invasive bariatric treatment. Concern that the learning curve may be slow, even among those proficient in endoscopic suturing, is a barrier to widespread implementation of the procedure. Therefore, we aimed to define the learning curve for ESG in a single endoscopist experienced in endoscopic suturing who participated in a 1-day ESG training program.  Consecutive patients who underwent ESG between February 2016 and November 2016 were included. The performing endoscopist, who is proficient in endoscopic suturing for non-ESG procedures, participated in a 1-day ESG training session before offering ESG to patients. The outcome measurements were length of procedure (LOP) and number of plications per procedure. Nonlinear regression was used to determine the learning plateau and calculate the learning rate.  Twenty-one consecutive patients (8 males), with mean age 47.7 ± 11.2 years and mean body mass index 41.8 ± 8.5 kg/m 2 underwent ESG. LOP decreased significantly across consecutive procedures, with a learning plateau at 101.5 minutes and a learning rate of 7 cases ( P  = 0.04). The number of plications per procedure also decreased significantly across consecutive procedures, with a plateau at 8 sutures and a learning rate of 9 cases ( P  < 0.001). Further, the average time per plication decreased significantly with consecutive procedures, reaching a plateau at 9 procedures ( P  < 0.001).  Endoscopists experienced in endoscopic suturing are expected to achieve a reduction in LOP and number of plications per procedure in successive cases, with progress plateauing at 7 and 9 cases, respectively.

Acquisition of neural learning in cerebellum and cerebral cortex for smooth pursuit eye movements

PubMed Central

Li, Jennifer X.; Medina, Javier F.; Frank, Loren M.; Lisberger, Stephen G.

2011-01-01

We have evaluated the emergence of neural learning in the frontal eye fields (FEFSEM) and the floccular complex of the cerebellum while monkeys learned a precisely-timed change in the direction of pursuit eye movement. For each neuron, we measured the time course of changes in neural response across a learning session that comprised at least 100 repetitions of an instructive change in target direction. In both areas, the average population learning curves tracked the behavioral changes with high fidelity, consistent with possible roles in driving learning. However, the learning curves of individual neurons sometimes bore little relation to the smooth, monotonic progression of behavioral learning. In the FEFSEM, neural learning was episodic. For individual neurons, learning appeared at different times during the learning session and sometimes disappeared by the end of the session. Different FEFSEM neurons expressed maximal learning at different times relative to the acquisition of behavioral learning. In the floccular complex, many Purkinje cells acquired learned simple-spike responses according to the same time course as behavioral learning and retained their learned responses throughout the learning session. A minority of Purkinje cells acquired learned responses late in the learning session, after behavioral learning had reached an asymptote. We conclude that learning in single neurons can follow a very different time course from behavioral learning. Both the FEFSEM and the floccular complex contain representations of multiple temporal components of learning, with different neurons contributing to learning at different times during the acquisition of a learned movement. PMID:21900551

Learning curves of theta/beta neurofeedback in children with ADHD.

PubMed

Janssen, Tieme W P; Bink, Marleen; Weeda, Wouter D; Geladé, Katleen; van Mourik, Rosa; Maras, Athanasios; Oosterlaan, Jaap

2017-05-01

Neurofeedback is widely applied as non-pharmacological intervention aimed at reducing symptoms of ADHD, even though efficacy has not been unequivocally established. Neuronal changes during the neurofeedback intervention that resemble learning can provide crucial evidence for the feasibility and specificity of this intervention. A total of 38 children (aged between 7 and 13 years) with a DSM-IV-TR diagnosis of ADHD, completed on average 29 sessions of theta (4-8 Hz)/beta (13-20 Hz) neurofeedback training. Dependent variables included training-related measures as well as theta and beta power during baseline and training runs for each session. Learning effects were analyzed both within and between sessions. To further specify findings, individual learning curves were explored and correlated with behavioral changes in ADHD symptoms. Over the course of the training, there was a linear increase in participants' mean training level, highest obtained training level and the number of earned credits (range b = 0.059, -0.750, p < 0.001). Theta remained unchanged over the course of the training, while beta activity increased linearly within training sessions (b = 0.004, 95% CI = [0.0013-0.0067], p = 0.005) and over the course of the intervention (b = 0.0052, 95% CI = [0.0039-0.0065], p < 0.001). In contrast to the group analyses, significant individual learning curves were found for both theta and beta over the course of the intervention in 39 and 53%, respectively. Individual learning curves were not significantly correlated with behavioral changes. This study shows that children with ADHD can gain control over EEG states during neurofeedback, although a lack of behavioral correlates may indicate insufficient transfer to daily functioning, or to confounding reinforcement of electromyographic activity. This trial is registered at the US National Institutes of Health (ClinicalTrials.gov, ref. no: NCT01363544); https://clinicaltrials.gov/show/NCT01363544 .

Assessing Teachers' Competencies to Read and Interpret Graphs from Learning Progress Assessment: Results from Tests and Interviews

ERIC Educational Resources Information Center

Zeuch, Nina; Förster, Natalie; Souvignier, Elmar

2017-01-01

Learning progress assessment (LPA) provides formative information about effectiveness of instructional decisions. Learning curves are usually presented as graphical illustrations. However, little is known about teachers understanding and interpreting of graphically presented information. An instrument to measure competencies in reading graphs from…

Stochastic Gain in Population Dynamics

NASA Astrophysics Data System (ADS)

Traulsen, Arne; Röhl, Torsten; Schuster, Heinz Georg

2004-07-01

We introduce an extension of the usual replicator dynamics to adaptive learning rates. We show that a population with a dynamic learning rate can gain an increased average payoff in transient phases and can also exploit external noise, leading the system away from the Nash equilibrium, in a resonancelike fashion. The payoff versus noise curve resembles the signal to noise ratio curve in stochastic resonance. Seen in this broad context, we introduce another mechanism that exploits fluctuations in order to improve properties of the system. Such a mechanism could be of particular interest in economic systems.

Rate Constants for Fine-Structure Excitations in O - H Collisions with Error Bars Obtained by Machine Learning

NASA Astrophysics Data System (ADS)

Vieira, Daniel; Krems, Roman

2017-04-01

Fine-structure transitions in collisions of O(3Pj) with atomic hydrogen are an important cooling mechanism in the interstellar medium; knowledge of the rate coefficients for these transitions has a wide range of astrophysical applications. The accuracy of the theoretical calculation is limited by inaccuracy in the ab initio interaction potentials used in the coupled-channel quantum scattering calculations from which the rate coefficients can be obtained. In this work we use the latest ab initio results for the O(3Pj) + H interaction potentials to improve on previous calculations of the rate coefficients. We further present a machine-learning technique based on Gaussian Process regression to determine the sensitivity of the rate coefficients to variations of the underlying adiabatic interaction potentials. To account for the inaccuracy inherent in the ab initio calculations we compute error bars for the rate coefficients corresponding to 20% variation in each of the interaction potentials. We obtain these error bars by fitting a Gaussian Process model to a data set of potential curves and rate constants. We use the fitted model to do sensitivity analysis, determining the relative importance of individual adiabatic potential curves to a given fine-structure transition. NSERC.

Augmented Reality Image Guidance in Minimally Invasive Prostatectomy

NASA Astrophysics Data System (ADS)

Cohen, Daniel; Mayer, Erik; Chen, Dongbin; Anstee, Ann; Vale, Justin; Yang, Guang-Zhong; Darzi, Ara; Edwards, Philip'eddie'

This paper presents our work aimed at providing augmented reality (AR) guidance of robot-assisted laparoscopic surgery (RALP) using the da Vinci system. There is a good clinical case for guidance due to the significant rate of complications and steep learning curve for this procedure. Patients who were due to undergo robotic prostatectomy for organ-confined prostate cancer underwent preoperative 3T MRI scans of the pelvis. These were segmented and reconstructed to form 3D images of pelvic anatomy. The reconstructed image was successfully overlaid onto screenshots of the recorded surgery post-procedure. Surgeons who perform minimally-invasive prostatectomy took part in a user-needs analysis to determine the potential benefits of an image guidance system after viewing the overlaid images. All surgeons stated that the development would be useful at key stages of the surgery and could help to improve the learning curve of the procedure and improve functional and oncological outcomes. Establishing the clinical need in this way is a vital early step in development of an AR guidance system. We have also identified relevant anatomy from preoperative MRI. Further work will be aimed at automated registration to account for tissue deformation during the procedure, using a combination of transrectal ultrasound and stereoendoscopic video.

Analysis of DoD Travel Management: An Application of Learning Curve Theory.

DTIC Science & Technology

1982-09-01

tI AFIT Control Number LSSR 72-82 AFIT RESEARCH ASSESSMENT The purpose of this questionnaire is to determine the potential for current and future...by the equivalent value that your agency received by virtue of AFIT performing the research . Can you estimate what this research would have cost if it...In-house). 4. Often it is not possible to attach equivalent dollar values to research , although the results of the research may, in fact, be

Development of a Portable Motor Learning Laboratory (PoMLab)

PubMed Central

Shinya, Masahiro

2016-01-01

Most motor learning experiments have been conducted in a laboratory setting. In this type of setting, a huge and expensive manipulandum is frequently used, requiring a large budget and wide open space. Subjects also need to travel to the laboratory, which is a burden for them. This burden is particularly severe for patients with neurological disorders. Here, we describe the development of a novel application based on Unity3D and smart devices, e.g., smartphones or tablet devices, that can be used to conduct motor learning experiments at any time and in any place, without requiring a large budget and wide open space and without the burden of travel on subjects. We refer to our application as POrtable Motor learning LABoratory, or PoMLab. PoMLab is a multiplatform application that is available and sharable for free. We investigated whether PoMLab could be an alternative to the laboratory setting using a visuomotor rotation paradigm that causes sensory prediction error, enabling the investigation of how subjects minimize the error. In the first experiment, subjects could adapt to a constant visuomotor rotation that was abruptly applied at a specific trial. The learning curve for the first experiment could be modeled well using a state space model, a mathematical model that describes the motor leaning process. In the second experiment, subjects could adapt to a visuomotor rotation that gradually increased each trial. The subjects adapted to the gradually increasing visuomotor rotation without being aware of the visuomotor rotation. These experimental results have been reported for conventional experiments conducted in a laboratory setting, and our PoMLab application could reproduce these results. PoMLab can thus be considered an alternative to the laboratory setting. We also conducted follow-up experiments in university physical education classes. A state space model that was fit to the data obtained in the laboratory experiments could predict the learning curves obtained in the follow-up experiments. Further, we investigated the influence of vibration function, weight, and screen size on learning curves. Finally, we compared the learning curves obtained in the PoMLab experiments to those obtained in the conventional reaching experiments. The results of the in-class experiments show that PoMLab can be used to conduct motor learning experiments at any time and place. PMID:27348223

Development of a Portable Motor Learning Laboratory (PoMLab).

PubMed

Takiyama, Ken; Shinya, Masahiro

2016-01-01

Most motor learning experiments have been conducted in a laboratory setting. In this type of setting, a huge and expensive manipulandum is frequently used, requiring a large budget and wide open space. Subjects also need to travel to the laboratory, which is a burden for them. This burden is particularly severe for patients with neurological disorders. Here, we describe the development of a novel application based on Unity3D and smart devices, e.g., smartphones or tablet devices, that can be used to conduct motor learning experiments at any time and in any place, without requiring a large budget and wide open space and without the burden of travel on subjects. We refer to our application as POrtable Motor learning LABoratory, or PoMLab. PoMLab is a multiplatform application that is available and sharable for free. We investigated whether PoMLab could be an alternative to the laboratory setting using a visuomotor rotation paradigm that causes sensory prediction error, enabling the investigation of how subjects minimize the error. In the first experiment, subjects could adapt to a constant visuomotor rotation that was abruptly applied at a specific trial. The learning curve for the first experiment could be modeled well using a state space model, a mathematical model that describes the motor leaning process. In the second experiment, subjects could adapt to a visuomotor rotation that gradually increased each trial. The subjects adapted to the gradually increasing visuomotor rotation without being aware of the visuomotor rotation. These experimental results have been reported for conventional experiments conducted in a laboratory setting, and our PoMLab application could reproduce these results. PoMLab can thus be considered an alternative to the laboratory setting. We also conducted follow-up experiments in university physical education classes. A state space model that was fit to the data obtained in the laboratory experiments could predict the learning curves obtained in the follow-up experiments. Further, we investigated the influence of vibration function, weight, and screen size on learning curves. Finally, we compared the learning curves obtained in the PoMLab experiments to those obtained in the conventional reaching experiments. The results of the in-class experiments show that PoMLab can be used to conduct motor learning experiments at any time and place.

Generalization error analysis: deep convolutional neural network in mammography

NASA Astrophysics Data System (ADS)

Richter, Caleb D.; Samala, Ravi K.; Chan, Heang-Ping; Hadjiiski, Lubomir; Cha, Kenny

2018-02-01

We conducted a study to gain understanding of the generalizability of deep convolutional neural networks (DCNNs) given their inherent capability to memorize data. We examined empirically a specific DCNN trained for classification of masses on mammograms. Using a data set of 2,454 lesions from 2,242 mammographic views, a DCNN was trained to classify masses into malignant and benign classes using transfer learning from ImageNet LSVRC-2010. We performed experiments with varying amounts of label corruption and types of pixel randomization to analyze the generalization error for the DCNN. Performance was evaluated using the area under the receiver operating characteristic curve (AUC) with an N-fold cross validation. Comparisons were made between the convergence times, the inference AUCs for both the training set and the test set of the original image patches without corruption, and the root-mean-squared difference (RMSD) in the layer weights of the DCNN trained with different amounts and methods of corruption. Our experiments observed trends which revealed that the DCNN overfitted by memorizing corrupted data. More importantly, this study improved our understanding of DCNN weight updates when learning new patterns or new labels. Although we used a specific classification task with the ImageNet as example, similar methods may be useful for analysis of the DCNN learning processes, especially those that employ transfer learning for medical image analysis where sample size is limited and overfitting risk is high.

Rotation Periods and Photometric Amplitudes for Cool Stars with TESS

NASA Astrophysics Data System (ADS)

Andrews, Hannah; Dominguez, Zechariah; Johnson, Sara; Buzasi, Derek L.

2018-06-01

The original Kepler mission observed 200000 stars in the same field nearly continuously for over four years, generating an unparalleled set of stellar rotation curves and new insights into the correlation between rotation periods and photometric variability on the lower main sequence. The continuation of Kepler in the guise of K2 has allowed us to examine a stellar sample comparable in size to that observed with Kepler, but drawn from new stellar populations. However, K2 observed each field for at most three months, limiting the inferences that can be drawn, particularly for older, slower-rotating stars. The upcoming TESS spacecraft will provide light curves for perhaps two orders of magnitude more stars, but with time windows as short as 27 days. In this work, we resample Kepler light curves using the TESS observing window, and study what can be learned from high-precision light curves of such short lengths, and how to compare those results to what we have learned from Kepler.

Life Span as the Measure of Performance and Learning in a Business Gaming Simulation

ERIC Educational Resources Information Center

Thavikulwat, Precha

2012-01-01

This study applies the learning curve method of measuring learning to participants of a computer-assisted business gaming simulation that includes a multiple-life-cycle feature. The study involved 249 participants. It verified the workability of the feature and estimated the participants' rate of learning at 17.4% for every doubling of experience.…

Tools to Make Online Students and Community Partners in a Service Learning Project More "AT-EASE"--Evidence from a Finance Class

ERIC Educational Resources Information Center

Butchey, Deanne

2014-01-01

The impact of service learning as a pedagogy to ensure efficient and effective experiential learning is well recognized, but in business schools, there is a perception that a steep learning curve exists for the students, faculty, and community. We use a tool to motivate and build competence in participants of a service learning project undertaken…

Searching for prostate cancer by fully automated magnetic resonance imaging classification: deep learning versus non-deep learning.

PubMed

Wang, Xinggang; Yang, Wei; Weinreb, Jeffrey; Han, Juan; Li, Qiubai; Kong, Xiangchuang; Yan, Yongluan; Ke, Zan; Luo, Bo; Liu, Tao; Wang, Liang

2017-11-13

Prostate cancer (PCa) is a major cause of death since ancient time documented in Egyptian Ptolemaic mummy imaging. PCa detection is critical to personalized medicine and varies considerably under an MRI scan. 172 patients with 2,602 morphologic images (axial 2D T2-weighted imaging) of the prostate were obtained. A deep learning with deep convolutional neural network (DCNN) and a non-deep learning with SIFT image feature and bag-of-word (BoW), a representative method for image recognition and analysis, were used to distinguish pathologically confirmed PCa patients from prostate benign conditions (BCs) patients with prostatitis or prostate benign hyperplasia (BPH). In fully automated detection of PCa patients, deep learning had a statistically higher area under the receiver operating characteristics curve (AUC) than non-deep learning (P = 0.0007 < 0.001). The AUCs were 0.84 (95% CI 0.78-0.89) for deep learning method and 0.70 (95% CI 0.63-0.77) for non-deep learning method, respectively. Our results suggest that deep learning with DCNN is superior to non-deep learning with SIFT image feature and BoW model for fully automated PCa patients differentiation from prostate BCs patients. Our deep learning method is extensible to image modalities such as MR imaging, CT and PET of other organs.

The learning curve for narrow-band imaging in the diagnosis of precancerous gastric lesions by using Web-based video.

PubMed

Dias-Silva, Diogo; Pimentel-Nunes, Pedro; Magalhães, Joana; Magalhães, Ricardo; Veloso, Nuno; Ferreira, Carlos; Figueiredo, Pedro; Moutinho, Pedro; Dinis-Ribeiro, Mário

2014-06-01

A simplified narrow-band imaging (NBI) endoscopy classification of gastric precancerous and cancerous lesions was derived and validated in a multicenter study. This classification comes with the need for dissemination through adequate training. To address the learning curve of this classification by endoscopists with differing expertise and to assess the feasibility of a YouTube-based learning program to disseminate it. Prospective study. Five centers. Six gastroenterologists (3 trainees, 3 fully trained endoscopists [FTs]). Twenty tests provided through a Web-based program containing 10 randomly ordered NBI videos of gastric mucosa were taken. Feedback was sent 7 days after every test submission. Measures of accuracy of the NBI classification throughout the time. From the first to the last 50 videos, a learning curve was observed with a 10% increase in global accuracy, for both trainees (from 64% to 74%) and FTs (from 56% to 65%). After 200 videos, sensitivity and specificity of 80% and higher for intestinal metaplasia were observed in half the participants, and a specificity for dysplasia greater than 95%, along with a relevant likelihood ratio for a positive result of 7 to 28 and likelihood ratio for a negative result of 0.21 to 0.82, were achieved by all of the participants. No constant learning curve was observed for the identification of Helicobacter pylori gastritis and sensitivity to dysplasia. The trainees had better results in all of the parameters, except specificity for dysplasia, compared with the FTs. Globally, participants agreed that the program's structure was adequate, except on the feedback, which should have consisted of a more detailed explanation of each answer. No formal sample size estimate. A Web-based learning program could be used to teach and disseminate classifications in the endoscopy field. In this study, an NBI classification for gastric mucosal features seems to be easily learned for the identification of gastric preneoplastic lesions. Copyright © 2014 American Society for Gastrointestinal Endoscopy. Published by Mosby, Inc. All rights reserved.

Perceptual learning as improved probabilistic inference in early sensory areas.

PubMed

Bejjanki, Vikranth R; Beck, Jeffrey M; Lu, Zhong-Lin; Pouget, Alexandre

2011-05-01

Extensive training on simple tasks such as fine orientation discrimination results in large improvements in performance, a form of learning known as perceptual learning. Previous models have argued that perceptual learning is due to either sharpening and amplification of tuning curves in early visual areas or to improved probabilistic inference in later visual areas (at the decision stage). However, early theories are inconsistent with the conclusions of psychophysical experiments manipulating external noise, whereas late theories cannot explain the changes in neural responses that have been reported in cortical areas V1 and V4. Here we show that we can capture both the neurophysiological and behavioral aspects of perceptual learning by altering only the feedforward connectivity in a recurrent network of spiking neurons so as to improve probabilistic inference in early visual areas. The resulting network shows modest changes in tuning curves, in line with neurophysiological reports, along with a marked reduction in the amplitude of pairwise noise correlations.

Characterizing the Learning Curve of the VBLaST-PT© (Virtual Basic Laparoscopic Skill Trainer)

PubMed Central

Zhang, Likun; Sankaranarayanan, Ganesh; Arikatla, Venkata Sreekanth; Ahn, Woojin; Grosdemouge, Cristol; Rideout, Jesse M.; Epstein, Scott K.; De, Suvranu; Schwaitzberg, Steven D.; Jones, Daniel B.; Cao, Caroline G.L.

2013-01-01

Background and study aim Mastering laparoscopic surgical skills requires considerable time and effort. The Virtual Basic Laparoscopic Skill Trainer (VBLaST-PT©) is being developed as a computerized version of the peg transfer task of the Fundamentals of Laparoscopic Surgery (FLS) system using virtual reality technology. We assessed the learning curve of trainees on the VBLaST-PT© using the cumulative summation (CUSUM) method and compared them with those on the FLS to establish convergent validity for the VBLaST-PT©. Methods Eighteen medical students from were assigned randomly to one of three groups: control, VBLaST-training and FLS-training. The VBLaST and the FLS groups performed a total of 150 trials of the peg-transfer task over a three week period, five days a week. Their CUSUM scores were computed based on pre-defined performance criteria (junior, intermediate and senior levels). Results Of the six subjects in the VBLaST-training group, five achieved at least the “junior” level, three achieved the “intermediate” level and one achieved the “senior” level of performance criterion by the end of the 150 trials. In comparison, for the FLS group, three students achieved the “senior” criterion and all six students achieved the “intermediate” and “junior” criteria by the 150th trials. Both the VBLaST-PT© and the FLS systems showed significant skill improvement and retention, albeit with system specificity as measured by transfer of learning in the retention test: The VBLaST-trained group performed better on the VBLaST-PT© than on FLS (p=0.003), while the FLS-trained group performed better on the FLS than on VBLaST-PT© (p=0.002). Conclusion We characterized the learning curve for a virtual peg transfer task on the VBLaST-PT© and compared it with the FLS using CUSUM analysis. Subjects in both training groups showed significant improvement in skill performance, but the transfer of training between systems was not significant. PMID:23572217

Objective assessment of gynecologic laparoscopic skills using the LapSimGyn virtual reality simulator.

PubMed

Larsen, C R; Grantcharov, T; Aggarwal, R; Tully, A; Sørensen, J L; Dalsgaard, T; Ottesen, B

2006-09-01

Safe realistic training and unbiased quantitative assessment of technical skills are required for laparoscopy. Virtual reality (VR) simulators may be useful tools for training and assessing basic and advanced surgical skills and procedures. This study aimed to investigate the construct validity of the LapSimGyn VR simulator, and to determine the learning curves of gynecologists with different levels of experience. For this study, 32 gynecologic trainees and consultants (juniors or seniors) were allocated into three groups: novices (0 advanced laparoscopic procedures), intermediate level (>20 and <60 procedures), and experts (>100 procedures). All performed 10 sets of simulations consisting of three basic skill tasks and an ectopic pregnancy program. The simulations were carried out on 3 days within a maximum period of 2 weeks. Assessment of skills was based on time, economy of movement, and error parameters measured by the simulator. The data showed that expert gynecologists performed significantly and consistently better than intermediate and novice gynecologists. The learning curves differed significantly between the groups, showing that experts start at a higher level and more rapidly reach the plateau of their learning curve than do intermediate and novice groups of surgeons. The LapSimGyn VR simulator package demonstrates construct validity on both the basic skills module and the procedural gynecologic module for ectopic pregnancy. Learning curves can be obtained, but to reach the maximum performance for the more complex tasks, 10 repetitions do not seem sufficient at the given task level and settings. LapSimGyn also seems to be flexible and widely accepted by the users.

Diffusion of robotics into clinical practice in the United States: process, patient safety, learning curves, and the public health.

PubMed

Mirheydar, Hossein S; Parsons, J Kellogg

2013-06-01

Robotic technology disseminated into urological practice without robust comparative effectiveness data. To review the diffusion of robotic surgery into urological practice. We performed a comprehensive literature review focusing on diffusion patterns, patient safety, learning curves, and comparative costs for robotic radical prostatectomy, partial nephrectomy, and radical cystectomy. Robotic urologic surgery diffused in patterns typical of novel technology spreading among practicing surgeons. Robust evidence-based data comparing outcomes of robotic to open surgery were sparse. Although initial Level 3 evidence for robotic prostatectomy observed complication outcomes similar to open prostatectomy, subsequent population-based Level 2 evidence noted an increased prevalence of adverse patient safety events and genitourinary complications among robotic patients during the early years of diffusion. Level 2 evidence indicated comparable to improved patient safety outcomes for robotic compared to open partial nephrectomy and cystectomy. Learning curve recommendations for robotic urologic surgery have drawn exclusively on Level 4 evidence and subjective, non-validated metrics. The minimum number of cases required to achieve competency for robotic prostatectomy has increased to unrealistically high levels. Most comparative cost-analyses have demonstrated that robotic surgery is significantly more expensive than open or laparoscopic surgery. Evidence-based data are limited but suggest an increased prevalence of adverse patient safety events for robotic prostatectomy early in the national diffusion period. Learning curves for robotic urologic surgery are subjective and based on non-validated metrics. The urological community should develop rigorous, evidence-based processes by which future technological innovations may diffuse in an organized and safe manner.

Learning curve of transumbilical single incision laparoscopic cholecystectomy (SILS): a preliminary study of 80 selected patients with benign gallbladder diseases.

PubMed

Qiu, Zhengjun; Sun, Jing; Pu, Ying; Jiang, Tao; Cao, Jun; Wu, Weidong

2011-09-01

Transumbilical single incision laparoscopic surgery (SILS) is a new laparoscopic procedure in which only one transumbilical incision is made, demonstrated as a scarless procedure. Here we report a single-center preliminary experience of transumbilical single incision laparoscopic cholecystectomy (SILC) in the treatment of benign gallbladder diseases, defining a single surgeon's learning curve. A total of 80 patients underwent SILC successfully by a single experienced laparoscopic surgeon. The operation was performed following the routine LC procedure. Then the perioperative demographics were recorded and the operative time was used to define the learning curve. The study group included 27 male and 53 female patients with gallstones (56 cases), cholesterol polyps (16 cases), an adenomatous polyp (3 cases), adenomyomatosis (1 case), or complex diseases (4 cases), and all consented to undergo SILC. No patient was converted to normal LC or open surgery. There were no perioperative port-related or surgical complications. The average operative time was 46.9 ± 14.6 min. The average postoperative hospital stay was 1.8 ± 1.3 days. The learning curve of the SILC procedures for this series of selected patients confirmed that SILC is a feasible, safe, and effective approach to the treatment of benign gallbladder diseases. For experienced laparoscopic surgeons, SILC is an easy and safe procedure. Patients benefit from milder pain, a lower incidence of port-related complications, better cosmesis, and fast recovery. The SILC procedure may become another option for the treatment of benign gallbladder diseases for selected patients.

Balloon dilation of the eustachian tube in a cadaver model: technical considerations, learning curve, and potential barriers.

PubMed

McCoul, Edward D; Singh, Ameet; Anand, Vijay K; Tabaee, Abtin

2012-04-01

The surgical management options for eustachian tube dysfunction have historically been limited. The goal of the current study was to evaluate the technical considerations, learning curve, and potential barriers for balloon dilation of the eustachian tube (BDET) as an alternative treatment modality. Prospective preclinical trial of BDET in a cadaver model. A novel balloon catheter device was used for eustachian tube dilation. Twenty-four BDET procedures were performed by three independent rhinologists with no prior experience with the procedure (eight procedures per surgeon). The duration and number of attempts of the individual steps and overall procedure were recorded. Endoscopic examination of the eustachian tube was performed after each procedure, and the surgeon was asked to rate the subjective difficulty on a five-point scale. Successful completion of the procedure occurred in each case. The overall mean duration of the procedure was 284 seconds, and a mean number of 1.15 attempts were necessary to perform the individual steps. The mean subjective procedure difficulty was noted as somewhat easy. Statistically shorter duration and subjectively easier procedure were noted in the second compared to the first half of the series, indicating a favorable learning curve. Linear fissuring within the eustachian tube lumen without submucosal disruption (nine procedures, 37%) and with submucosal disruption (five procedures, 21%) were noted. The significance of these physical findings is unclear. Preclinical testing of BDET is associated with favorable duration, learning curve, and overall ease of completion. Clinical trials are necessary to evaluate safety and efficacy. Copyright © 2012 The American Laryngological, Rhinological, and Otological Society, Inc.

Transanal total mesorectal excision for rectal cancer: evaluation of the learning curve.

PubMed

Koedam, T W A; Veltcamp Helbach, M; van de Ven, P M; Kruyt, Ph M; van Heek, N T; Bonjer, H J; Tuynman, J B; Sietses, C

2018-04-01

Transanal total mesorectal excision (TaTME) provides an excellent view of the resection margins for rectal cancer from below, but is challenging due to few anatomical landmarks. During implementation of this technique, patient safety and optimal outcomes need to be ensured. The aim of this study was to evaluate the learning curve of TaTME in patients with rectal cancer in order to optimize future training programs. All consecutive patients after TaTME for rectal cancer between February 2012 and January 2017 were included in a single-center database. Influence of surgical experience on major postoperative complications, leakage rate and operating time was evaluated using cumulative sum charts and the splitting model. Correction for potential case-mix differences was performed. Over a period of 60 months, a total of 138 patients were included in this study. Adjusted for case-mix, improvement in postoperative outcomes was clearly seen after the first 40 patients, showing a decrease in major postoperative complications from 47.5 to 17.5% and leakage rate from 27.5 to 5%. Mean operating time (42 min) and conversion rate (from 10% to zero) was lower after transition to a two-team approach, but neither endpoint decreased with experience. Readmission and reoperation rates were not influenced by surgical experience. The learning curve of TaTME affected major (surgical) postoperative complications for the first 40 patients. A two-team approach decreased operative time and conversion rate. When implementing this new technique, a thorough teaching and supervisory program is recommended to shorten the learning curve and improve the clinical outcomes of the first patients.

Adoption of Robotic vs Fluoroscopic Guidance in Total Hip Arthroplasty: Is Acetabular Positioning Improved in the Learning Curve?

PubMed

Kamara, Eli; Robinson, Jonathon; Bas, Marcel A; Rodriguez, Jose A; Hepinstall, Matthew S

2017-01-01

Acetabulum positioning affects dislocation rates, component impingement, bearing surface wear rates, and need for revision surgery. Novel techniques purport to improve the accuracy and precision of acetabular component position, but may have a significant learning curve. Our aim was to assess whether adopting robotic or fluoroscopic techniques improve acetabulum positioning compared to manual total hip arthroplasty (THA) during the learning curve. Three types of THAs were compared in this retrospective cohort: (1) the first 100 fluoroscopically guided direct anterior THAs (fluoroscopic anterior [FA]) done by a surgeon learning the anterior approach, (2) the first 100 robotic-assisted posterior THAs done by a surgeon learning robotic-assisted surgery (robotic posterior [RP]), and (3) the last 100 manual posterior (MP) THAs done by each surgeon (200 THAs) before adoption of novel techniques. Component position was measured on plain radiographs. Radiographic measurements were taken by 2 blinded observers. The percentage of hips within the surgeons' "target zone" (inclination, 30°-50°; anteversion, 10°-30°) was calculated, along with the percentage within the "safe zone" of Lewinnek (inclination, 30°-50°; anteversion, 5°-25°) and Callanan (inclination, 30°-45°; anteversion, 5°-25°). Relative risk (RR) and absolute risk reduction (ARR) were calculated. Variances (square of the standard deviations) were used to describe the variability of cup position. Seventy-six percentage of MP THAs were within the surgeons' target zone compared with 84% of FA THAs and 97% of RP THAs. This difference was statistically significant, associated with a RR reduction of 87% (RR, 0.13 [0.04-0.40]; P < .01; ARR, 21%; number needed to treat, 5) for RP compared to MP THAs. Compared to FA THAs, RP THAs were associated with a RR reduction of 81% (RR, 0.19 [0.06-0.62]; P < .01; ARR, 13%; number needed to treat, 8). Variances were lower for acetabulum inclination and anteversion in RP THAs (14.0 and 19.5) as compared to the MP (37.5 and 56.3) and FA (24.5 and 54.6) groups. These differences were statistically significant (P < .01). Adoption of robotic techniques delivers significant and immediate improvement in the precision of acetabular component positioning during the learning curve. While fluoroscopy has been shown to be beneficial with experience, a learning curve exists before precision improves significantly. Copyright © 2016 Elsevier Inc. All rights reserved.

A Glance at Institutional Support for Faculty Teaching in an Online Learning Environment

ERIC Educational Resources Information Center

Lion, Robert W.; Stark, Gary

2010-01-01

With continued advances in web-based learning, colleges and universities strive to meet the needs and interests of students, faculty, and staff. New instructional technologies have at least one thing in common: the learning curve associated with users becoming adept. Mastery requires significant time and attention. Providing the best quality…

Curved Saccade Trajectories Reveal Conflicting Predictions in Associative Learning

ERIC Educational Resources Information Center

Koenig, Stephan; Lachnit, Harald

2011-01-01

We report how the trajectories of saccadic eye movements are affected by memory interference acquired during associative learning. Human participants learned to perform saccadic choice responses based on the presentation of arbitrary central cues A, B, AC, BC, AX, BY, X, and Y that were trained to predict the appearance of a peripheral target…

Verbal Knowledge, Working Memory, and Processing Speed as Predictors of Verbal Learning in Older Adults

ERIC Educational Resources Information Center

Rast, Philippe

2011-01-01

The present study aimed at modeling individual differences in a verbal learning task by means of a latent structured growth curve approach based on an exponential function that yielded 3 parameters: initial recall, learning rate, and asymptotic performance. Three cognitive variables--speed of information processing, verbal knowledge, working…

Measuring the surgical 'learning curve': methods, variables and competency.

PubMed

Khan, Nuzhath; Abboudi, Hamid; Khan, Mohammed Shamim; Dasgupta, Prokar; Ahmed, Kamran

2014-03-01

To describe how learning curves are measured and what procedural variables are used to establish a 'learning curve' (LC). To assess whether LCs are a valuable measure of competency. A review of the surgical literature pertaining to LCs was conducted using the Medline and OVID databases. Variables should be fully defined and when possible, patient-specific variables should be used. Trainee's prior experience and level of supervision should be quantified; the case mix and complexity should ideally be constant. Logistic regression may be used to control for confounding variables. Ideally, a learning plateau should reach a predefined/expert-derived competency level, which should be fully defined. When the group splitting method is used, smaller cohorts should be used in order to narrow the range of the LC. Simulation technology and competence-based objective assessments may be used in training and assessment in LC studies. Measuring the surgical LC has potential benefits for patient safety and surgical education. However, standardisation in the methods and variables used to measure LCs is required. Confounding variables, such as participant's prior experience, case mix, difficulty of procedures and level of supervision, should be controlled. Competency and expert performance should be fully defined. © 2013 The Authors. BJU International © 2013 BJU International.

Modeling cascading diffusion of new energy technologies: case study of residential solid oxide fuel cells in the US and internationally.

PubMed

Herron, Seth; Williams, Eric

2013-08-06

Subsidy programs for new energy technologies are motivated by the experience curve: increased adoption of a technology leads to learning and economies of scale that lower costs. Geographic differences in fuel prices and climate lead to large variability in the economic performance of energy technologies. The notion of cascading diffusion is that regions with favorable economic conditions serve as the basis to build scale and reduce costs so that the technology becomes attractive in new regions. We develop a model of cascading diffusion and implement via a case study of residential solid oxide fuel cells (SOFCs) for combined heating and power. We consider diffusion paths within the U.S. and internationally. We construct market willingness-to-pay curves and estimate future manufacturing costs via an experience curve. Combining market and cost results, we find that for rapid cost reductions (learning rate = 25%), a modest public subsidy can make SOFC investment profitable for 20-160 million households. If cost reductions are slow however (learning rate = 15%), residential SOFCs may not become economically competitive. Due to higher energy prices in some countries, international diffusion is more favorable than domestic, mitigating much of the uncertainty in the learning rate.

PHOTOMETRIC SUPERNOVA CLASSIFICATION WITH MACHINE LEARNING

DOE Office of Scientific and Technical Information (OSTI.GOV)

Lochner, Michelle; Peiris, Hiranya V.; Lahav, Ofer

Automated photometric supernova classification has become an active area of research in recent years in light of current and upcoming imaging surveys such as the Dark Energy Survey (DES) and the Large Synoptic Survey Telescope, given that spectroscopic confirmation of type for all supernovae discovered will be impossible. Here, we develop a multi-faceted classification pipeline, combining existing and new approaches. Our pipeline consists of two stages: extracting descriptive features from the light curves and classification using a machine learning algorithm. Our feature extraction methods vary from model-dependent techniques, namely SALT2 fits, to more independent techniques that fit parametric models tomore » curves, to a completely model-independent wavelet approach. We cover a range of representative machine learning algorithms, including naive Bayes, k -nearest neighbors, support vector machines, artificial neural networks, and boosted decision trees (BDTs). We test the pipeline on simulated multi-band DES light curves from the Supernova Photometric Classification Challenge. Using the commonly used area under the curve (AUC) of the Receiver Operating Characteristic as a metric, we find that the SALT2 fits and the wavelet approach, with the BDTs algorithm, each achieve an AUC of 0.98, where 1 represents perfect classification. We find that a representative training set is essential for good classification, whatever the feature set or algorithm, with implications for spectroscopic follow-up. Importantly, we find that by using either the SALT2 or the wavelet feature sets with a BDT algorithm, accurate classification is possible purely from light curve data, without the need for any redshift information.« less

Development of the Mathematics of Learning Curve Models for Evaluating Small Modular Reactor Economics

DOE Office of Scientific and Technical Information (OSTI.GOV)

Harrison, T. J.

2014-02-01

The cost of nuclear power is a straightforward yet complicated topic. It is straightforward in that the cost of nuclear power is a function of the cost to build the nuclear power plant, the cost to operate and maintain it, and the cost to provide fuel for it. It is complicated in that some of those costs are not necessarily known, introducing uncertainty into the analysis. For large light water reactor (LWR)-based nuclear power plants, the uncertainty is mainly contained within the cost of construction. The typical costs of operations and maintenance (O&M), as well as fuel, are well knownmore » based on the current fleet of LWRs. However, the last currently operating reactor to come online was Watts Bar 1 in May 1996; thus, the expected construction costs for gigawatt (GW)-class reactors in the United States are based on information nearly two decades old. Extrapolating construction, O&M, and fuel costs from GW-class LWRs to LWR-based small modular reactors (SMRs) introduces even more complication. The per-installed-kilowatt construction costs for SMRs are likely to be higher than those for the GW-class reactors based on the property of the economy of scale. Generally speaking, the economy of scale is the tendency for overall costs to increase slower than the overall production capacity. For power plants, this means that doubling the power production capacity would be expected to cost less than twice as much. Applying this property in the opposite direction, halving the power production capacity would be expected to cost more than half as much. This can potentially make the SMRs less competitive in the electricity market against the GW-class reactors, as well as against other power sources such as natural gas and subsidized renewables. One factor that can potentially aid the SMRs in achieving economic competitiveness is an economy of numbers, as opposed to the economy of scale, associated with learning curves. The basic concept of the learning curve is that the more a new process is repeated, the more efficient the process can be made. Assuming that efficiency directly relates to cost means that the more a new process is repeated successfully and efficiently, the less costly the process can be made. This factor ties directly into the factory fabrication and modularization aspect of the SMR paradigm—manufacturing serial, standardized, identical components for use in nuclear power plants can allow the SMR industry to use the learning curves to predict and optimize deployment costs.« less

Learning curve for gastric cancer patients with laparoscopy-assisted distal gastrectomy: 6-year experience from a single institution in western China.

PubMed

Zhao, Lin-Yong; Zhang, Wei-Han; Sun, Yan; Chen, Xin-Zu; Yang, Kun; Liu, Kai; Chen, Xiao-Long; Wang, Yi-Gao; Song, Xiao-Hai; Xue, Lian; Zhou, Zong-Guang; Hu, Jian-Kun

2016-09-01

Laparoscopy-assisted distal gastrectomy (LADG) is widely used for gastric cancer (GC) patients nowadays. This study aimed to investigate the time trend of outcomes so as to describe the learning curve for GC patients with LADG at a single medical institution in western China over a 6-year period.A total of 246 consecutive GC patients with LADG were divided into 5 groups (group A: 46 patients from 2006 to 2007; group B: 47 patients in 2008; group C: 49 patients in 2009; group D: 73 patients in 2010; and group E: 31 patients in 2011). All surgeries were conducted by the same surgeon. Comparative analyses were successively performed by Mann-Whitney U test or Student t test among the 5 different groups for the clinical data, including clinicopathologic characteristics, surgical parameters, postoperative course, and survival outcomes, through which the learning curve was described.There were no differences in the baseline information among the 5 groups (P > 0.05), and the proportion of advanced GC patients with LADG slightly increased from 58.7% to 77.4% during the 6 years. Besides, the proportion of D2/D2+ lymphadenectomy and the number of retrieved lymph nodes gradually grew from 60.9% to 80.6% and from 20.0 to 28.8, respectively. In addition, the operation time decreased from 299.2 to 267.8 minutes, while the estimated blood loss dropped from 175.2 to 146.8 mL. Furthermore, some surgical parameters (surgical duration and blood loss) and postoperative course (such as postoperative complications, the time to ambulation, to first flatus, and to first liquid intake as well as the length of hospital stay) were all observed to be significantly different between group A and other groups (P < 0.05), illustrating a similar downward trend and remaining stable to form a plateau after 46 cases in group A. However, no difference on overall survival was found among these 5 groups, and multivariate analysis indicated that factors, such as age, tumor differentiation, tumor size, and T stage as well as N stage, were independent prognostic factors for patients with LADG.Improvement on surgical parameters and postoperative course can be seen over the past years, and the cutoff value of the learning curve of LADG for surgeons with rich experience in open operation might be 46 cases.

Crashes and near-crashes on horizontal curves along rural two-lane highways: Analysis of naturalistic driving data.

PubMed

Wang, Bo; Hallmark, Shauna; Savolainen, Peter; Dong, Jing

2017-12-01

Prior research has shown the probability of a crash occurring on horizontal curves to be significantly higher than on similar tangent segments, and a disproportionally higher number of curve-related crashes occurred in rural areas. Challenges arise when analyzing the safety of horizontal curves due to imprecision in integrating information as to the temporal and spatial characteristics of each crash with specific curves. The second Strategic Highway Research Program(SHRP 2) conducted a large-scale naturalistic driving study (NDS),which provides a unique opportunity to better understand the contributing factors leading to crash or near-crash events. This study utilizes high-resolution behavioral data from the NDS to identify factors associated with 108 safety critical events (i.e., crashes or near-crashes) on rural two-lane curves. A case-control approach is utilized wherein these events are compared to 216 normal, baseline-driving events. The variables examined in this study include driver demographic characteristics, details of the traffic environment and roadway geometry, as well as driver behaviors such as in-vehicle distractions. Logistic regression models are estimated to discern those factors affecting the likelihood of a driver being crash-involved. These factors include high-risk behaviors, such as speeding and visual distractions, as well as curve design elements and other roadway characteristics such as pavement surface conditions. This paper successfully integrated driver behavior, vehicle characteristics, and roadway environments into the same model. Logistic regression model was found to be an effective way to investigate crash risks using naturalistic driving data. This paper revealed a number of contributing factors to crashes on rural two-lane curves, which has important implications in traffic safety policy and curve geometry design. This paper also discussed limitations and lessons learned from working with the SHRP 2 NDS data. It will benefit future researchers who work with similar type of data. Copyright © 2017 National Safety Council and Elsevier Ltd. All rights reserved.

Early results after laparoscopic Roux-en-Y gastric bypass: effect of the learning curve

PubMed Central

Andrew, Christopher G.; Hanna, Wael; Look, Didier; McLean, Alexander P.H.; Christou, Nicolas V.

2006-01-01

Introduction This study was performed to evaluate the safety and short-term efficacy of laparoscopic Roux-en-Y gastric bypass (LRYGB) for morbid obesity and to describe the relation between learning curve and short-term outcomes. Methods We collected a prospective database on the first 201 consecutive patients who underwent LRYGB by a single university-based, experienced bariatric surgeon over 24 months. We divided patients into 3 consecutive groups of 67 patients for analysis (Group 1, Group 2 and Group 3). Results The mean patient age was 37 (standard deviation [SD] 9) years; mean body mass index (BMI) was 49.2 (SD 8.3) kg/m2. BMI was similar in Groups 1 and 2 (mean 47.1, SD 5.9 and mean 48.7, SD 8.9 kg/m2) but increased in Group 3 (mean 52, SD 9.7 kg/m2, p < 0.01). Operative time decreased from 145 (SD 30) minutes in Group 1 to 114 (SD 24) minutes in Group 2 (p < 0.01) and was maintained at 119 (SD 23) minutes in Group 3. Early and late complication rates were 14.9% and 12.4%, respectively. Leak rates decreased from 6.0% in the first group to 1.5% in Groups 2 and 3, but they did not reach statistical significance. Anastomotic stricture rates decreased from 11.9% in Group 1 to 3.0% in Group 2 (p < 0.01). Overall excess weight loss for the entire series was 31.5% (SD 11.9%), 54.5% (SD 14.1%), 77.1% (SD 18.5%) and 82.1% (SD 17.5%) at 3, 6, 12 and 18 months, respectively. Conclusion LRYGB can be performed with acceptable morbidity and short-term results during the learning curve. In our series, operative time and anastomotic stricture rates decreased with experience, despite an increase in mean BMI. PMID:17234071

Deep 3D convolution neural network for CT brain hemorrhage classification

NASA Astrophysics Data System (ADS)

Jnawali, Kamal; Arbabshirani, Mohammad R.; Rao, Navalgund; Patel, Alpen A.

2018-02-01

Intracranial hemorrhage is a critical conditional with the high mortality rate that is typically diagnosed based on head computer tomography (CT) images. Deep learning algorithms, in particular, convolution neural networks (CNN), are becoming the methodology of choice in medical image analysis for a variety of applications such as computer-aided diagnosis, and segmentation. In this study, we propose a fully automated deep learning framework which learns to detect brain hemorrhage based on cross sectional CT images. The dataset for this work consists of 40,367 3D head CT studies (over 1.5 million 2D images) acquired retrospectively over a decade from multiple radiology facilities at Geisinger Health System. The proposed algorithm first extracts features using 3D CNN and then detects brain hemorrhage using the logistic function as the last layer of the network. Finally, we created an ensemble of three different 3D CNN architectures to improve the classification accuracy. The area under the curve (AUC) of the receiver operator characteristic (ROC) curve of the ensemble of three architectures was 0.87. Their results are very promising considering the fact that the head CT studies were not controlled for slice thickness, scanner type, study protocol or any other settings. Moreover, the proposed algorithm reliably detected various types of hemorrhage within the skull. This work is one of the first applications of 3D CNN trained on a large dataset of cross sectional medical images for detection of a critical radiological condition

UNSUPERVISED TRANSIENT LIGHT CURVE ANALYSIS VIA HIERARCHICAL BAYESIAN INFERENCE

DOE Office of Scientific and Technical Information (OSTI.GOV)

Sanders, N. E.; Soderberg, A. M.; Betancourt, M., E-mail: nsanders@cfa.harvard.edu

2015-02-10

Historically, light curve studies of supernovae (SNe) and other transient classes have focused on individual objects with copious and high signal-to-noise observations. In the nascent era of wide field transient searches, objects with detailed observations are decreasing as a fraction of the overall known SN population, and this strategy sacrifices the majority of the information contained in the data about the underlying population of transients. A population level modeling approach, simultaneously fitting all available observations of objects in a transient sub-class of interest, fully mines the data to infer the properties of the population and avoids certain systematic biases. Wemore » present a novel hierarchical Bayesian statistical model for population level modeling of transient light curves, and discuss its implementation using an efficient Hamiltonian Monte Carlo technique. As a test case, we apply this model to the Type IIP SN sample from the Pan-STARRS1 Medium Deep Survey, consisting of 18,837 photometric observations of 76 SNe, corresponding to a joint posterior distribution with 9176 parameters under our model. Our hierarchical model fits provide improved constraints on light curve parameters relevant to the physical properties of their progenitor stars relative to modeling individual light curves alone. Moreover, we directly evaluate the probability for occurrence rates of unseen light curve characteristics from the model hyperparameters, addressing observational biases in survey methodology. We view this modeling framework as an unsupervised machine learning technique with the ability to maximize scientific returns from data to be collected by future wide field transient searches like LSST.« less

Training anesthesiology residents in providing anesthesia for awake craniotomy: learning curves and estimate of needed case load.

PubMed

Bilotta, Federico; Titi, Luca; Lanni, Fabiana; Stazi, Elisabetta; Rosa, Giovanni

2013-08-01

To measure the learning curves of residents in anesthesiology in providing anesthesia for awake craniotomy, and to estimate the case load needed to achieve a "good-excellent" level of competence. Prospective study. Operating room of a university hospital. 7 volunteer residents in anesthesiology. Residents underwent a dedicated training program of clinical characteristics of anesthesia for awake craniotomy. The program was divided into three tasks: local anesthesia, sedation-analgesia, and intraoperative hemodynamic management. The learning curve for each resident for each task was recorded over 10 procedures. Quantitative assessment of the individual's ability was based on the resident's self-assessment score and the attending anesthesiologist's judgment, and rated by modified 12 mm Likert scale, reported ability score visual analog scale (VAS). This ability VAS score ranged from 1 to 12 (ie, very poor, mild, moderate, sufficient, good, excellent). The number of requests for advice also was recorded (ie, resident requests for practical help and theoretical notions to accomplish the procedures). Each task had a specific learning rate; the number of procedures necessary to achieve "good-excellent" ability with confidence, as determined by the recorded results, were 10 procedures for local anesthesia, 15 to 25 procedures for sedation-analgesia, and 20 to 30 procedures for intraoperative hemodynamic management. Awake craniotomy is an approach used increasingly in neuroanesthesia. A dedicated training program based on learning specific tasks and building confidence with essential features provides "good-excellent" ability. © 2013 Elsevier Inc. All rights reserved.

Optical Coherence Tomography Machine Learning Classifiers for Glaucoma Detection: A Preliminary Study

PubMed Central

Burgansky-Eliash, Zvia; Wollstein, Gadi; Chu, Tianjiao; Ramsey, Joseph D.; Glymour, Clark; Noecker, Robert J.; Ishikawa, Hiroshi; Schuman, Joel S.

2007-01-01

Purpose Machine-learning classifiers are trained computerized systems with the ability to detect the relationship between multiple input parameters and a diagnosis. The present study investigated whether the use of machine-learning classifiers improves optical coherence tomography (OCT) glaucoma detection. Methods Forty-seven patients with glaucoma (47 eyes) and 42 healthy subjects (42 eyes) were included in this cross-sectional study. Of the glaucoma patients, 27 had early disease (visual field mean deviation [MD] ≥ −6 dB) and 20 had advanced glaucoma (MD < −6 dB). Machine-learning classifiers were trained to discriminate between glaucomatous and healthy eyes using parameters derived from OCT output. The classifiers were trained with all 38 parameters as well as with only 8 parameters that correlated best with the visual field MD. Five classifiers were tested: linear discriminant analysis, support vector machine, recursive partitioning and regression tree, generalized linear model, and generalized additive model. For the last two classifiers, a backward feature selection was used to find the minimal number of parameters that resulted in the best and most simple prediction. The cross-validated receiver operating characteristic (ROC) curve and accuracies were calculated. Results The largest area under the ROC curve (AROC) for glaucoma detection was achieved with the support vector machine using eight parameters (0.981). The sensitivity at 80% and 95% specificity was 97.9% and 92.5%, respectively. This classifier also performed best when judged by cross-validated accuracy (0.966). The best classification between early glaucoma and advanced glaucoma was obtained with the generalized additive model using only three parameters (AROC = 0.854). Conclusions Automated machine classifiers of OCT data might be useful for enhancing the utility of this technology for detecting glaucomatous abnormality. PMID:16249492

The U-Curve of E-Learning: Course Website and Online Video Use in Blended and Distance Learning

ERIC Educational Resources Information Center

Geri, Nitza; Gafni, Ruti; Winer, Amir

2014-01-01

Procrastination is a common challenge for students. While course Websites and online video lectures enable studying anytime, anywhere, and expand learning opportunities, their availability may increase procrastination by making it easier for students to defer until tomorrow. This research used Google Analytics to examine temporal use patterns of…

What Is the Future of Learning in Canada?

ERIC Educational Resources Information Center

Canadian Council on Learning, 2011

2011-01-01

In its final report to Canadians, the Canadian Council on Learning (CCL) reveals that Canada is slipping down the international learning curve. The needs in this area are stark. The potential rewards are enormous. But Canada is falling behind competitor countries and economies. It is on the wrong road and must make a dramatic change in the course…

Learning Curve Analyses in Neurodevelopmental Disorders: Are Children with Autism Spectrum Disorder Truly Visual Learners?

ERIC Educational Resources Information Center

Erdodi, Laszlo; Lajiness-O'Neill, Renee; Schmitt, Thomas A.

2013-01-01

Visual and auditory verbal learning using a selective reminding format was studied in a mixed clinical sample of children with autism spectrum disorder (ASD) (n = 42), attention-deficit hyperactivity disorder (n = 83), velocardiofacial syndrome (n = 17) and neurotypicals (n = 38) using the Test of Memory and Learning to (1) more thoroughly…

Inter-Labeler and Intra-Labeler Variability of Condition Severity Classification Models Using Active and Passive Learning Methods

PubMed Central

Nissim, Nir; Shahar, Yuval; Boland, Mary Regina; Tatonetti, Nicholas P; Elovici, Yuval; Hripcsak, George; Moskovitch, Robert

2018-01-01

Background and Objectives Labeling instances by domain experts for classification is often time consuming and expensive. To reduce such labeling efforts, we had proposed the application of active learning (AL) methods, introduced our CAESAR-ALE framework for classifying the severity of clinical conditions, and shown its significant reduction of labeling efforts. The use of any of three AL methods (one well known [SVM-Margin], and two that we introduced [Exploitation and Combination_XA]) significantly reduced (by 48% to 64%) condition labeling efforts, compared to standard passive (random instance-selection) SVM learning. Furthermore, our new AL methods achieved maximal accuracy using 12% fewer labeled cases than the SVM-Margin AL method. However, because labelers have varying levels of expertise, a major issue associated with learning methods, and AL methods in particular, is how to best to use the labeling provided by a committee of labelers. First, we wanted to know, based on the labelers’ learning curves, whether using AL methods (versus standard passive learning methods) has an effect on the Intra-labeler variability (within the learning curve of each labeler) and inter-labeler variability (among the learning curves of different labelers). Then, we wanted to examine the effect of learning (either passively or actively) from the labels created by the majority consensus of a group of labelers. Methods We used our CAESAR-ALE framework for classifying the severity of clinical conditions, the three AL methods and the passive learning method, as mentioned above, to induce the classifications models. We used a dataset of 516 clinical conditions and their severity labeling, represented by features aggregated from the medical records of 1.9 million patients treated at Columbia University Medical Center. We analyzed the variance of the classification performance within (intra-labeler), and especially among (inter-labeler) the classification models that were induced by using the labels provided by seven labelers. We also compared the performance of the passive and active learning models when using the consensus label. Results The AL methods produced, for the models induced from each labeler, smoother Intra-labeler learning curves during the training phase, compared to the models produced when using the passive learning method. The mean standard deviation of the learning curves of the three AL methods over all labelers (mean: 0.0379; range: [0.0182 to 0.0496]), was significantly lower (p = 0.049) than the Intra-labeler standard deviation when using the passive learning method (mean: 0.0484; range: [0.0275 to 0.0724). Using the AL methods resulted in a lower mean Inter-labeler AUC standard deviation among the AUC values of the labelers’ different models during the training phase, compared to the variance of the induced models’ AUC values when using passive learning. The Inter-labeler AUC standard deviation, using the passive learning method (0.039), was almost twice as high as the Inter-labeler standard deviation using our two new AL methods (0.02 and 0.019, respectively). The SVM-Margin AL method resulted in an Inter-labeler standard deviation (0.029) that was higher by almost 50% than that of our two AL methods. The difference in the inter-labeler standard deviation between the passive learning method and the SVM-Margin learning method was significant (p = 0.042). The difference between the SVM-Margin and Exploitation method was insignificant (p = 0.29), as was the difference between the Combination_XA and Exploitation methods (p = 0.67). Finally, using the consensus label led to a learning curve that had a higher mean intra-labeler variance, but resulted eventually in an AUC that was at least as high as the AUC achieved using the gold standard label and that was always higher than the expected mean AUC of a randomly selected labeler, regardless of the choice of learning method (including a passive learning method). Using a paired t-test, the difference between the intra-labeler AUC standard deviation when using the consensus label, versus that value when using the other two labeling strategies, was significant only when using the passive learning method (p = 0.014), but not when using any of the three AL methods. Conclusions The use of AL methods, (a) reduces intra-labeler variability in the performance of the induced models during the training phase, and thus reduces the risk of halting the process at a local minimum that is significantly different in performance from the rest of the learned models; and (b) reduces Inter-labeler performance variance, and thus reduces the dependence on the use of a particular labeler. In addition, the use of a consensus label, agreed upon by a rather uneven group of labelers, might be at least as good as using the gold standard labeler, who might not be available, and certainly better than randomly selecting one of the group’s individual labelers. Finally, using the AL methods when provided by the consensus label reduced the intra-labeler AUC variance during the learning phase, compared to using passive learning. PMID:28456512

Inter-labeler and intra-labeler variability of condition severity classification models using active and passive learning methods.

PubMed

Nissim, Nir; Shahar, Yuval; Elovici, Yuval; Hripcsak, George; Moskovitch, Robert

2017-09-01

Labeling instances by domain experts for classification is often time consuming and expensive. To reduce such labeling efforts, we had proposed the application of active learning (AL) methods, introduced our CAESAR-ALE framework for classifying the severity of clinical conditions, and shown its significant reduction of labeling efforts. The use of any of three AL methods (one well known [SVM-Margin], and two that we introduced [Exploitation and Combination_XA]) significantly reduced (by 48% to 64%) condition labeling efforts, compared to standard passive (random instance-selection) SVM learning. Furthermore, our new AL methods achieved maximal accuracy using 12% fewer labeled cases than the SVM-Margin AL method. However, because labelers have varying levels of expertise, a major issue associated with learning methods, and AL methods in particular, is how to best to use the labeling provided by a committee of labelers. First, we wanted to know, based on the labelers' learning curves, whether using AL methods (versus standard passive learning methods) has an effect on the Intra-labeler variability (within the learning curve of each labeler) and inter-labeler variability (among the learning curves of different labelers). Then, we wanted to examine the effect of learning (either passively or actively) from the labels created by the majority consensus of a group of labelers. We used our CAESAR-ALE framework for classifying the severity of clinical conditions, the three AL methods and the passive learning method, as mentioned above, to induce the classifications models. We used a dataset of 516 clinical conditions and their severity labeling, represented by features aggregated from the medical records of 1.9 million patients treated at Columbia University Medical Center. We analyzed the variance of the classification performance within (intra-labeler), and especially among (inter-labeler) the classification models that were induced by using the labels provided by seven labelers. We also compared the performance of the passive and active learning models when using the consensus label. The AL methods: produced, for the models induced from each labeler, smoother Intra-labeler learning curves during the training phase, compared to the models produced when using the passive learning method. The mean standard deviation of the learning curves of the three AL methods over all labelers (mean: 0.0379; range: [0.0182 to 0.0496]), was significantly lower (p=0.049) than the Intra-labeler standard deviation when using the passive learning method (mean: 0.0484; range: [0.0275-0.0724). Using the AL methods resulted in a lower mean Inter-labeler AUC standard deviation among the AUC values of the labelers' different models during the training phase, compared to the variance of the induced models' AUC values when using passive learning. The Inter-labeler AUC standard deviation, using the passive learning method (0.039), was almost twice as high as the Inter-labeler standard deviation using our two new AL methods (0.02 and 0.019, respectively). The SVM-Margin AL method resulted in an Inter-labeler standard deviation (0.029) that was higher by almost 50% than that of our two AL methods The difference in the inter-labeler standard deviation between the passive learning method and the SVM-Margin learning method was significant (p=0.042). The difference between the SVM-Margin and Exploitation method was insignificant (p=0.29), as was the difference between the Combination_XA and Exploitation methods (p=0.67). Finally, using the consensus label led to a learning curve that had a higher mean intra-labeler variance, but resulted eventually in an AUC that was at least as high as the AUC achieved using the gold standard label and that was always higher than the expected mean AUC of a randomly selected labeler, regardless of the choice of learning method (including a passive learning method). Using a paired t-test, the difference between the intra-labeler AUC standard deviation when using the consensus label, versus that value when using the other two labeling strategies, was significant only when using the passive learning method (p=0.014), but not when using any of the three AL methods. The use of AL methods, (a) reduces intra-labeler variability in the performance of the induced models during the training phase, and thus reduces the risk of halting the process at a local minimum that is significantly different in performance from the rest of the learned models; and (b) reduces Inter-labeler performance variance, and thus reduces the dependence on the use of a particular labeler. In addition, the use of a consensus label, agreed upon by a rather uneven group of labelers, might be at least as good as using the gold standard labeler, who might not be available, and certainly better than randomly selecting one of the group's individual labelers. Finally, using the AL methods: when provided by the consensus label reduced the intra-labeler AUC variance during the learning phase, compared to using passive learning. Copyright © 2017 Elsevier B.V. All rights reserved.

Practically perfect: learning by doing at AVS congress.

PubMed

2017-02-18

It has been some time since Cambridge vet school last hosted the annual AVS congress, which meant that this year's congress committee faced a steep learning curve. However, as Gill Harris reports, it rose to the occasion. British Veterinary Association.

Machine Learning Algorithms Outperform Conventional Regression Models in Predicting Development of Hepatocellular Carcinoma

PubMed Central

Singal, Amit G.; Mukherjee, Ashin; Elmunzer, B. Joseph; Higgins, Peter DR; Lok, Anna S.; Zhu, Ji; Marrero, Jorge A; Waljee, Akbar K

2015-01-01

Background Predictive models for hepatocellular carcinoma (HCC) have been limited by modest accuracy and lack of validation. Machine learning algorithms offer a novel methodology, which may improve HCC risk prognostication among patients with cirrhosis. Our study's aim was to develop and compare predictive models for HCC development among cirrhotic patients, using conventional regression analysis and machine learning algorithms. Methods We enrolled 442 patients with Child A or B cirrhosis at the University of Michigan between January 2004 and September 2006 (UM cohort) and prospectively followed them until HCC development, liver transplantation, death, or study termination. Regression analysis and machine learning algorithms were used to construct predictive models for HCC development, which were tested on an independent validation cohort from the Hepatitis C Antiviral Long-term Treatment against Cirrhosis (HALT-C) Trial. Both models were also compared to the previously published HALT-C model. Discrimination was assessed using receiver operating characteristic curve analysis and diagnostic accuracy was assessed with net reclassification improvement and integrated discrimination improvement statistics. Results After a median follow-up of 3.5 years, 41 patients developed HCC. The UM regression model had a c-statistic of 0.61 (95%CI 0.56-0.67), whereas the machine learning algorithm had a c-statistic of 0.64 (95%CI 0.60–0.69) in the validation cohort. The machine learning algorithm had significantly better diagnostic accuracy as assessed by net reclassification improvement (p<0.001) and integrated discrimination improvement (p=0.04). The HALT-C model had a c-statistic of 0.60 (95%CI 0.50-0.70) in the validation cohort and was outperformed by the machine learning algorithm (p=0.047). Conclusion Machine learning algorithms improve the accuracy of risk stratifying patients with cirrhosis and can be used to accurately identify patients at high-risk for developing HCC. PMID:24169273

Learning curves and impact of previous operative experience on performance on a virtual reality simulator to test laparoscopic surgical skills.

PubMed

Grantcharov, Teodor P; Bardram, Linda; Funch-Jensen, Peter; Rosenberg, Jacob

2003-02-01

The study was carried out to analyze the learning rate for laparoscopic skills on a virtual reality training system and to establish whether the simulator was able to differentiate between surgeons with different laparoscopic experience. Forty-one surgeons were divided into three groups according to their experience in laparoscopic surgery: masters (group 1, performed more than 100 cholecystectomies), intermediates (group 2, between 15 and 80 cholecystectomies), and beginners (group 3, fewer than 10 cholecystectomies) were included in the study. The participants were tested on the Minimally Invasive Surgical Trainer-Virtual Reality (MIST-VR) 10 consecutive times within a 1-month period. Assessment of laparoscopic skills included time, errors, and economy of hand movement, measured by the simulator. The learning curves regarding time reached plateau after the second repetition for group 1, the fifth repetition for group 2, and the seventh repetition for group 3 (Friedman's tests P <0.05). Experienced surgeons did not improve their error or economy of movement scores (Friedman's tests, P >0.2) indicating the absence of a learning curve for these parameters. Group 2 error scores reached plateau after the first repetition, and group 3 after the fifth repetition. Group 2 improved their economy of movement score up to the third repetition and group 3 up to the sixth repetition (Friedman's tests, P <0.05). Experienced surgeons (group 1) demonstrated best performance parameters, followed by group 2 and group 3 (Mann-Whitney test P <0.05). Different learning curves existed for surgeons with different laparoscopic background. The familiarization rate on the simulator was proportional to the operative experience of the surgeons. Experienced surgeons demonstrated best laparoscopic performance on the simulator, followed by those with intermediate experience and the beginners. These differences indicate that the scoring system of MIST-VR is sensitive and specific to measuring skills relevant for laparoscopic surgery.

Using Supervised Machine Learning to Classify Real Alerts and Artifact in Online Multisignal Vital Sign Monitoring Data.

PubMed

Chen, Lujie; Dubrawski, Artur; Wang, Donghan; Fiterau, Madalina; Guillame-Bert, Mathieu; Bose, Eliezer; Kaynar, Ata M; Wallace, David J; Guttendorf, Jane; Clermont, Gilles; Pinsky, Michael R; Hravnak, Marilyn

2016-07-01

The use of machine-learning algorithms to classify alerts as real or artifacts in online noninvasive vital sign data streams to reduce alarm fatigue and missed true instability. Observational cohort study. Twenty-four-bed trauma step-down unit. Two thousand one hundred fifty-three patients. Noninvasive vital sign monitoring data (heart rate, respiratory rate, peripheral oximetry) recorded on all admissions at 1/20 Hz, and noninvasive blood pressure less frequently, and partitioned data into training/validation (294 admissions; 22,980 monitoring hours) and test sets (2,057 admissions; 156,177 monitoring hours). Alerts were vital sign deviations beyond stability thresholds. A four-member expert committee annotated a subset of alerts (576 in training/validation set, 397 in test set) as real or artifact selected by active learning, upon which we trained machine-learning algorithms. The best model was evaluated on test set alerts to enact online alert classification over time. The Random Forest model discriminated between real and artifact as the alerts evolved online in the test set with area under the curve performance of 0.79 (95% CI, 0.67-0.93) for peripheral oximetry at the instant the vital sign first crossed threshold and increased to 0.87 (95% CI, 0.71-0.95) at 3 minutes into the alerting period. Blood pressure area under the curve started at 0.77 (95% CI, 0.64-0.95) and increased to 0.87 (95% CI, 0.71-0.98), whereas respiratory rate area under the curve started at 0.85 (95% CI, 0.77-0.95) and increased to 0.97 (95% CI, 0.94-1.00). Heart rate alerts were too few for model development. Machine-learning models can discern clinically relevant peripheral oximetry, blood pressure, and respiratory rate alerts from artifacts in an online monitoring dataset (area under the curve > 0.87).

[Evaluation of the learning curve of residents in localizing a phantom target with ultrasonography].

PubMed

Dessieux, T; Estebe, J-P; Bloc, S; Mercadal, L; Ecoffey, C

2008-10-01

Few information are available regarding the learning curve in ultrasonography and even less for ultrasound-guided regional anesthesia. This study aimed to evaluate in a training program the learning curve on a phantom of 12 residents novice in ultrasonography. Twelve trainees inexperienced in ultrasonography were given introductory training consisting of didactic formation on the various components of the portable ultrasound machine (i.e. on/off button, gain, depth, resolution, and image storage). Then, students performed three trials, in two sets of increased difficulty, at executing these predefined tasks: adjustments of the machine, then localization of a small plastic piece introduced into roasting pork (3 cm below the surface). At the end of the evaluation, the residents were asked to insert a 22 G needle into an exact predetermined target (i.e. point of fascia intersection). The progression of the needle was continuously controlled by ultrasound visualization using injection of a small volume of water (needle perpendicular to the longitudinal plane of the ultrasound beam). Two groups of two different examiners evaluated for each three trials the skill of the residents (quality, time to perform the machine adjustments, to localize the plastic target, and to hydrolocalize, and volume used for hydrolocalization). After each trial, residents evaluated their performance using a difficulty scale (0: easy to 10: difficult). All residents performed the adjustments from the last trial of each set, with a learning curve observed in terms of duration. Localization of the plastic piece was achieved by all residents at the 6th trial, with a shorter duration of localization. Hydrolocalization was achieved after the 4th trial by all subjects. Difficulty scale was correlated to the number of trials. All these results were independent of the experience of residents in regional anesthesia. Four trials were necessary to adjust correctly the machine, to localize a target, and to complete hydrolocalization. Ultrasonography in regional anesthesia seems to be a fast-learning technique, using this kind of practical training.

Analyzing microtomography data with Python and the scikit-image library.

PubMed

Gouillart, Emmanuelle; Nunez-Iglesias, Juan; van der Walt, Stéfan

2017-01-01

The exploration and processing of images is a vital aspect of the scientific workflows of many X-ray imaging modalities. Users require tools that combine interactivity, versatility, and performance. scikit-image is an open-source image processing toolkit for the Python language that supports a large variety of file formats and is compatible with 2D and 3D images. The toolkit exposes a simple programming interface, with thematic modules grouping functions according to their purpose, such as image restoration, segmentation, and measurements. scikit-image users benefit from a rich scientific Python ecosystem that contains many powerful libraries for tasks such as visualization or machine learning. scikit-image combines a gentle learning curve, versatile image processing capabilities, and the scalable performance required for the high-throughput analysis of X-ray imaging data.

Computerised curve deconvolution of TL/OSL curves using a popular spreadsheet program.

PubMed

Afouxenidis, D; Polymeris, G S; Tsirliganis, N C; Kitis, G

2012-05-01

This paper exploits the possibility of using commercial software for thermoluminescence and optically stimulated luminescence curve deconvolution analysis. The widely used software package Microsoft Excel, with the Solver utility has been used to perform deconvolution analysis to both experimental and reference glow curves resulted from the GLOw Curve ANalysis INtercomparison project. The simple interface of this programme combined with the powerful Solver utility, allows the analysis of complex stimulated luminescence curves into their components and the evaluation of the associated luminescence parameters.

Automated analysis of individual sperm cells using stain-free interferometric phase microscopy and machine learning.

PubMed

Mirsky, Simcha K; Barnea, Itay; Levi, Mattan; Greenspan, Hayit; Shaked, Natan T

2017-09-01

Currently, the delicate process of selecting sperm cells to be used for in vitro fertilization (IVF) is still based on the subjective, qualitative analysis of experienced clinicians using non-quantitative optical microscopy techniques. In this work, a method was developed for the automated analysis of sperm cells based on the quantitative phase maps acquired through use of interferometric phase microscopy (IPM). Over 1,400 human sperm cells from 8 donors were imaged using IPM, and an algorithm was designed to digitally isolate sperm cell heads from the quantitative phase maps while taking into consideration both the cell 3D morphology and contents, as well as acquire features describing sperm head morphology. A subset of these features was used to train a support vector machine (SVM) classifier to automatically classify sperm of good and bad morphology. The SVM achieves an area under the receiver operating characteristic curve of 88.59% and an area under the precision-recall curve of 88.67%, as well as precisions of 90% or higher. We believe that our automatic analysis can become the basis for objective and automatic sperm cell selection in IVF. © 2017 International Society for Advancement of Cytometry. © 2017 International Society for Advancement of Cytometry.

Mentor Tutoring: An Efficient Method for Teaching Laparoscopic Colorectal Surgical Skills in a General Hospital.

PubMed

Ichikawa, Nobuki; Homma, Shigenori; Yoshida, Tadashi; Ohno, Yosuke; Kawamura, Hideki; Wakizaka, Kazuki; Nakanishi, Kazuaki; Kazui, Keizo; Iijima, Hiroaki; Shomura, Hiroki; Funakoshi, Tohru; Nakano, Shiro; Taketomi, Akinobu

2017-12-01

We retrospectively assessed the efficacy of our mentor tutoring system for teaching laparoscopic colorectal surgical skills in a general hospital. A series of 55 laparoscopic colectomies performed by 1 trainee were evaluated. Next, the learning curves for high anterior resection performed by the trainee (n=20) were compared with those of a self-trained surgeon (n=19). Cumulative sum analysis and multivariate regression analyses showed that 38 completed cases were needed to reduce the operative time. In high anterior resection, the mean operative times were significantly shorter after the seventh average for the tutored surgeon compared with that for the self-trained surgeon. In cumulative sum charting, the curve reached a plateau by the seventh case for the tutored surgeon, but continued to increase for the self-trained surgeon. Mentor tutoring effectively teaches laparoscopic colorectal surgical skills in a general hospital setting.

PyKE3: data analysis tools for NASA's Kepler, K2, and TESS missions

NASA Astrophysics Data System (ADS)

Hedges, Christina L.; Cardoso, Jose Vinicius De Miranda; Barentsen, Geert; Gully-Santiago, Michael A.; Cody, Ann Marie; Barclay, Thomas; Still, Martin; BAY AREA ENVIRONMENTAL RESEARCH IN

2018-01-01

The PyKE package is a set of easy to use tools for working with Kepler/K2 data. This includes tools to correct light curves for cotrending basis vectors, turn the raw Target Pixel File data into motion corrected light curves, check for exoplanet false positives and run new PSF photometry. We are now releasing PyKE 3, which is compatible with Python 3, is pip installable and no longer depends on PyRAF. Tools are available both as Python routines and from the command line. New tutorials are available and under construction for users to learn about Kepler and K2 data and how to best use it for their science goals. PyKE is open source and welcomes contributions from the community. Routines and more information are available on the PyKE repository on GitHub.

Learning Curves for Ultrasound Assessment of Lumbar Puncture Insertion Sites: When is Competency Established?

PubMed

Rankin, Jessica H; Elkhunovich, Marsha A; Rangarajan, Vijayeta; Chilstrom, Mikaela; Mailhot, Tom

2016-07-01

Ultrasound (US) can be used to improve lumbar puncture (LP) success. How to achieve competency in LP US has not been defined. Cumulative sum statistics (CUSUM) characterized competency acquisition in other skills. Identify the learning curve for 80% success rate in LP US insertion site (IS) identification among pediatric emergency medicine fellows. This prospective study took place in a single pediatric emergency department. Fellows with limited ultrasound experience received didactics, training, and three proctored examinations. Skills were evaluated in three 2-h sessions: using US, subjects identified LP ISs on a convenience sample of patients ages 0-20 years old. Subjects' IS markings were compared to markings by an expert, an emergency US fellowship-trained attending. Successful IS identification was defined as markings within 2 mm or 5 mm of the expert mark in infants and older children, respectively. A second expert marked 17 cases for interrater agreement. CUSUM was used to analyze individual learning curves. Five fellows evaluated 72 patients (mean age 11.4 years [SD = 4, range 3-20], mean body mass index 20.5 [SD = 4.4, range 13.1-37.7]) over a 3-month period. Mean number of attempts per fellow was 14.4 ± 3.1 (R 11-19); mean time to landmark identification was 72 ± 46 s (R 27-240). The two experts demonstrated 100% observed agreement. Aggregate success rate for all fellows was 75% (54/72). Four fellows showed learning curves that trended toward, but did not achieve, the acceptable success rate of 80%. Nineteen attempts are insufficient among fellows to achieve competency in US-guided LP IS identification. Copyright © 2016 Elsevier Inc. All rights reserved.

[HoLEP learning curve: Toward a standardised formation and a team strategy].

PubMed

Baron, M; Nouhaud, F-X; Delcourt, C; Grise, P; Pfister, C; Cornu, J-N; Sibert, L

2016-09-01

Holmium laser enucleation of prostate (HoLEP) is renowned for the difficulty of its learning curve. Our aim was to evaluate the interest of a three-step tutorial in the HoLEP learning curve, in a university center. It is a retrospective, monocentric study of the 82 first procedures done consecutively by the same operator with a proctoring in early experience and after 40 procedures. For all patients were noted: enucleation efficiency (g/min), morcellation efficiency (g/min), percentage of enucleated tissue (enucleated tissue/adenome weigth evaluated by ultrasonography. g/g), perioperative morbidity (Clavien), length of hospital stay, length of urinary drainage, functional outcomes at short and middle term (Qmax, post-void residual volume [PVR], QOL scores and IPSS at 3 and 6months). Enucleation and morcellation efficiency were significantly higher after the second proctoring (0.87 vs 0.44g/min; P<0.0001 and 4.2 vs 3.37g/min, P=0.038, respectively) so as the prostatic volume (43.5 vs 68.1mL, P=0.0001). Percentage of enucleated tissue was higher in the second group, however, the difference was not significant (69.5% vs 80.4%, P=0.03). Per- and postoperative complications, hospital length of stay, urinary drainage length and functional results at 3 and 6months were not significantly different. The learning curve did not interfere with functional results. The second proctoring was essential to us in order to grasp the technique. These data underlined the necessity of a pedagogic reflexion in order to built a standardized formation technique to the HoLEP. 4. Copyright © 2016 Elsevier Masson SAS. All rights reserved.

Learning curve for peroral endoscopic myotomy

PubMed Central

El Zein, Mohamad; Kumbhari, Vivek; Ngamruengphong, Saowanee; Carson, Kathryn A.; Stein, Ellen; Tieu, Alan; Chaveze, Yamile; Ismail, Amr; Dhalla, Sameer; Clarke, John; Kalloo, Anthony; Canto, Marcia Irene; Khashab, Mouen A.

2016-01-01

Background and study aims: Although peroral endoscopic myotomy (POEM) is being performed more frequently, the learning curve for gastroenterologists performing the procedure has not been well studied. The aims of this study were to define the learning curve for POEM and determine which preoperative and intraoperative factors predict the time that will be taken to complete the procedure and its different steps. Patients and methods: Consecutive patients who underwent POEM performed by a single expert gastroenterologist for the treatment of achalasia or spastic esophageal disorders were included. The POEM procedure was divided into four steps: mucosal entry, submucosal tunneling, myotomy, and closure. Nonlinear regression was used to determine the POEM learning plateau and calculate the learning rate. Results: A total of 60 consecutive patients underwent POEM in an endoscopy suite. The median length of procedure (LOP) was 88 minutes (range 36 – 210), and the mean (± standard deviation [SD]) LOP per centimeter of myotomy was 9 ± 5 minutes. The total operative time decreased significantly as experience increased (P < 0.001), with a “learning plateau” at 102 minutes and a “learning rate” of 13 cases. The mucosal entry, tunneling, and closure times decreased significantly with experience (P < 0.001). The myotomy time showed no significant decrease with experience (P = 0.35). When the mean (± SD) total procedure times for the learning phase and the corresponding comparator groups were compared, a statistically significant difference was observed between procedures 11 – 15 and procedures 16 – 20 (15.5 ± 2.4 min/cm and 10.1 ± 2.7 min/cm, P = 0.01) but not thereafter. A higher case number was significantly associated with a decreased LOP (P < 0.001). Conclusion: In this single-center retrospective study, the minimum threshold number of cases required for an expert interventional endoscopist performing POEM to reach a plateau approached 13. PMID:27227118

Searching for exoplanets using artificial intelligence

NASA Astrophysics Data System (ADS)

Pearson, Kyle A.; Palafox, Leon; Griffith, Caitlin A.

2018-02-01

In the last decade, over a million stars were monitored to detect transiting planets. Manual interpretation of potential exoplanet candidates is labor intensive and subject to human error, the results of which are difficult to quantify. Here we present a new method of detecting exoplanet candidates in large planetary search projects which, unlike current methods uses a neural network. Neural networks, also called "deep learning" or "deep nets" are designed to give a computer perception into a specific problem by training it to recognize patterns. Unlike past transit detection algorithms deep nets learn to recognize planet features instead of relying on hand-coded metrics that humans perceive as the most representative. Our convolutional neural network is capable of detecting Earth-like exoplanets in noisy time-series data with a greater accuracy than a least-squares method. Deep nets are highly generalizable allowing data to be evaluated from different time series after interpolation without compromising performance. As validated by our deep net analysis of Kepler light curves, we detect periodic transits consistent with the true period without any model fitting. Our study indicates that machine learning will facilitate the characterization of exoplanets in future analysis of large astronomy data sets.

Kepler

NASA Technical Reports Server (NTRS)

Howell, Steve B.

2011-01-01

The NASA Kepler mission recently announced over 1200 exoplanet candidates. While some are common Hot Jupiters, a large number are Neptune size and smaller, transit depths suggest sizes down to the radius of Earth. The Kepler project has a fairly high confidence that most of these candidates are real exoplanets. Many analysis steps and lessons learned from Kepler light curves are used during the vetting process. This talk will cover some new results in the areas of stellar variability, solar systems with multiple planets, and how transit-like signatures are vetted for false positives, especially those indicative of small planets.

Curved-line search algorithm for ab initio atomic structure relaxation

NASA Astrophysics Data System (ADS)

Chen, Zhanghui; Li, Jingbo; Li, Shushen; Wang, Lin-Wang

2017-09-01

Ab initio atomic relaxations often take large numbers of steps and long times to converge, especially when the initial atomic configurations are far from the local minimum or there are curved and narrow valleys in the multidimensional potentials. An atomic relaxation method based on on-the-flight force learning and a corresponding curved-line search algorithm is presented to accelerate this process. Results demonstrate the superior performance of this method for metal and magnetic clusters when compared with the conventional conjugate-gradient method.

Differentiation of arterioles from venules in mouse histology images using machine learning

NASA Astrophysics Data System (ADS)

Elkerton, J. S.; Xu, Yiwen; Pickering, J. G.; Ward, Aaron D.

2016-03-01

Analysis and morphological comparison of arteriolar and venular networks are essential to our understanding of multiple diseases affecting every organ system. We have developed and evaluated the first fully automatic software system for differentiation of arterioles from venules on high-resolution digital histology images of the mouse hind limb immunostained for smooth muscle α-actin. Classifiers trained on texture and morphologic features by supervised machine learning provided excellent classification accuracy for differentiation of arterioles and venules, achieving an area under the receiver operating characteristic curve of 0.90 and balanced false-positive and false-negative rates. Feature selection was consistent across cross-validation iterations, and a small set of three features was required to achieve the reported performance, suggesting potential generalizability of the system. This system eliminates the need for laborious manual classification of the hundreds of microvessels occurring in a typical sample, and paves the way for high-throughput analysis the arteriolar and venular networks in the mouse.

Minimally invasive transforaminal lumbar interbody fusion for spondylolisthesis and degenerative spondylosis: 5-year results.

PubMed

Park, Yung; Ha, Joong Won; Lee, Yun Tae; Sung, Na Young

2014-06-01

Multiple studies have reported favorable short-term results after treatment of spondylolisthesis and other degenerative lumbar diseases with minimally invasive transforaminal lumbar interbody fusion. However, to our knowledge, results at a minimum of 5 years have not been reported. We determined (1) changes to the Oswestry Disability Index, (2) frequency of radiographic fusion, (3) complications and reoperations, and (4) the learning curve associated with minimally invasive transforaminal lumbar interbody fusion at minimum 5-year followup. We reviewed our first 124 patients who underwent minimally invasive transforaminal lumbar interbody fusion to treat low-grade spondylolisthesis and degenerative lumbar diseases and did not need a major deformity correction. This represented 63% (124 of 198) of the transforaminal lumbar interbody fusion procedures we performed for those indications during the study period (2003-2007). Eighty-three (67%) patients had complete 5-year followup. Plain radiographs and CT scans were evaluated by two reviewers. Trends of surgical time, blood loss, and hospital stay over time were examined by logarithmic curve fit-regression analysis to evaluate the learning curve. At 5 years, mean Oswestry Disability Index improved from 60 points preoperatively to 24 points and 79 of 83 patients (95%) had improvement of greater than 10 points. At 5 years, 67 of 83 (81%) achieved radiographic fusion, including 64 of 72 patients (89%) who had single-level surgery. Perioperative complications occurred in 11 of 124 patients (9%), and another surgical procedure was performed in eight of 124 patients (6.5%) involving the index level and seven of 124 patients (5.6%) at adjacent levels. There were slowly decreasing trends of surgical time and hospital stay only in single-level surgery and almost no change in intraoperative blood loss over time, suggesting a challenging learning curve. Oswestry Disability Index scores improved for patients with spondylolisthesis and degenerative lumbar diseases treated with minimally invasive transforaminal lumbar interbody fusion at minimum 5-year followup. We suggest this procedure is reasonable for properly selected patients with these indications; however, traditional approaches should still be performed for patients with high-grade spondylolisthesis, patients with a severely collapsed disc space and no motion seen on the dynamic radiographs, patients who need multilevel decompression and arthrodesis, and patients with kyphoscoliosis needing correction. Level IV, therapeutic study. See the Instructions for Authors for a complete description of levels of evidence.

Using a High-Fidelity Patient Simulator with First-Year Medical Students to Facilitate Learning of Cardiovascular Function Curves

ERIC Educational Resources Information Center

Harris, David M.; Ryan, Kathleen; Rabuck, Cynthia

2012-01-01

Students are relying on technology for learning more than ever, and educators need to adapt to facilitate student learning. High-fidelity patient simulators (HFPS) are usually reserved for the clinical years of medical education and are geared to improve clinical decision skills, teamwork, and patient safety. Finding ways to incorporate HFPS into…

How to develop renewable power in China? A cost-effective perspective.

PubMed

Cong, Rong-Gang; Shen, Shaochuan

2014-01-01

To address the problems of climate change and energy security, Chinese government strived to develop renewable power as an important alternative of conventional electricity. In this paper, the learning curve model is employed to describe the decreasing unit investment cost due to accumulated installed capacity; the technology diffusion model is used to analyze the potential of renewable power. Combined with the investment cost, the technology potential, and scenario analysis of China social development in the future, we develop the Renewable Power Optimization Model (RPOM) to analyze the optimal development paths of three sources of renewable power from 2009 to 2020 in a cost-effective way. Results show that (1) the optimal accumulated installed capacities of wind power, solar power, and biomass power will reach 169000, 20000, and 30000 MW in 2020; (2) the developments of renewable power show the intermittent feature; (3) the unit investment costs of wind power, solar power, and biomass power will be 4500, 11500, and 5700 Yuan/KW in 2020; (4) the discounting effect dominates the learning curve effect for solar and biomass powers; (5) the rise of on-grid ratio of renewable power will first promote the development of wind power and then solar power and biomass power.

How to Develop Renewable Power in China? A Cost-Effective Perspective

PubMed Central

2014-01-01

To address the problems of climate change and energy security, Chinese government strived to develop renewable power as an important alternative of conventional electricity. In this paper, the learning curve model is employed to describe the decreasing unit investment cost due to accumulated installed capacity; the technology diffusion model is used to analyze the potential of renewable power. Combined with the investment cost, the technology potential, and scenario analysis of China social development in the future, we develop the Renewable Power Optimization Model (RPOM) to analyze the optimal development paths of three sources of renewable power from 2009 to 2020 in a cost-effective way. Results show that (1) the optimal accumulated installed capacities of wind power, solar power, and biomass power will reach 169000, 20000, and 30000 MW in 2020; (2) the developments of renewable power show the intermittent feature; (3) the unit investment costs of wind power, solar power, and biomass power will be 4500, 11500, and 5700 Yuan/KW in 2020; (4) the discounting effect dominates the learning curve effect for solar and biomass powers; (5) the rise of on-grid ratio of renewable power will first promote the development of wind power and then solar power and biomass power. PMID:24578672

Novel Uses of Video to Accelerate the Surgical Learning Curve.

PubMed

Ibrahim, Andrew M; Varban, Oliver A; Dimick, Justin B

2016-04-01

Surgeons are under enormous pressure to continually improve and learn new surgical skills. Novel uses of surgical video in the preoperative, intraoperative, and postoperative setting are emerging to accelerate the learning curve of surgical skill and minimize harm to patients. In the preoperative setting, social media outlets provide a valuable platform for surgeons to collaborate and plan for difficult operative cases. Live streaming of video has allowed for intraoperative telementoring. Finally, postoperative use of video has provided structure for peer coaching to evaluate and improve surgical skill. Applying these approaches into practice is becoming easier as most of our surgical platforms (e.g., laparoscopic, and endoscopy) now have video recording technology built in and video editing software has become more user friendly. Future applications of video technology are being developed, including possible integration into accreditation and board certification.

Advances in Laparoscopic Colorectal Surgery.

PubMed

Parker, James Michael; Feldmann, Timothy F; Cologne, Kyle G

2017-06-01

Laparoscopic colorectal surgery has now become widely adopted for the treatment of colorectal neoplasia, with steady increases in utilization over the past 15 years. Common minimally invasive techniques include multiport laparoscopy, single-incision laparoscopy, and hand-assisted laparoscopy, with the choice of technique depending on several patient and surgeon factors. Laparoscopic colorectal surgery involves a robust learning curve, and fellowship training often lays the foundation for a high-volume laparoscopic practice. This article provides a summary of the various techniques for laparoscopic colorectal surgery, including operative steps, the approach to difficult patients, and the learning curve for proficiency. Copyright © 2017 Elsevier Inc. All rights reserved.

Brief periods of NREM sleep do not promote early offline gains but subsequent on-task performance in motor skill learning.

PubMed

Maier, Jonathan G; Piosczyk, Hannah; Holz, Johannes; Landmann, Nina; Deschler, Christoph; Frase, Lukas; Kuhn, Marion; Klöppel, Stefan; Spiegelhalder, Kai; Sterr, Annette; Riemann, Dieter; Feige, Bernd; Voderholzer, Ulrich; Nissen, Christoph

2017-11-01

Sleep modulates motor learning, but its detailed impact on performance curves remains to be fully characterized. This study aimed to further determine the impact of brief daytime periods of NREM sleep on 'offline' (task discontinuation after initial training) and 'on-task' (performance within the test session) changes in motor skill performance (finger tapping task). In a mixed design (combined parallel group and repeated measures) sleep laboratory study (n=17 'active' wake vs. sleep, n=19 'passive' wake vs. sleep), performance curves were assessed prior to and after a 90min period containing either sleep, active or passive wakefulness. We observed a highly significant, but state- (that is, sleep/wake)-independent early offline gain and improved on-task performance after sleep in comparison to wakefulness. Exploratory curve fitting suggested that the observed sleep effect most likely emerged from an interaction of training-induced improvement and detrimental 'time-on-task' processes, such as fatigue. Our results indicate that brief periods of NREM sleep do not promote early offline gains but subsequent on-task performance in motor skill learning. Copyright © 2017 Elsevier Inc. All rights reserved.

A variation reduction allocation model for quality improvement to minimize investment and quality costs by considering suppliers’ learning curve

NASA Astrophysics Data System (ADS)

Rosyidi, C. N.; Jauhari, WA; Suhardi, B.; Hamada, K.

2016-02-01

Quality improvement must be performed in a company to maintain its product competitiveness in the market. The goal of such improvement is to increase the customer satisfaction and the profitability of the company. In current practice, a company needs several suppliers to provide the components in assembly process of a final product. Hence quality improvement of the final product must involve the suppliers. In this paper, an optimization model to allocate the variance reduction is developed. Variation reduction is an important term in quality improvement for both manufacturer and suppliers. To improve suppliers’ components quality, the manufacturer must invest an amount of their financial resources in learning process of the suppliers. The objective function of the model is to minimize the total cost consists of investment cost, and quality costs for both internal and external quality costs. The Learning curve will determine how the employee of the suppliers will respond to the learning processes in reducing the variance of the component.

Digital mammographic tumor classification using transfer learning from deep convolutional neural networks.

PubMed

Huynh, Benjamin Q; Li, Hui; Giger, Maryellen L

2016-07-01

Convolutional neural networks (CNNs) show potential for computer-aided diagnosis (CADx) by learning features directly from the image data instead of using analytically extracted features. However, CNNs are difficult to train from scratch for medical images due to small sample sizes and variations in tumor presentations. Instead, transfer learning can be used to extract tumor information from medical images via CNNs originally pretrained for nonmedical tasks, alleviating the need for large datasets. Our database includes 219 breast lesions (607 full-field digital mammographic images). We compared support vector machine classifiers based on the CNN-extracted image features and our prior computer-extracted tumor features in the task of distinguishing between benign and malignant breast lesions. Five-fold cross validation (by lesion) was conducted with the area under the receiver operating characteristic (ROC) curve as the performance metric. Results show that classifiers based on CNN-extracted features (with transfer learning) perform comparably to those using analytically extracted features [area under the ROC curve [Formula: see text

What is the learning curve for robotic-assisted pedicle screw placement in spine surgery?

PubMed

Hu, Xiaobang; Lieberman, Isador H

2014-06-01

Some early studies with robotic-assisted pedicle screw implantation have suggested these systems increase accuracy of screw placement. However, the relationship between the success rate of screw placement and the learning curve of this new technique has not been evaluated. We determined whether, as a function of surgeon experience, (1) the success rate of robotic-assisted pedicle screw placement improved, (2) the frequency of conversion from robotic to manual screw placement decreased, and (3) the frequency of malpositioned screws decreased. Between June 2010 and August 2012, the senior surgeon (IHL) performed 174 posterior spinal procedures using pedicle screws, 162 of which were attempted with robotic assistance. The use of the robotic system was aborted in 12 of the 162 procedures due to technical issues (registration failure, software crash, etc). The robotic system was successfully used in the remaining 150 procedures. These were the first procedures performed with the robot by the senior surgeon, and in this study, we divided the early learning curve into five groups: Group 1 (Patients 1-30), Group 2 (Patients 31-60), Group 3 (Patients 61-90), Group 4 (Patients 91-120), and Group 5 (Patients 121-150). One hundred twelve patients (75%) had spinal deformity and 80 patients (53%) had previous spine surgery. The accuracy of screw placement in the groups was assessed based on intraoperative biplanar fluoroscopy and postoperative radiographs. The results from these five groups were compared to determine the effect on the learning curve. The numbers of attempted pedicle screw placements were 359, 312, 349, 359, and 320 in Groups 1 to 5, respectively. The rates of successfully placed screws using robotic guidance were 82%, 93%, 91%, 95%, and 93% in Groups 1 to 5. The rates of screws converted to manual placement were 17%, 7%, 8%, 4%, and 7%. Of the robotically placed screws, the screw malposition rates were 0.8%, 0.3%, 1.4%, 0.8%, and 0%. The rate of successfully placed pedicle screws improved with increasing experience. The rate of the screws that were converted to manual placement decreased with increasing experience. The frequency of screw malposition was similar over the learning curve at 0% to 1.4%. Future studies will need to determine whether this finding is generalizable to others. Level III, therapeutic study. See the Instructions for Authors for a complete description of levels of evidence.

On the necessity of U-shaped learning.

PubMed

Carlucci, Lorenzo; Case, John

2013-01-01

A U-shaped curve in a cognitive-developmental trajectory refers to a three-step process: good performance followed by bad performance followed by good performance once again. U-shaped curves have been observed in a wide variety of cognitive-developmental and learning contexts. U-shaped learning seems to contradict the idea that learning is a monotonic, cumulative process and thus constitutes a challenge for competing theories of cognitive development and learning. U-shaped behavior in language learning (in particular in learning English past tense) has become a central topic in the Cognitive Science debate about learning models. Antagonist models (e.g., connectionism versus nativism) are often judged on their ability of modeling or accounting for U-shaped behavior. The prior literature is mostly occupied with explaining how U-shaped behavior occurs. Instead, we are interested in the necessity of this kind of apparently inefficient strategy. We present and discuss a body of results in the abstract mathematical setting of (extensions of) Gold-style computational learning theory addressing a mathematically precise version of the following question: Are there learning tasks that require U-shaped behavior? All notions considered are learning in the limit from positive data. We present results about the necessity of U-shaped learning in classical models of learning as well as in models with bounds on the memory of the learner. The pattern emerges that, for parameterized, cognitively relevant learning criteria, beyond very few initial parameter values, U-shapes are necessary for full learning power! We discuss the possible relevance of the above results for the Cognitive Science debate about learning models as well as directions for future research. Copyright © 2013 Cognitive Science Society, Inc.

Learning a New Selection Rule in Visual and Frontal Cortex.

PubMed

van der Togt, Chris; Stănişor, Liviu; Pooresmaeili, Arezoo; Albantakis, Larissa; Deco, Gustavo; Roelfsema, Pieter R

2016-08-01

How do you make a decision if you do not know the rules of the game? Models of sensory decision-making suggest that choices are slow if evidence is weak, but they may only apply if the subject knows the task rules. Here, we asked how the learning of a new rule influences neuronal activity in the visual (area V1) and frontal cortex (area FEF) of monkeys. We devised a new icon-selection task. On each day, the monkeys saw 2 new icons (small pictures) and learned which one was relevant. We rewarded eye movements to a saccade target connected to the relevant icon with a curve. Neurons in visual and frontal cortex coded the monkey's choice, because the representation of the selected curve was enhanced. Learning delayed the neuronal selection signals and we uncovered the cause of this delay in V1, where learning to select the relevant icon caused an early suppression of surrounding image elements. These results demonstrate that the learning of a new rule causes a transition from fast and random decisions to a more considerate strategy that takes additional time and they reveal the contribution of visual and frontal cortex to the learning process. © The Author 2016. Published by Oxford University Press.

Learning the Constellations: From Junior High to Undergraduate Descriptive Astronomy Class

NASA Astrophysics Data System (ADS)

Stephens, Denise C.; Hintz, Eric G.; Hintz, Maureen; Lawler, Jeannette; Jones, Michael; Bench, Nathan

2015-01-01

As part of two separate studies we have examined the ability of students to learn and remember a group of constellations, bright stars, and deep sky objects. For a group of junior high students we tested their knowledge of only the constellations by giving them a 'constellation quiz' without any instruction. We then provided the students with a lab session, and retested. We also tested a large number of undergraduate students in our descriptive astronomy classes, but in this case there were the same 30 constellations, 17 bright stars, and 3 deep sky objects. The undergraduate students were tested in a number of ways: 1) pre-testing without instruction, 2) self-reporting of knowledge, 3) normal constellation quizzes as part of the class, and 4) retesting students from previous semesters. This provided us with a set of baseline measurements, allowed us to track the learning curve, and test retention of the material. We will present our early analysis of the data.

Spaghetti Sine Curves: Virtual Environments for Reasoning and Sense Making

ERIC Educational Resources Information Center

Özgün-Koca, S. Asli; Edwards, Michael Todd; Meagher, Michael

2013-01-01

In a recent collaboration with an area high school teacher, the authors were asked to develop an introductory sinusoidal curves lesson for a group of second-year algebra students. Because the topic was abstract and unfamiliar to these tenth graders, they looked for hands-on lessons to support their learning. One lesson that they found, which they…

Developing a Drawing Task to Differentiate Group Average Time Course vs. Dynamics in the Individual

ERIC Educational Resources Information Center

Blech, Christine; Gaschler, Robert

2017-01-01

Teaching and theorizing in psychology has long been torn between targeting general underlying principles by observing dynamics in the individual or focusing on average behavior. As dealing with group averages is common practice in psychology, it is important for students to understand how individual learning curves relate to group average curves.…

Quantitative Ultrasound for Measuring Obstructive Severity in Children with Hydronephrosis.

PubMed

Cerrolaza, Juan J; Peters, Craig A; Martin, Aaron D; Myers, Emmarie; Safdar, Nabile; Linguraru, Marius George

2016-04-01

We define sonographic biomarkers for hydronephrotic renal units that can predict the necessity of diuretic nuclear renography. We selected a cohort of 50 consecutive patients with hydronephrosis of varying severity in whom 2-dimensional sonography and diuretic mercaptoacetyltriglycine renography had been performed. A total of 131 morphological parameters were computed using quantitative image analysis algorithms. Machine learning techniques were then applied to identify ultrasound based safety thresholds that agreed with the t½ for washout. A best fit model was then derived for each threshold level of t½ that would be clinically relevant at 20, 30 and 40 minutes. Receiver operating characteristic curve analysis was performed. Sensitivity, specificity and area under the receiver operating characteristic curve were determined. Improvement obtained by the quantitative imaging method compared to the Society for Fetal Urology grading system and the hydronephrosis index was statistically verified. For the 3 thresholds considered and at 100% sensitivity the specificities of the quantitative imaging method were 94%, 70% and 74%, respectively. Corresponding area under the receiver operating characteristic curve values were 0.98, 0.94 and 0.94, respectively. Improvement obtained by the quantitative imaging method over the Society for Fetal Urology grade and hydronephrosis index was statistically significant (p <0.05 in all cases). Quantitative imaging analysis of renal sonograms in children with hydronephrosis can identify thresholds of clinically significant washout times with 100% sensitivity to decrease the number of diuretic renograms in up to 62% of children. Copyright © 2016 American Urological Association Education and Research, Inc. Published by Elsevier Inc. All rights reserved.

Experience in the management of ECMO therapy as a mortality risk factor.

PubMed

Guilló Moreno, V; Gutiérrez Martínez, A; Romero Berrocal, A; Sánchez Castilla, M; García-Fernández, J

2018-02-01

The extracorporeal oxygenation membrane (ECMO) is a system that provides circulatory and respiratory assistance to patients in cardiac or respiratory failure refractory to conventional treatment. It is a therapy with numerous associated complications and high mortality. Multidisciplinary management and experienced teams increase survival. Our purpose is to evaluate and analyse the effect of the learning curve on mortality. Retrospective and observational study of 31 patients, from January 2012 to December 2015. Patients were separated into 2periods. These periods were divided by the establishment of an ECMO protocol. We compared the quantitative variables by performing the Mann-Whitney U test. For the categorical qualitative variables we performed the chi-square test or Fisher exact statistic as appropriate. The survival curve was computed using the Kaplan-Meier method, and the analysis of statistical significance using the Log-rank test. Data analysis was performed with the STATA programme 14. Survival curves show the tendency to lower mortality in the subsequent period (P=0.0601). The overall mortality rate in the initial period was higher than in the subsequent period (P=0.042). In another analysis, we compared the characteristics of the 2groups and concluded that they were homogeneous. The degree of experience is an independent factor for mortality. The application of a care protocol is fundamental to facilitate the management of ECMO therapy. Copyright © 2017 Sociedad Española de Anestesiología, Reanimación y Terapéutica del Dolor. Publicado por Elsevier España, S.L.U. All rights reserved.

An interplay of fusiform gyrus and hippocampus enables prototype- and exemplar-based category learning.

PubMed

Lech, Robert K; Güntürkün, Onur; Suchan, Boris

2016-09-15

The aim of the present study was to examine the contributions of different brain structures to prototype- and exemplar-based category learning using functional magnetic resonance imaging (fMRI). Twenty-eight subjects performed a categorization task in which they had to assign prototypes and exceptions to two different families. This test procedure usually produces different learning curves for prototype and exception stimuli. Our behavioral data replicated these previous findings by showing an initially superior performance for prototypes and typical stimuli and a switch from a prototype-based to an exemplar-based categorization for exceptions in the later learning phases. Since performance varied, we divided participants into learners and non-learners. Analysis of the functional imaging data revealed that the interaction of group (learners vs. non-learners) and block (Block 5 vs. Block 1) yielded an activation of the left fusiform gyrus for the processing of prototypes, and an activation of the right hippocampus for exceptions after learning the categories. Thus, successful prototype- and exemplar-based category learning is associated with activations of complementary neural substrates that constitute object-based processes of the ventral visual stream and their interaction with unique-cue representations, possibly based on sparse coding within the hippocampus. Copyright © 2016 Elsevier B.V. All rights reserved.

Laparoscopic colorectal surgery in learning curve: Role of implementation of a standardized technique and recovery protocol. A cohort study

PubMed Central

Luglio, Gaetano; De Palma, Giovanni Domenico; Tarquini, Rachele; Giglio, Mariano Cesare; Sollazzo, Viviana; Esposito, Emanuela; Spadarella, Emanuela; Peltrini, Roberto; Liccardo, Filomena; Bucci, Luigi

2015-01-01

Background Despite the proven benefits, laparoscopic colorectal surgery is still under utilized among surgeons. A steep learning is one of the causes of its limited adoption. Aim of the study is to determine the feasibility and morbidity rate after laparoscopic colorectal surgery in a single institution, “learning curve” experience, implementing a well standardized operative technique and recovery protocol. Methods The first 50 patients treated laparoscopically were included. All the procedures were performed by a trainee surgeon, supervised by a consultant surgeon, according to the principle of complete mesocolic excision with central vascular ligation or TME. Patients underwent a fast track recovery programme. Recovery parameters, short-term outcomes, morbidity and mortality have been assessed. Results Type of resections: 20 left side resections, 8 right side resections, 14 low anterior resection/TME, 5 total colectomy and IRA, 3 total panproctocolectomy and pouch. Mean operative time: 227 min; mean number of lymph-nodes: 18.7. Conversion rate: 8%. Mean time to flatus: 1.3 days; Mean time to solid stool: 2.3 days. Mean length of hospital stay: 7.2 days. Overall morbidity: 24%; major morbidity (Dindo–Clavien III): 4%. No anastomotic leak, no mortality, no 30-days readmission. Conclusion Proper laparoscopic colorectal surgery is safe and leads to excellent results in terms of recovery and short term outcomes, even in a learning curve setting. Key factors for better outcomes and shortening the learning curve seem to be the adoption of a standardized technique and training model along with the strict supervision of an expert colorectal surgeon. PMID:25859386

Applying Active Learning to Assertion Classification of Concepts in Clinical Text

PubMed Central

Chen, Yukun; Mani, Subramani; Xu, Hua

2012-01-01

Supervised machine learning methods for clinical natural language processing (NLP) research require a large number of annotated samples, which are very expensive to build because of the involvement of physicians. Active learning, an approach that actively samples from a large pool, provides an alternative solution. Its major goal in classification is to reduce the annotation effort while maintaining the quality of the predictive model. However, few studies have investigated its uses in clinical NLP. This paper reports an application of active learning to a clinical text classification task: to determine the assertion status of clinical concepts. The annotated corpus for the assertion classification task in the 2010 i2b2/VA Clinical NLP Challenge was used in this study. We implemented several existing and newly developed active learning algorithms and assessed their uses. The outcome is reported in the global ALC score, based on the Area under the average Learning Curve of the AUC (Area Under the Curve) score. Results showed that when the same number of annotated samples was used, active learning strategies could generate better classification models (best ALC – 0.7715) than the passive learning method (random sampling) (ALC – 0.7411). Moreover, to achieve the same classification performance, active learning strategies required fewer samples than the random sampling method. For example, to achieve an AUC of 0.79, the random sampling method used 32 samples, while our best active learning algorithm required only 12 samples, a reduction of 62.5% in manual annotation effort. PMID:22127105

Machine Learning of Three-dimensional Right Ventricular Motion Enables Outcome Prediction in Pulmonary Hypertension: A Cardiac MR Imaging Study.

PubMed

Dawes, Timothy J W; de Marvao, Antonio; Shi, Wenzhe; Fletcher, Tristan; Watson, Geoffrey M J; Wharton, John; Rhodes, Christopher J; Howard, Luke S G E; Gibbs, J Simon R; Rueckert, Daniel; Cook, Stuart A; Wilkins, Martin R; O'Regan, Declan P

2017-05-01

Purpose To determine if patient survival and mechanisms of right ventricular failure in pulmonary hypertension could be predicted by using supervised machine learning of three-dimensional patterns of systolic cardiac motion. Materials and Methods The study was approved by a research ethics committee, and participants gave written informed consent. Two hundred fifty-six patients (143 women; mean age ± standard deviation, 63 years ± 17) with newly diagnosed pulmonary hypertension underwent cardiac magnetic resonance (MR) imaging, right-sided heart catheterization, and 6-minute walk testing with a median follow-up of 4.0 years. Semiautomated segmentation of short-axis cine images was used to create a three-dimensional model of right ventricular motion. Supervised principal components analysis was used to identify patterns of systolic motion that were most strongly predictive of survival. Survival prediction was assessed by using difference in median survival time and area under the curve with time-dependent receiver operating characteristic analysis for 1-year survival. Results At the end of follow-up, 36% of patients (93 of 256) died, and one underwent lung transplantation. Poor outcome was predicted by a loss of effective contraction in the septum and free wall, coupled with reduced basal longitudinal motion. When added to conventional imaging and hemodynamic, functional, and clinical markers, three-dimensional cardiac motion improved survival prediction (area under the receiver operating characteristic curve, 0.73 vs 0.60, respectively; P < .001) and provided greater differentiation according to difference in median survival time between high- and low-risk groups (13.8 vs 10.7 years, respectively; P < .001). Conclusion A machine-learning survival model that uses three-dimensional cardiac motion predicts outcome independent of conventional risk factors in patients with newly diagnosed pulmonary hypertension. Online supplemental material is available for this article.

5 Myths about Classroom Technology: How Do We Integrate Digital Tools to Truly Enhance Learning?

ERIC Educational Resources Information Center

Renwick, Matt

2015-01-01

What's keeping your school behind the technology curve? Is it a fear of the unfamiliar? Expenses? Or some other myth? Have you considered how students with special needs or students learning a second language may benefit from using digital tools? If you've fallen for the perception that technology is too expensive, unnecessary for real learning,…

Implementing Machine Learning in the PCWG Tool

DOE Office of Scientific and Technical Information (OSTI.GOV)

Clifton, Andrew; Ding, Yu; Stuart, Peter

The Power Curve Working Group (www.pcwg.org) is an ad-hoc industry-led group to investigate the performance of wind turbines in real-world conditions. As part of ongoing experience-sharing exercises, machine learning has been proposed as a possible way to predict turbine performance. This presentation provides some background information about machine learning and how it might be implemented in the PCWG exercises.

Predictive inference for best linear combination of biomarkers subject to limits of detection.

PubMed

Coolen-Maturi, Tahani

2017-08-15

Measuring the accuracy of diagnostic tests is crucial in many application areas including medicine, machine learning and credit scoring. The receiver operating characteristic (ROC) curve is a useful tool to assess the ability of a diagnostic test to discriminate between two classes or groups. In practice, multiple diagnostic tests or biomarkers are combined to improve diagnostic accuracy. Often, biomarker measurements are undetectable either below or above the so-called limits of detection (LoD). In this paper, nonparametric predictive inference (NPI) for best linear combination of two or more biomarkers subject to limits of detection is presented. NPI is a frequentist statistical method that is explicitly aimed at using few modelling assumptions, enabled through the use of lower and upper probabilities to quantify uncertainty. The NPI lower and upper bounds for the ROC curve subject to limits of detection are derived, where the objective function to maximize is the area under the ROC curve. In addition, the paper discusses the effect of restriction on the linear combination's coefficients on the analysis. Examples are provided to illustrate the proposed method. Copyright © 2017 John Wiley & Sons, Ltd. Copyright © 2017 John Wiley & Sons, Ltd.

Mastery versus the standard proficiency target for basic laparoscopic skill training: effect on skill transfer and retention.

PubMed

Kolozsvari, Nicoleta O; Kaneva, Pepa; Brace, Chantalle; Chartrand, Genevieve; Vaillancourt, Marilou; Cao, Jiguo; Banaszek, Daniel; Demyttenaere, Sebastian; Vassiliou, Melina C; Fried, Gerald M; Feldman, Liane S

2011-07-01

Little evidence exists to guide educators in the best way to implement simulation within surgical skills curricula. This study investigated whether practicing a basic Fundamentals of Laparoscopic Surgery (FLS) simulator task [peg transfer (PT)] facilitates learning a more complex skill [intracorporeal suturing (ICS)] and compared the effect of PT training to mastery with training to the passing level on PT retention and on learning ICS. For this study, 98 surgically naïve subjects were randomized to one of three PT training groups: control, standard training, and overtraining. All the participants then trained in ICS. The learning curves for ICS were analyzed by estimating the learning plateau and rate using nonlinear regression. Skill retention was assessed by retesting participants 1 month after training. The groups were compared using analysis of variance (ANOVA). Effectiveness of skill transfer was calculated using the transfer effectiveness ratio (TER). Data are presented as mean±standard deviation (p<0.05). The study was completed by 77 participants (28 control, 26 standard, and 23 overtrained subjects). The ICS learning plateau rose with increasing PT training (452±10 vs. 459±10 vs. 467±10; p<0.01). Increased PT training was associated with a trend toward higher initial ICS scores (128±107 vs. 127±110 vs. 183±106; p=0.13) and faster learning rates (15±4 vs. 14±4 vs. 13±4 trials; p=0.10). At retention, there were no differences in PT scores (p=0.5). The PT training took 20±10 min for standard training and 39±20 min for overtraining (p<0.01). Overtrained participants saved 11±5 min in ICS training compared with the control subjects (p=0.04). However, TER was 0.165 for the overtraining group and 0.160 for the standard training group, suggesting that PT overtraining took longer than the time saved on ICS training. For surgically naïve subjects, part-task training with PT alone was associated with slight improvements in the learning curve for ICS. However, overtraining with PT did not improve skill retention, and peg training alone was not an efficient strategy for learning ICS.

Training on N.O.T.E.S.: from history we learn.

PubMed

Al-Akash, M; Boyle, E; Tanner, W A

2009-06-01

Surgical errors occurring early in the learning curve of laparoscopic surgery providers delayed the uptake and progress of minimally invasive surgery (MIS) for years. This taught us a valuable lesson; innovations in surgical techniques should not be rapidly implemented until all aspects including applicability, feasibility and safety have been fully tested. In 2005, the Natural Orifice Surgery Consortium for Assessment and Research (NOSCAR) published a white paper highlighting the barriers to NOTES development and identifying key elements for its progress. One of these elements is the training of future providers. Proficiency-based, virtual reality simulation will offer a feasible alternative to animal testing once the safety and efficacy parameters of NOTES are established. Recent advances in imaging including computed tomography (CT) scanning, magnetic resonance imaging (MRI) scanning, and ultrasound (US) scanning can offer improved image registration and real-time tracking. Combining these advanced imaging technologies with the newly designed virtual reality simulators will result in a fully comprehensive simulation curriculum which will offer a unique facility for future NOTES providers to train anytime, anywhere, and as much as they need to in order to achieve the pre-set proficiency levels for a variety of NOTES procedures. Furthermore they will incorporate patient-specific anatomical models obtained from patient imaging and uploaded onto the simulator to ensure face reliability and validity assurance. Training in a clean, safe environment with proximate feedback and performance analysis will help accelerate the learning curve and therefore improve patients' safety and outcomes in order to maximize the benefits of innovative access procedures such as NOTES.

Learning Under Stress: The Inverted-U-Shape Function Revisited

ERIC Educational Resources Information Center

Salehi, Basira; Cordero, M. Isabel; Sandi, Carmen

2010-01-01

Although the relationship between stress intensity and memory function is generally believed to follow an inverted-U-shaped curve, strikingly this phenomenon has not been demonstrated under the same experimental conditions. We investigated this phenomenon for rats' performance in a hippocampus-dependent learning task, the radial arm water maze…

Tale of the Tape: International Teaching Assistant Noticing during Videotaped Classroom Observations

ERIC Educational Resources Information Center

Williams, Gwendolyn M.; Case, Rod E.

2015-01-01

International teaching assistants face challenges in learning the norms for teaching in American universities. In order to address this learning curve this article describes a qualitative study of twenty international teaching assistants that examined how these participants viewed observations as part of their professional development. The study…

Assessment of Postflight Locomotor Performance Utilizing a Test of Functional Mobility: Strategic and Adaptive Responses

NASA Technical Reports Server (NTRS)

Warren, L. E.; Mulavara, A. P.; Peters, B. T.; Cohen, H. S.; Richards, J. T.; Miller, C. A.; Brady, R.; Ruttley, T. M.; Bloomberg, J. J.

2006-01-01

Space flight induces adaptive modification in sensorimotor function, allowing crewmembers to operate in the unique microgravity environment. This adaptive state, however, is inappropriate for a terrestrial environment. During a re-adaptation period upon their return to Earth, crewmembers experience alterations in sensorimotor function, causing various disturbances in perception, spatial orientation, posture, gait, and eye-head coordination. Following long duration space flight, sensorimotor dysfunction would prevent or extend the time required to make an emergency egress from the vehicle; compromising crew safety and mission objectives. We are investigating two types of motor learning that may interact with each other and influence a crewmember's ability to re-adapt to Earth's gravity environment. In strategic learning, crewmembers make rapid modifications in their motor control strategy emphasizing error reduction. This type of learning may be critical during the first minutes and hours after landing. In adaptive learning, long-term plastic transformations occur, involving morphological changes and synaptic modification. In recent literature these two behavioral components have been associated with separate brain structures that control the execution of motor strategies: the strategic component was linked to the posterior parietal cortex and the adaptive component was linked to the cerebellum (Pisella, et al. 2004). The goal of this paper was to demonstrate the relative contributions of the strategic and adaptive components to the re-adaptation process in locomotor control after long duration space flight missions on the International Space Station (ISS). The Functional Mobility Test (FMT) was developed to assess crewmember s ability to ambulate postflight from an operational and functional perspective. Sixteen crewmembers were tested preflight (3 sessions) and postflight (days 1, 2, 4, 7, 25) following a long duration space flight (approx 6 months) on the ISS. We have further analyzed the FMT data to characterize strategic and adaptive components during the postflight readaptation period. Crewmembers walked at a preferred pace through an obstacle course set up on a base of 10 cm thick medium density foam (Sunmate Foam, Dynamic Systems, Inc., Leicester, NC). The 6.0m X 4.0m course consisted of several pylons made of foam; a Styrofoam barrier 46.0cm high that crewmembers stepped over; and a portal constructed of two Styrofoam blocks, each 31cm high, with a horizontal bar covered by foam and suspended from the ceiling which was adjusted to the height of the crewmember s shoulder. The portal required crewmembers to bend at the waist and step over a barrier simultaneously. All obstacles were lightweight, soft and easily knocked over. Crewmembers were instructed to walk through the course as quickly and as safely as possible without touching any of the objects on the course. This task was performed three times in the clockwise direction and three times in the counterclockwise direction that was randomly chosen. The dependent measures for each trial were: time to complete the course (seconds) and the number of obstacles touched or knocked down. For each crewmember, the time to complete each FMT trial from postflight days 1, 2, 4, 7 and 25 were further analyzed. A single logarithmic curve using a least squares calculation was fit through these data to produce a single comprehensive curve (macro). This macro curve composed of data spanning 25 days, illustrates the re-adaptive learning function over the longer time scale term. Additionally, logarithmic curves were fit to the 6 data trials within each individual post flight test day to produce 5 separate daily curves. These micro curves, produced from data obtained over the course of minutes, illustrates the strategic learning function exhibited over a relative shorter time scale. The macro curve for all subjects exhibited adaptive motor learning patterns over the 25 day period. Howev, 9/16 crewmembers exhibited significant strategic motor learning patterns in their micro curves, as defined by m > 1 in the equation of the line y=m*LN(x) +b. These data indicate that postflight recovery in locomotor function involves both strategic and adaptive mechanisms. Future countermeasures will be designed to enhance both recovery processes.

Extensions to decision curve analysis, a novel method for evaluating diagnostic tests, prediction models and molecular markers

PubMed Central

Vickers, Andrew J; Cronin, Angel M; Elkin, Elena B; Gonen, Mithat

2008-01-01

Background Decision curve analysis is a novel method for evaluating diagnostic tests, prediction models and molecular markers. It combines the mathematical simplicity of accuracy measures, such as sensitivity and specificity, with the clinical applicability of decision analytic approaches. Most critically, decision curve analysis can be applied directly to a data set, and does not require the sort of external data on costs, benefits and preferences typically required by traditional decision analytic techniques. Methods In this paper we present several extensions to decision curve analysis including correction for overfit, confidence intervals, application to censored data (including competing risk) and calculation of decision curves directly from predicted probabilities. All of these extensions are based on straightforward methods that have previously been described in the literature for application to analogous statistical techniques. Results Simulation studies showed that repeated 10-fold crossvalidation provided the best method for correcting a decision curve for overfit. The method for applying decision curves to censored data had little bias and coverage was excellent; for competing risk, decision curves were appropriately affected by the incidence of the competing risk and the association between the competing risk and the predictor of interest. Calculation of decision curves directly from predicted probabilities led to a smoothing of the decision curve. Conclusion Decision curve analysis can be easily extended to many of the applications common to performance measures for prediction models. Software to implement decision curve analysis is provided. PMID:19036144

Extensions to decision curve analysis, a novel method for evaluating diagnostic tests, prediction models and molecular markers.

PubMed

Vickers, Andrew J; Cronin, Angel M; Elkin, Elena B; Gonen, Mithat

2008-11-26

Decision curve analysis is a novel method for evaluating diagnostic tests, prediction models and molecular markers. It combines the mathematical simplicity of accuracy measures, such as sensitivity and specificity, with the clinical applicability of decision analytic approaches. Most critically, decision curve analysis can be applied directly to a data set, and does not require the sort of external data on costs, benefits and preferences typically required by traditional decision analytic techniques. In this paper we present several extensions to decision curve analysis including correction for overfit, confidence intervals, application to censored data (including competing risk) and calculation of decision curves directly from predicted probabilities. All of these extensions are based on straightforward methods that have previously been described in the literature for application to analogous statistical techniques. Simulation studies showed that repeated 10-fold crossvalidation provided the best method for correcting a decision curve for overfit. The method for applying decision curves to censored data had little bias and coverage was excellent; for competing risk, decision curves were appropriately affected by the incidence of the competing risk and the association between the competing risk and the predictor of interest. Calculation of decision curves directly from predicted probabilities led to a smoothing of the decision curve. Decision curve analysis can be easily extended to many of the applications common to performance measures for prediction models. Software to implement decision curve analysis is provided.

Hydrologic Impacts of Oak Harvesting and Evaluation of the Modified Universal Soil Loss Equation

Treesearch

Charlette R. Epifanio; Michael J. Singer; Xiaohong Huang

1991-01-01

Two Sierra foothill watersheds were monitored to learn what effects selective oak removal would have on watershed hydrology and water quality. We also used the data to generate sediment rating curves and evaluate the modified universal soil loss equation (MUSLE). Annual sediment rating curves better accounted for the variability in precipitation events from year to...

What Does a Graphical Representation Mean for Students at the Beginning of Function Teaching?

ERIC Educational Resources Information Center

Yavuz, Ilyas

2010-01-01

This study examines how students in the early stages of learning about the concept of functions, describe a curve and, in particular, evaluate the appropriateness of their argument about the representation of a function. Students are offered a message game which is related to a curve drawn on a coordinate system, representing an "imaginary…

Longitudinal Examination of Procrastination and Anxiety, and Their Relation to Self-Efficacy for Self- Regulated Learning: Latent Growth Curve Modeling

ERIC Educational Resources Information Center

Yerdelen, Sündüs; McCaffrey, Adam; Klassen, Robert M.

2016-01-01

This study investigated the longitudinal association between students' anxiety and procrastination and the relation of self-efficacy for self-regulation to these constructs. Latent Growth Curve Modeling was used to analyze data gathered from 182 undergraduate students (134 female, 48 male) at 4 times during a semester. Our results showed that…

The "U" Curve Hypothesis: A Framework for Making Sense of Learning to Teach in Diverse Settings

ERIC Educational Resources Information Center

Birrell, James R.; Tinney, Mari Vawn

2008-01-01

Experiences in this research study started in 1991 before many teacher educators were aware of the "?U"? curve hypothesis or predictable stages of culture shock and the recognizable stages used on the path to gaining intercultural competence. This study of student teachers is used here as an illustration of what happens when teachers are…

The Sociocultural Adjustment Trajectory of International University Students and the Role of University Structures: A Qualitative Investigation

ERIC Educational Resources Information Center

Coles, Rebecca; Swami, Viren

2012-01-01

The present research explores the sociocultural adjustment of international students and the role played by university structures in the process. The adjustment process of international students has been modelled in psychological literature as a U-curve, a learning curve and most recently as a long, uneven and unending process of change. Yet,…

Medical learning curves and the Kantian ideal.

PubMed

Le Morvan, P; Stock, B

2005-09-01

A hitherto unexamined problem for the "Kantian ideal" that one should always treat patients as ends in themselves, and never only as a means to other ends, is explored in this paper. The problem consists of a prima facie conflict between this Kantian ideal and the reality of medical practice. This conflict arises because, at least presently, medical practitioners can only acquire certain skills and abilities by practising on live, human patients, and given the inevitability and ubiquity of learning curves, this learning requires some patients to be treated only as a means to this end. A number of ways of attempting to establish the compatibility of the Kantian Ideal with the reality of medical practice are considered. Each attempt is found to be unsuccessful. Accordingly, until a way is found to reconcile them, we conclude that the Kantian ideal is inconsistent with the reality of medical practice.

Medical learning curves and the Kantian ideal

PubMed Central

Le Morvan, P; Stock, B

2005-01-01

A hitherto unexamined problem for the "Kantian ideal" that one should always treat patients as ends in themselves, and never only as a means to other ends, is explored in this paper. The problem consists of a prima facie conflict between this Kantian ideal and the reality of medical practice. This conflict arises because, at least presently, medical practitioners can only acquire certain skills and abilities by practising on live, human patients, and given the inevitability and ubiquity of learning curves, this learning requires some patients to be treated only as a means to this end. A number of ways of attempting to establish the compatibility of the Kantian Ideal with the reality of medical practice are considered. Each attempt is found to be unsuccessful. Accordingly, until a way is found to reconcile them, we conclude that the Kantian ideal is inconsistent with the reality of medical practice. PMID:16131552

TOOKUIL: A case study in user interface development for safety code application

DOE Office of Scientific and Technical Information (OSTI.GOV)

Gray, D.L.; Harkins, C.K.; Hoole, J.G.

1997-07-01

Traditionally, there has been a very high learning curve associated with using nuclear power plant (NPP) analysis codes. Even for seasoned plant analysts and engineers, the process of building or modifying an input model for present day NPP analysis codes is tedious, error prone, and time consuming. Current cost constraints and performance demands place an additional burden on today`s safety analysis community. Advances in graphical user interface (GUI) technology have been applied to obtain significant productivity and quality assurance improvements for the Transient Reactor Analysis Code (TRAC) input model development. KAPL Inc. has developed an X Windows-based graphical user interfacemore » named TOOKUIL which supports the design and analysis process, acting as a preprocessor, runtime editor, help system, and post processor for TRAC. This paper summarizes the objectives of the project, the GUI development process and experiences, and the resulting end product, TOOKUIL.« less

Ensemble Learning Method for Outlier Detection and its Application to Astronomical Light Curves

NASA Astrophysics Data System (ADS)

Nun, Isadora; Protopapas, Pavlos; Sim, Brandon; Chen, Wesley

2016-09-01

Outlier detection is necessary for automated data analysis, with specific applications spanning almost every domain from financial markets to epidemiology to fraud detection. We introduce a novel mixture of the experts outlier detection model, which uses a dynamically trained, weighted network of five distinct outlier detection methods. After dimensionality reduction, individual outlier detection methods score each data point for “outlierness” in this new feature space. Our model then uses dynamically trained parameters to weigh the scores of each method, allowing for a finalized outlier score. We find that the mixture of experts model performs, on average, better than any single expert model in identifying both artificially and manually picked outliers. This mixture model is applied to a data set of astronomical light curves, after dimensionality reduction via time series feature extraction. Our model was tested using three fields from the MACHO catalog and generated a list of anomalous candidates. We confirm that the outliers detected using this method belong to rare classes, like Novae, He-burning, and red giant stars; other outlier light curves identified have no available information associated with them. To elucidate their nature, we created a website containing the light-curve data and information about these objects. Users can attempt to classify the light curves, give conjectures about their identities, and sign up for follow up messages about the progress made on identifying these objects. This user submitted data can be used further train of our mixture of experts model. Our code is publicly available to all who are interested.

A Study of Learning in the Operations of a Viscous Damped Traversing Unit.

DTIC Science & Technology

1978-06-01

Finally I would like to express special appreciation to my wife, Donna Pardue Robinson, and my children, Kristin, Keith and Stephanie for their...indication of learning. Bahrick, Fitts and Briggs dealt with learning curves in a 1957 article which supported their earlier work. They used the same data...Olof and Kaare Rodahl, Textbook of Work Physiology, New York: McGraw-Hill Book Compnay, 1970. 2. Bahrick, H. P., F. M. Fitts and G. E. Briggs , "Learning

An evidence-based virtual reality training program for novice laparoscopic surgeons.

PubMed

Aggarwal, Rajesh; Grantcharov, Teodor P; Eriksen, Jens R; Blirup, Dorthe; Kristiansen, Viggo B; Funch-Jensen, Peter; Darzi, Ara

2006-08-01

To develop an evidence-based virtual reality laparoscopic training curriculum for novice laparoscopic surgeons to achieve a proficient level of skill prior to participating in live cases. Technical skills for laparoscopic surgery must be acquired within a competency-based curriculum that begins in the surgical skills laboratory. Implementation of this program necessitates the definition of the validity, learning curves and proficiency criteria on the training tool. The study recruited 40 surgeons, classified into experienced (performed >100 laparoscopic cholecystectomies) or novice groups (<10 laparoscopic cholecystectomies). Ten novices and 10 experienced surgeons were tested on basic tasks, and 11 novices and 9 experienced surgeons on a procedural module for dissection of Calot triangle. Performance of the 2 groups was assessed using time, error, and economy of movement parameters. All basic tasks demonstrated construct validity (Mann-Whitney U test, P < 0.05), and learning curves for novices plateaued at a median of 7 repetitions (Friedman's test, P < 0.05). Expert surgeons demonstrated a learning rate at a median of 2 repetitions (P < 0.05). Performance on the dissection module demonstrated significant differences between experts and novices (P < 0.002); learning curves for novice subjects plateaued at the fourth repetition (P < 0.05). Expert benchmark criteria were defined for validated parameters on each task. A competency-based training curriculum for novice laparoscopic surgeons has been defined. This can serve to ensure that junior trainees have acquired prerequisite levels of skill prior to entering the operating room, and put them directly into practice.

Increased reward in ankle robotics training enhances motor control and cortical efficiency in stroke.

PubMed

Goodman, Ronald N; Rietschel, Jeremy C; Roy, Anindo; Jung, Brian C; Diaz, Jason; Macko, Richard F; Forrester, Larry W

2014-01-01

Robotics is rapidly emerging as a viable approach to enhance motor recovery after disabling stroke. Current principles of cognitive motor learning recognize a positive relationship between reward and motor learning. Yet no prior studies have established explicitly whether reward improves the rate or efficacy of robotics-assisted rehabilitation or produces neurophysiologic adaptations associated with motor learning. We conducted a 3 wk, 9-session clinical pilot with 10 people with chronic hemiparetic stroke, randomly assigned to train with an impedance-controlled ankle robot (anklebot) under either high reward (HR) or low reward conditions. The 1 h training sessions entailed playing a seated video game by moving the paretic ankle to hit moving onscreen targets with the anklebot only providing assistance as needed. Assessments included paretic ankle motor control, learning curves, electroencephalograpy (EEG) coherence and spectral power during unassisted trials, and gait function. While both groups exhibited changes in EEG, the HR group had faster learning curves (p = 0.05), smoother movements (p

Recession curve analysis for groundwater levels: case study in Latvia

NASA Astrophysics Data System (ADS)

Gailuma, A.; VÄ«tola, I.; Abramenko, K.; Lauva, D.; Vircavs, V.; Veinbergs, A.; Dimanta, Z.

2012-04-01

Recession curve analysis is powerful and effective analysis technique in many research areas related with hydrogeology where observations have to be made, such as water filtration and absorption of moisture, irrigation and drainage, planning of hydroelectric power production and chemical leaching (elution of chemical substances) as well as in other areas. The analysis of the surface runoff hydrograph`s recession curves, which is performed to conceive the after-effects of interaction of precipitation and surface runoff, has approved in practice. The same method for analysis of hydrograph`s recession curves can be applied for the observations of the groundwater levels. There are manually prepared hydrograph for analysis of recession curves for observation wells (MG2, BG2 and AG1) in agricultural monitoring sites in Latvia. Within this study from the available monitoring data of groundwater levels were extracted data of declining periods, splitted by month. The drop-down curves were manually (by changing the date) moved together, until to find the best match, thereby obtaining monthly drop-down curves, representing each month separately. Monthly curves were combined and manually joined, for obtaining characterizing drop-down curves of the year for each well. Within the process of decreased recession curve analysis, from the initial curve was cut out upward areas, leaving only the drops of the curve, consequently, the curve is transformed more closely to the groundwater flow, trying to take out the impact of rain or drought periods from the curve. Respectively, the drop-down curve is part of the data, collected with hydrograph, where data with the discharge dominates, without considering impact of precipitation. Using the recession curve analysis theory, ready tool "A Visual Basic Spreadsheet Macro for Recession Curve Analysis" was used for selection of data and logarithmic functions matching (K. Posavec et.al., GROUND WATER 44, no. 5: 764-767, 2006), as well as functions were developed by manual processing of data. For displaying data the mathematical model of data equalization was used, finding the corresponding or closest logarithmic function of the recession for the graph. Obtained recession curves were similar but not identical. With full knowledge of the fluctuations of ground water level, it is possible to indirectly (without taking soil samples) determine the filtration coefficient: more rapid decline in the recession curve correspond for the better filtration conditions. This research could be very useful in construction planning, road constructions, agriculture etc. Acknowledgments The authors gratefully acknowledge the funding from ESF Project "Establishment of interdisciplinary scientist group and modeling system for groundwater research" (Agreement No. 2009/0212/1DP/1.1.1.2.0/09/APIA/VIAA/060EF7)

Mathcad in the Chemistry Curriculum Symbolic Software in the Chemistry Curriculum

NASA Astrophysics Data System (ADS)

Zielinski, Theresa Julia

2000-05-01

Physical chemistry is such a broad discipline that the topics we expect average students to complete in two semesters usually exceed their ability for meaningful learning. Consequently, the number and kind of topics and the efficiency with which students can learn them are important concerns. What topics are essential and what can we do to provide efficient and effective access to those topics? How do we accommodate the fact that students come to upper-division chemistry courses with a variety of nonuniformly distributed skills, a bit of calculus, and some physics studied one or more years before physical chemistry? The critical balance between depth and breadth of learning in courses and curricula may be achieved through appropriate use of technology and especially through the use of symbolic mathematics software. Software programs such as Mathcad, Mathematica, and Maple, however, have learning curves that diminish their effectiveness for novices. There are several ways to address the learning curve conundrum. First, basic instruction in the software provided during laboratory sessions should be followed by requiring laboratory reports that use the software. Second, one should assign weekly homework that requires the software and builds student skills within the discipline and with the software. Third, a complementary method, supported by this column, is to provide students with Mathcad worksheets or templates that focus on one set of related concepts and incorporate a variety of features of the software that they are to use to learn chemistry. In this column we focus on two significant topics for young chemists. The first is curve-fitting and the statistical analysis of the fitting parameters. The second is the analysis of the rotation/vibration spectrum of a diatomic molecule, HCl. A broad spectrum of Mathcad documents exists for teaching chemistry. One collection of 50 documents can be found at http://www.monmouth.edu/~tzielins/mathcad/Lists/index.htm. Another collection of peer-reviewed documents is developing through this column at the JCE Internet Web site, http://jchemed.chem.wisc.edu/JCEWWW/Features/ McadInChem/index.html. With this column we add three peer-reviewed and tested Mathcad documents to the JCE site. In Linear Least-Squares Regression, Sidney H. Young and Andrzej Wierzbicki demonstrate various implicit and explicit methods for determining the slope and intercept of the regression line for experimental data. The document shows how to determine the standard deviation for the slope, the intercept, and the standard deviation of the overall fit. Students are next given the opportunity to examine the confidence level for the fit through the Student's t-test. Examination of the residuals of the fit leads students to explore the possibility of rejecting points in a set of data. The document concludes with a discussion of and practice with adding a quadratic term to create a polynomial fit to a set of data and how to determine if the quadratic term is statistically significant. There is full documentation of the various steps used throughout the exposition of the statistical concepts. Although the statistical methods presented in this worksheet are generally accessible to average physical chemistry students, an instructor would be needed to explain the finer points of the matrix methods used in some sections of the worksheet. The worksheet is accompanied by a set of data for students to use to practice the techniques presented. It would be worthwhile for students to spend one or two laboratory periods learning to use the concepts presented and then to apply them to experimental data they have collected for themselves. Any linear or linearizable data set would be appropriate for use with this Mathcad worksheet. Alternatively, instructors may select sections of the document suited to the skill level of their students and the laboratory tasks at hand. In a second Mathcad document, Non-Linear Least-Squares Regression, Young and Wierzbicki introduce the basic concepts of nonlinear curve-fitting and develop the techniques needed to fit a variety of mathematical functions to experimental data. This approach is especially important when mathematical models for chemical processes cannot be linearized. In Mathcad the Levenberg-Marquardt algorithm is used to determine the best fitting parameters for a particular mathematical model. As in linear least-squares, the goal of the fitting process is to find the values for the fitting parameters that minimize the sum of the squares of the deviations between the data and the mathematical model. Students are asked to determine the fitting parameters, use the Hessian matrix to compute the standard deviation of the fitting parameters, test for the significance of the parameters using Student's t-test, use residual analysis to test for data points to remove, and repeat the calculations for another set of data. The nonlinear least-squares procedure follows closely on the pattern set up for linear least-squares by the same authors (see above). If students master the linear least-squares worksheet content they will be able to master the nonlinear least-squares technique (see also refs 1, 2). In the third document, The Analysis of the Vibrational Spectrum of a Linear Molecule by Richard Schwenz, William Polik, and Sidney Young, the authors build on the concepts presented in the curve fitting worksheets described above. This vibrational analysis document, which supports a classic experiment performed in the physical chemistry laboratory, shows how a Mathcad worksheet can increase the efficiency by which a set of complicated manipulations for data reduction can be made more accessible for students. The increase in efficiency frees up time for students to develop a fuller understanding of the physical chemistry concepts important to the interpretation of spectra and understanding of bond vibrations in general. The analysis of the vibration/rotation spectrum for a linear molecule worksheet builds on the rich literature for this topic (3). Before analyzing their own spectral data, students practice and learn the concepts and methods of the HCl spectral analysis by using the fundamental and first harmonic vibrational frequencies provided by the authors. This approach has a fundamental pedagogical advantage. Most explanations in laboratory texts are very concise and lack mathematical details required by average students. This Mathcad worksheet acts as a tutor; it guides students through the essential concepts for data reduction and lets them focus on learning important spectroscopic concepts. The Mathcad worksheet is amply annotated. Students who have moderate skill with the software and have learned about regression analysis from the curve-fitting worksheets described in this column will be able to complete and understand their analysis of the IR spectrum of HCl. The three Mathcad worksheets described here stretch the physical chemistry curriculum by presenting important topics in forms that students can use with only moderate Mathcad skills. The documents facilitate learning by giving students opportunities to interact with the material in meaningful ways in addition to using the documents as sources of techniques for building their own data-reduction worksheets. However, working through these Mathcad worksheets is not a trivial task for the average student. Support needs to be provided by the instructor to ease students through more advanced mathematical and Mathcad processes. These worksheets raise the question of how much we can ask diligent students to do in one course and how much time they need to spend to master the essential concepts of that course. The Mathcad documents and associated PDF versions are available at the JCE Internet WWW site. The Mathcad documents require Mathcad version 6.0 or higher and the PDF files require Adobe Acrobat. Every effort has been made to make the documents fully compatible across the various Mathcad versions. Users may need to refer to Mathcad manuals for functions that vary with the Mathcad version number. Literature Cited 1. Bevington, P. R. Data Reduction and Error Analysis for the Physical Sciences; McGraw-Hill: New York, 1969. 2. Zielinski, T. J.; Allendoerfer, R. D. J. Chem. Educ. 1997, 74, 1001. 3. Schwenz, R. W.; Polik, W. F. J. Chem. Educ. 1999, 76, 1302.

Bayesian Inference and Application of Robust Growth Curve Models Using Student's "t" Distribution

ERIC Educational Resources Information Center

Zhang, Zhiyong; Lai, Keke; Lu, Zhenqiu; Tong, Xin

2013-01-01

Despite the widespread popularity of growth curve analysis, few studies have investigated robust growth curve models. In this article, the "t" distribution is applied to model heavy-tailed data and contaminated normal data with outliers for growth curve analysis. The derived robust growth curve models are estimated through Bayesian…

SpecTracer: A Python-Based Interactive Solution for Echelle Spectra Reduction

NASA Astrophysics Data System (ADS)

Romero Matamala, Oscar Fernando; Petit, Véronique; Caballero-Nieves, Saida Maria

2018-01-01

SpecTracer is a newly developed interactive solution to reduce cross dispersed echelle spectra. The use of widgets saves the user the steep learning curves of currently available reduction software. SpecTracer uses well established image processing techniques based on IRAF to succesfully extract the stellar spectra. Comparisons with other reduction software, like IRAF, show comparable results, with the added advantages of ease of use, platform independence and portability. This tool can obtain meaningful scientific data and serve also as a training tool, especially for undergraduates doing research, in the procedure for spectroscopic analysis.

Defining the learning curve of laparoendoscopic single-site Heller myotomy.

PubMed

Ross, Sharona B; Luberice, Kenneth; Kurian, Tony J; Paul, Harold; Rosemurgy, Alexander S

2013-08-01

Initial outcomes suggest laparoendoscopic single-site (LESS) Heller myotomy with anterior fundoplication provides safe, efficacious, and cosmetically superior outcomes relative to conventional laparoscopy. This study was undertaken to define the learning curve of LESS Heller myotomy with anterior fundoplication. One hundred patients underwent LESS Heller myotomy with anterior fundoplication. Symptom frequency and severity were scored using a Likert scale (0 = never/not bothersome to 10 = always/very bothersome). Symptom resolution, additional trocars, and complications were compared among patient quartiles. Median data are presented. Preoperative frequency/severity scores were: dysphagia = 10/8 and regurgitation = 8/7. Additional trocars were placed in 12 patients (10%), of whom all were in the first two quartiles. Esophagotomy/gastrotomy occurred in three patients. Postoperative complications occurred in 9 per cent. No conversions to "open" operations occurred. Length of stay was 1 day. Postoperative frequency/severity scores were: dysphagia = 2/0 and regurgitation = 0/0; scores were less than before myotomy (P < 0.001). There were no apparent scars, except where additional trocars were placed. LESS Heller myotomy with anterior fundoplication well palliates symptoms of achalasia with no apparent scar. Placement of additional trocars only occurred early in the experience. For surgeons proficient with the conventional laparoscopic approach, the learning curve of LESS Heller myotomy with anterior fundoplication is short and safe, because proficiency is quickly attained.

Learning curve of thyroid fine-needle aspiration biopsy.

PubMed

Penín, Manuel; Martín, M Ángeles; San Millán, Beatriz; García, Juana

2017-12-01

Fine-needle aspiration biopsy (FNAB) is the reference procedure for thyroid nodule evaluation. Its main limitation are inadequate samples, which should be less than 20%. To analyze the learning curve of the procedure by comparing the results of a non-experienced endocrinologist (endocrinologist 2) to those of an experienced one (endocrinologist 1). Sixty FNABs were analyzed from February to June 2016. Each endocrinologist made 2punctures of every nodule in a random order. This order and the professional making every puncture were unknown to the pathologist who examined the samples. Endocrinologist 1 had a higher percentage of diagnoses than endocrinologist 2 (82% vs. 72%, P=.015). In the first 20 FNABs, the difference between both physicians was remarkable and statistically significant (80% vs. 50%, P=.047). In the following 20 FNABs, the difference narrowed and was not statistically significant (90% vs. 65%, P=.058). In the final 20 FNABs, the difference was minimal and not statistically significant (75% vs. 70%, P=.723). The learning curve of ultrasound-guided FNAB may be completed in a suitable environment by performing it at least 60 times. Although the guidelines recommend at least 3punctures per nodule, 2are enough to achieve an accurate percentage of diagnoses. Copyright © 2017 SEEN y SED. Publicado por Elsevier España, S.L.U. All rights reserved.

Outcomes of robotic versus laparoscopic surgery for mid and low rectal cancer after neoadjuvant chemoradiation therapy and the effect of learning curve

PubMed Central

Huang, Yu-Min; Huang, Yan Jiun; Wei, Po-Li

2017-01-01

Abstract Randomized controlled trials have demonstrated that laparoscopic surgery for rectal cancer is safe and can accelerate recovery without compromising oncological outcomes. However, such a surgery is technically demanding, limiting its application in nonspecialized centers. The operational features of a robotic system may facilitate overcoming this limitation. Studies have reported the potential advantages of robotic surgery. However, only a few of them have featured the application of this surgery in patients with advanced rectal cancer undergoing neoadjuvant chemoradiation therapy (nCRT). From January 2012 to April 2015, after undergoing nCRT, 40 patients with mid or low rectal cancer were operated using the robotic approach at our institution. Another 38 patients who were operated using the conventional laparoscopic approach were matched to patients in the robotic group by sex, age, the body mass index, and procedure. All operations were performed by a single surgical team. The clinicopathological characteristics and short-term outcomes of these patients were compared. To assess the effect of the learning curve on the outcomes, patients in the robotic group were further subdivided into 2 groups according to the sequential order of their procedures, with an equal number of patients in each group. Their outcome measures were compared. The robotic and laparoscopic groups were comparable with regard to pretreatment characteristics, rectal resection type, and pathological examination result. After undergoing nCRT, more patients in the robotic group exhibited clinically advanced diseases. The complication rate was similar between the 2 groups. The operation time and the time to the resumption of a soft diet were significantly prolonged in the robotic group. Further analysis revealed that the difference was mainly observed in the first robotic group. No significant difference was observed between the second robotic and laparoscopic groups. Although the robotic approach may offer potential advantages for rectal surgery, comparable short-term outcomes may be achieved when laparoscopic surgery is performed by experienced surgeons. However, our results suggested a shorter learning curve for robotic surgery for rectal cancer, even in patients who exhibited more advanced disease after undergoing nCRT. PMID:28984767

Automatic assessment of average diaphragm motion trajectory from 4DCT images through machine learning.

PubMed

Li, Guang; Wei, Jie; Huang, Hailiang; Gaebler, Carl Philipp; Yuan, Amy; Deasy, Joseph O

2015-12-01

To automatically estimate average diaphragm motion trajectory (ADMT) based on four-dimensional computed tomography (4DCT), facilitating clinical assessment of respiratory motion and motion variation and retrospective motion study. We have developed an effective motion extraction approach and a machine-learning-based algorithm to estimate the ADMT. Eleven patients with 22 sets of 4DCT images (4DCT1 at simulation and 4DCT2 at treatment) were studied. After automatically segmenting the lungs, the differential volume-per-slice (dVPS) curves of the left and right lungs were calculated as a function of slice number for each phase with respective to the full-exhalation. After 5-slice moving average was performed, the discrete cosine transform (DCT) was applied to analyze the dVPS curves in frequency domain. The dimensionality of the spectrum data was reduced by using several lowest frequency coefficients ( f v ) to account for most of the spectrum energy (Σ f v 2 ). Multiple linear regression (MLR) method was then applied to determine the weights of these frequencies by fitting the ground truth-the measured ADMT, which are represented by three pivot points of the diaphragm on each side. The 'leave-one-out' cross validation method was employed to analyze the statistical performance of the prediction results in three image sets: 4DCT1, 4DCT2, and 4DCT1 + 4DCT2. Seven lowest frequencies in DCT domain were found to be sufficient to approximate the patient dVPS curves ( R = 91%-96% in MLR fitting). The mean error in the predicted ADMT using leave-one-out method was 0.3 ± 1.9 mm for the left-side diaphragm and 0.0 ± 1.4 mm for the right-side diaphragm. The prediction error is lower in 4DCT2 than 4DCT1, and is the lowest in 4DCT1 and 4DCT2 combined. This frequency-analysis-based machine learning technique was employed to predict the ADMT automatically with an acceptable error (0.2 ± 1.6 mm). This volumetric approach is not affected by the presence of the lung tumors, providing an automatic robust tool to evaluate diaphragm motion.

Millennials Invading: Building Training for Today's Admissions Counselors

ERIC Educational Resources Information Center

Barnds, W. Kent

2009-01-01

As chief admissions officer at two small colleges, the author has been responsible, in part, for ensuring that entry-level admissions counselors are trained properly. He learned through trial and error, and has adapted his methods to be increasingly sensitive to the learning curve of new employees. His thoughts about training new admissions…

Learning, Retention, and Forgetting of Newton's Third Law throughout University Physics

ERIC Educational Resources Information Center

Sayre, Eleanor C.; Franklin, Scott V.; Dymek, Stephanie; Clark, Jessica; Sun, Yifei

2012-01-01

We present data from a between-student study on student response to questions on Newton's third law given in two introductory calculus-based physics classes (Mechanics and Electromagnetism) at a large northeastern university. Construction of a response curve reveals subtle dynamics in student learning not capturable by pretesting and post-testing.…

Fractal dimension based damage identification incorporating multi-task sparse Bayesian learning

NASA Astrophysics Data System (ADS)

Huang, Yong; Li, Hui; Wu, Stephen; Yang, Yongchao

2018-07-01

Sensitivity to damage and robustness to noise are critical requirements for the effectiveness of structural damage detection. In this study, a two-stage damage identification method based on the fractal dimension analysis and multi-task Bayesian learning is presented. The Higuchi’s fractal dimension (HFD) based damage index is first proposed, directly examining the time-frequency characteristic of local free vibration data of structures based on the irregularity sensitivity and noise robustness analysis of HFD. Katz’s fractal dimension is then presented to analyze the abrupt irregularity change of the spatial curve of the displacement mode shape along the structure. At the second stage, the multi-task sparse Bayesian learning technique is employed to infer the final damage localization vector, which borrow the dependent strength of the two fractal dimension based damage indication information and also incorporate the prior knowledge that structural damage occurs at a limited number of locations in a structure in the absence of its collapse. To validate the capability of the proposed method, a steel beam and a bridge, named Yonghe Bridge, are analyzed as illustrative examples. The damage identification results demonstrate that the proposed method is capable of localizing single and multiple damages regardless of its severity, and show superior robustness under heavy noise as well.

Studying Weather and Climate Using Atmospheric Retrospective Analyses

NASA Astrophysics Data System (ADS)

Bosilovich, M. G.

2014-12-01

Over the last 35 years, tremendous amounts of satellite observations of the Earth's atmosphere have been collected along side the much longer and diverse record of in situ measurements. The satellite data records have disparate qualities, structure and uncertainty which make comparing weather from the 80s and 2000s a challenging prospect. Likewise, in-situ data records lack complete coverage of the earth in both space and time. Atmospheric reanalyses use the observations with numerical models and data assimilation to produce continuous and consistent weather data records for periods longer than decades. The result is a simplified data format with a relatively straightforward learning curve that includes many more variables available (through the modeling component of the system), but driven by a full suite of observational data. The simplified data format allows introduction into weather and climate data analysis. Some examples are provided from undergraduate meteorology program internship projects. We will present the students progression through the projects from their initial understanding and competencies to some final results and the skills learned along the way. Reanalyses are a leading research tool in weather and climate, but can also provide an introductory experience as well, allowing students to develop an understanding of the physical system while learning basic programming and analysis skills.

Preliminary Results and Learning Curve of the Minimally Invasive Chevron Akin Operation for Hallux Valgus.

PubMed

Jowett, Charlie R J; Bedi, Harvinder S

Minimally invasive surgery is increasing in popularity. It is relevant in hallux valgus surgery owing to the potential for reduced disruption of the soft tissues and improved wound healing. We present our results and assess the learning curve of the minimally invasive Chevron Akin operation for hallux valgus. A total of 120 consecutive feet underwent minimally invasive Chevron Akin for symptomatic hallux valgus, of which 14 were excluded. They were followed up for a mean of 25 (range 18 to 38) months. The patients were clinically assessed using the American Orthopaedic Foot and Ankle Society score. Complications and patient satisfaction were recorded. The radiographs were analyzed and measurements recorded for hallux valgus and intermetatarsal angle correction. The mean age of the patients undergoing surgery was 55 (range 25 to 81) years. Of the 78 patients, 76 (97.4%) were female and 2 (2.6%) were male; 28 (35.9%) cases were bilateral. The mean American Orthopaedic Foot and Ankle Society score improved from 56 (range 23 to 76) preoperatively to 87 (range 50 to 100) postoperatively (p < .001). The mean hallux valgus and intermetatarsal angles preoperatively were 29.7° (range 12° to 46°) and 14.0° (range 8° to 20°). The corresponding postoperative angles were 10.3° (range 0° to 25°) and 7.6° (range 3° to 15°; p < .001). The patients were satisfied with the results of surgery in 87% of cases (92 of 106). The incidence of reoperation was 14% (15 of 106). These are the only reported results for this technique. They display a steep associated learning curve. However, the results are promising, and the learning curve is comparable to that for open hallux valgus surgery. Copyright © 2017 American College of Foot and Ankle Surgeons. Published by Elsevier Inc. All rights reserved.

Learning curves and perioperative outcomes after endoscopic enucleation of the prostate: a comparison between GreenLight 532-nm and holmium lasers.

PubMed

Peyronnet, Benoit; Robert, Grégoire; Comat, Vincent; Rouprêt, Morgan; Gomez-Sancha, Fernando; Cornu, Jean-Nicolas; Misrai, Vincent

2017-06-01

To compare the learning curves, perioperative and early functional outcomes after HoLEP and GreenLEP. Data from the first 100 consecutive cases treated by GreenLEP and HoLEP by two surgeons were prospectively collected from dedicated databases and analysed retrospectively. En-bloc GreenLEP and two-lobar HoLEP enucleations were conducted using the GreenLight HPS™ 2090 laser and Lumenis™ holmium laser. Patients' characteristics, perioperative outcomes and functional outcomes after 1, 3 and 6 months were compared between groups. Total energy delivered and operative times were significantly shorter for GreenLEP (58 vs. 110 kJ, p < 0.0001; 60 vs. 90 min, p < 0.0001). Operative time reached a plateau after 30 procedures in each group. Length of catheterization and hospital stay were significantly shorter in the HoLEP group (2 vs. 1 day, p < 0.0001; 2 vs. 1 day, p < 0.0001). Postoperative complications were comparable between GreenLEP and HoLEP (19 vs. 25 %; p = 0.13). There was a greater increase of Q max at 3 months and a greater IPSS decrease at 1 month for GreenLEP, whereas decreases in IPSS and IPSS-Q8 at 6 months were greater for HoLEP. Transient stress urinary incontinence was comparable between both groups (6 vs. 9 % at 3 months; p = 0.42). Pentafecta was achieved in four consecutive patients after the 18th and the 40th procedure in the GreenLEP and HoLEP group, respectively. Learning curves ranged from 14 to 30 cases for GreenLEP and 22 to 40 cases for HoLEP. Learning curves of GreenLEP and HoLEP provided roughly similar peri-operative and short-term functional outcomes.

Nasoenteral feeding tube placement by nurses using an electromagnetic guidance system (with video).

PubMed

Mathus-Vliegen, Elisabeth M H; Duflou, Ann; Spanier, Marcel B W; Fockens, Paul

2010-04-01

The early institution of feeding in patients who need postpyloric feeding tubes is often hampered by a limited availability of endoscopists experienced in safe tube positioning. To test the feasibility of having nurses place postpyloric feeding tubes by using a universal path finding system device. Prospective study. Academic hospital. The success rate and learning curve of a senior nurse placing postpyloric feeding tubes in 50 patients was studied, followed by a study in 160 patients on the success rates and learning curves of 4 inexperienced nurses instructed by the senior nurse. Finally, the success rate of postpyloric feeding tube placement by the senior nurse in 50 critically ill patients was investigated. Postpyloric feeding tube positioning by nurses using an electromagnetic universal path-finding system device enabling them to follow the path of the tip of the feeding tube on a monitor screen. Success was defined by postpyloric positioning of the feeding tube. The ultimate aim was to reach at least the duodenojejunal flexure. In the first part, the senior nurse was successful in 72% of cases. There was a clear learning curve. In the second part, the 4 newly instructed nurses had a success rate of 89.4% without an evident learning curve. In the third part, successful feeding tube positioning was achieved in 78% of critically ill patients. Of the 217 successfully positioned tubes, 74% reached at least the duodenojejunal flexure. In half of the unsuccessful cases, an explanation for the failure was found at endoscopy. No complications were seen. The generalization to less-specialized hospitals should be investigated. Postpyloric positioning of feeding tubes by nurses at the bedside without endoscopy is feasible and safe. Nurses may take over some of the tasks of doctors in a time of high endoscopic needs. Copyright 2010 American Society for Gastrointestinal Endoscopy. Published by Mosby, Inc. All rights reserved.

Learning curve for the management of tyrosine kinase inhibitors as the first line of treatment for patients with metastatic renal cancer.

PubMed

Lendínez-Cano, G; Osman García, I; Congregado Ruiz, C B; Conde Sánchez, J M; Medina López, R A

2018-03-07

To analyse the learning curve for the management of tyrosine kinase inhibitors as the first line of treatment for patients with metastatic renal cancer. We evaluated 32 consecutive patients treated in our department for metastatic renal cancer with tyrosine kinase inhibitors (pazopanib or sunitinib) as first-line treatment between September 2012 and November 2015. We retrospectively analysed this sample. We measured the time to the withdrawal of the first-line treatment, the time to progression and overall survival using Kaplan-Meier curves. The learning curve was analysed with the cumulative sum (CUSUM) methodology. In our series, the median time to the withdrawal of the first-line treatment was 11 months (95% CI 4.9-17.1). The mean time to progression was 30.4 months (95% CI 22.7-38.1), and the mean overall survival was 34.9 months (95% CI 27.8-42). By applying the CUSUM methodology, we obtained a graph for the CUSUM value of the time to withdrawal of the first-line treatment (CUSUM TW), observing 3 well-differentiated phases: phase 1 or initial learning phase (1-15), phase 2 (16-26) in which the management of the drug progressively improved and phase 3 (27-32) of maximum experience or mastery of the management of these drugs. The number of treated patients needed to achieve the proper management of these patients was estimated at 15. Despite the limitations of the sample size and follow-up time, we estimated (in 15 patients) the number needed to reach the necessary experience in the management of these patients with tyrosine kinase inhibitors. We observed no relationship between the time to the withdrawal of the first-line treatment for any cause and progression. Copyright © 2018 AEU. Publicado por Elsevier España, S.L.U. All rights reserved.

Using a course pilot in the development of an online problem-based learning (PBL) therapeutics course in a post-professional PharmD program.

PubMed

Nagge, Jeff J; Killeen, Rosemary; Jennings, Brad

2018-02-01

To assess whether the traditional problem-based learning (PBL) process can be replicated in an online environment, and to identify any barriers and facilitators to learning using a course pilot. Eight alumni and one experienced tutor participated in a two-week simulated PBL course comprised of two three-hour synchronous online tutorials. Blackboard Collaborate ® software was used to permit audio and visual interaction. The PBL tutorials were recorded and observed by the researchers. Participants completed satisfaction surveys after the pilot, and were invited to take part in a focus group to debrief about their experience. Once the steep learning curve with the technology was overcome, the quality of the PBL process was similar in the online course as it was in the face-to-face course. Several key factors for success were identified through analysis of the videotaped sessions, and interviews with the participants in the course pilot. Conducting a course pilot study demonstrated that an online PBL course is feasible, and identified some considerations to facilitate success. Copyright © 2017 Elsevier Inc. All rights reserved.

Brain-state invariant thalamo-cortical coordination revealed by non-linear encoders.

PubMed

Viejo, Guillaume; Cortier, Thomas; Peyrache, Adrien

2018-03-01

Understanding how neurons cooperate to integrate sensory inputs and guide behavior is a fundamental problem in neuroscience. A large body of methods have been developed to study neuronal firing at the single cell and population levels, generally seeking interpretability as well as predictivity. However, these methods are usually confronted with the lack of ground-truth necessary to validate the approach. Here, using neuronal data from the head-direction (HD) system, we present evidence demonstrating how gradient boosted trees, a non-linear and supervised Machine Learning tool, can learn the relationship between behavioral parameters and neuronal responses with high accuracy by optimizing the information rate. Interestingly, and unlike other classes of Machine Learning methods, the intrinsic structure of the trees can be interpreted in relation to behavior (e.g. to recover the tuning curves) or to study how neurons cooperate with their peers in the network. We show how the method, unlike linear analysis, reveals that the coordination in thalamo-cortical circuits is qualitatively the same during wakefulness and sleep, indicating a brain-state independent feed-forward circuit. Machine Learning tools thus open new avenues for benchmarking model-based characterization of spike trains.

Brain-state invariant thalamo-cortical coordination revealed by non-linear encoders

PubMed Central

Cortier, Thomas; Peyrache, Adrien

2018-01-01

Understanding how neurons cooperate to integrate sensory inputs and guide behavior is a fundamental problem in neuroscience. A large body of methods have been developed to study neuronal firing at the single cell and population levels, generally seeking interpretability as well as predictivity. However, these methods are usually confronted with the lack of ground-truth necessary to validate the approach. Here, using neuronal data from the head-direction (HD) system, we present evidence demonstrating how gradient boosted trees, a non-linear and supervised Machine Learning tool, can learn the relationship between behavioral parameters and neuronal responses with high accuracy by optimizing the information rate. Interestingly, and unlike other classes of Machine Learning methods, the intrinsic structure of the trees can be interpreted in relation to behavior (e.g. to recover the tuning curves) or to study how neurons cooperate with their peers in the network. We show how the method, unlike linear analysis, reveals that the coordination in thalamo-cortical circuits is qualitatively the same during wakefulness and sleep, indicating a brain-state independent feed-forward circuit. Machine Learning tools thus open new avenues for benchmarking model-based characterization of spike trains. PMID:29565979

Conditional withholding of proboscis extension in honeybees (Apis mellifera) during discriminative punishment.

PubMed

Smith, B H; Abramson, C I; Tobin, T R

1991-12-01

Proboscis extension conditioning of honeybee workers was used to test the ability of bees to respond to appetitive and aversive stimuli while restrained in a harness that allows subjects to move their antennae and mouthparts (Kuwabara, 1957; Menzel, Erber, & Masuhr, 1974). Subjects were conditioned to discriminate between two odors, one associated with sucrose feeding and the other associated with a 10 V AC shock if they responded to the sucrose unconditioned stimulus (US) in the context of that odor. Most Ss readily learned to respond to the odor followed by sucrose feeding and not to the odor associated with sucrose stimulation plus shock. Furthermore, in the context of the odor associated with shock, significantly more subjects withheld or delayed proboscis extension on stimulation with the sucrose US than they did in the context of the odor associated with feeding. Thus, restrained honeybees can readily learn to avoid shock according to an odor context by withholding proboscis extension to a normally powerful releaser. Analysis of individual learning curves revealed that subjects differed markedly in performance on this task. Some learn the discrimination quickly, whereas others show different kinds of response patterns.

ENKI - A tool for analysing the learning efficiency

NASA Astrophysics Data System (ADS)

Simona, Dudáková; Boris, Lacsný; Aba, Teleki

2017-01-01

Long-term memory plays a crucial role in learning mechanisms. We start to build up a probability model of learning (ENKI) ten years ago based on findings of micro genetics published in [1]. We accomplished a number of experiments in our department to testify the validity of the model with success. We described ENKI in detail here, giving the general mathematical formula of the learning curve. This paper pointed out that the model ENKI can detect its own strategy of learning in the brain as well as the simulation of the process of learning that will lead to the development of this method using its own strategy.

Automated Dermoscopy Image Analysis of Pigmented Skin Lesions

PubMed Central

Baldi, Alfonso; Quartulli, Marco; Murace, Raffaele; Dragonetti, Emanuele; Manganaro, Mario; Guerra, Oscar; Bizzi, Stefano

2010-01-01

Dermoscopy (dermatoscopy, epiluminescence microscopy) is a non-invasive diagnostic technique for the in vivo observation of pigmented skin lesions (PSLs), allowing a better visualization of surface and subsurface structures (from the epidermis to the papillary dermis). This diagnostic tool permits the recognition of morphologic structures not visible by the naked eye, thus opening a new dimension in the analysis of the clinical morphologic features of PSLs. In order to reduce the learning-curve of non-expert clinicians and to mitigate problems inherent in the reliability and reproducibility of the diagnostic criteria used in pattern analysis, several indicative methods based on diagnostic algorithms have been introduced in the last few years. Recently, numerous systems designed to provide computer-aided analysis of digital images obtained by dermoscopy have been reported in the literature. The goal of this article is to review these systems, focusing on the most recent approaches based on content-based image retrieval systems (CBIR). PMID:24281070

Machine learning-based analysis of MR radiomics can help to improve the diagnostic performance of PI-RADS v2 in clinically relevant prostate cancer.

PubMed

Wang, Jing; Wu, Chen-Jiang; Bao, Mei-Ling; Zhang, Jing; Wang, Xiao-Ning; Zhang, Yu-Dong

2017-10-01

To investigate whether machine learning-based analysis of MR radiomics can help improve the performance PI-RADS v2 in clinically relevant prostate cancer (PCa). This IRB-approved study included 54 patients with PCa undergoing multi-parametric (mp) MRI before prostatectomy. Imaging analysis was performed on 54 tumours, 47 normal peripheral (PZ) and 48 normal transitional (TZ) zone based on histological-radiological correlation. Mp-MRI was scored via PI-RADS, and quantified by measuring radiomic features. Predictive model was developed using a novel support vector machine trained with: (i) radiomics, (ii) PI-RADS scores, (iii) radiomics and PI-RADS scores. Paired comparison was made via ROC analysis. For PCa versus normal TZ, the model trained with radiomics had a significantly higher area under the ROC curve (Az) (0.955 [95% CI 0.923-0.976]) than PI-RADS (Az: 0.878 [0.834-0.914], p < 0.001). The Az between them was insignificant for PCa versus PZ (0.972 [0.945-0.988] vs. 0.940 [0.905-0.965], p = 0.097). When radiomics was added, performance of PI-RADS was significantly improved for PCa versus PZ (Az: 0.983 [0.960-0.995]) and PCa versus TZ (Az: 0.968 [0.940-0.985]). Machine learning analysis of MR radiomics can help improve the performance of PI-RADS in clinically relevant PCa. • Machine-based analysis of MR radiomics outperformed in TZ cancer against PI-RADS. • Adding MR radiomics significantly improved the performance of PI-RADS. • DKI-derived Dapp and Kapp were two strong markers for the diagnosis of PCa.

Analysis of In-Situ Spectral Reflectance of Sago and Other Palms: Implications for Their Detection in Optical Satellite Images

NASA Astrophysics Data System (ADS)

Rendon Santillan, Jojene; Makinano-Santillan, Meriam

2018-04-01

We present a characterization, comparison and analysis of in-situ spectral reflectance of Sago and other palms (coconut, oil palm and nipa) to ascertain on which part of the electromagnetic spectrum these palms are distinguishable from each other. The analysis also aims to reveal information that will assist in selecting which band to use when mapping Sago palms using the images acquired by these sensors. The datasets used in the analysis consisted of averaged spectral reflectance curves of each palm species measured within the 345-1045 nm wavelength range using an Ocean Optics USB4000-VIS-NIR Miniature Fiber Optic Spectrometer. This in-situ reflectance data was also resampled to match the spectral response of the 4 bands of ALOS AVNIR-2, 3 bands of ASTER VNIR, 4 bands of Landsat 7 ETM+, 5 bands of Landsat 8, and 8 bands of Worldview-2 (WV2). Examination of the spectral reflectance curves showed that the near infra-red region, specifically at 770, 800 and 875 nm, provides the best wavelengths where Sago palms can be distinguished from other palms. The resampling of the in-situ reflectance spectra to match the spectral response of optical sensors made possible the analysis of the differences in reflectance values of Sago and other palms in different bands of the sensors. Overall, the knowledge learned from the analysis can be useful in the actual analysis of optical satellite images, specifically in determining which band to include or to exclude, or whether to use all bands of a sensor in discriminating and mapping Sago palms.

Proceedings of the Defense Acquisition Reform: Challenge to Government, Industry, and Academia, Held at Washington, DC on 26 April 1994.

DTIC Science & Technology

1994-04-26

Obviously what we want to do is come down to ber of problems we have had on these contracts, and it the bottom of the cost curve . Many years ago, 50 years...many of our systems the military and answer your first one. To be realistic, there is not primes have had many years of learning curve , not only much...risk levels. In this section I offer some assertions to add more precision to the shape of the boundary curves . D-4 * Assertion 1: Probability is the

[Nootropic and analgesic effects of Semax following different routes of administration].

PubMed

Manchenko, D M; Glazova, N Iu; Levitskaia, N G; Andreeva, L A; Kamenskiĭ, A A; Miasoedov, N F

2010-10-01

Heptapeptide Semax (MEHFPGP) is the fragment of ACTH(4-10) analogue with prolonged neurotropic activity. The aim of the present work was to study the Semax effects on learning capability and pain sensitivity in white rats following intraperitoneal and intranasal administration in different doses. Semax nootropic effects were studied in the test of acquisition of passive avoidance task. Pain sensitivity was estimated in Randall-Selitto paw-withdrawal test. It was shown that Semax exerts nootropic and analgesic activities following intraperitoneal administration. Analysis of dependence of these effects on dose resulted in different dose-response curves. Following intranasal administration, Semax was more potent in learning improvement compared to intraperitoneal administration. The peptide failed to affect the animal pain sensitivity following intranasal administration as opposed to intraperitoneal administration. The data obtained suggest different mechanisms and brain structures involved in realization of the nootropic and analgesic effects of Semax.

[Simulation in surgical training].

PubMed

Nabavi, A; Schipper, J

2017-01-01

Patient safety during operations hinges on the surgeon's skills and abilities. However, surgical training has come under a variety of restrictions. To acquire dexterity with decreasingly "simple" cases, within the legislative time constraints and increasing expectations for surgical results is the future challenge. Are there alternatives to traditional master-apprentice learning? A literature review and analysis of the development, implementation, and evaluation of surgical simulation are presented. Simulation, using a variety of methods, most important physical and virtual (computer-generated) models, provides a safe environment to practice basic and advanced skills without endangering patients. These environments have specific strengths and weaknesses. Simulations can only serve to decrease the slope of learning curves, but cannot be a substitute for the real situation. Thus, they have to be an integral part of a comprehensive training curriculum. Our surgical societies have to take up that challenge to ensure the training of future generations.

Comparison of success rates, learning curves, and inter-subject performance variability of robot-assisted and manual ultrasound-guided nerve block needle guidance in simulation.

PubMed

Morse, J; Terrasini, N; Wehbe, M; Philippona, C; Zaouter, C; Cyr, S; Hemmerling, T M

2014-06-01

This study focuses on a recently developed robotic nerve block system and its impact on learning regional anaesthesia skills. We compared success rates, learning curves, performance times, and inter-subject performance variability of robot-assisted vs manual ultrasound (US)-guided nerve block needle guidance. The hypothesis of this study is that robot assistance will result in faster skill acquisition than manual needle guidance. Five co-authors with different experience with nerve blocks and the robotic system performed both manual and robot-assisted, US-guided nerve blocks on two different nerves of a nerve phantom. Ten trials were performed for each of the four procedures. Time taken to move from a shared starting position till the needle was inserted into the target nerve was defined as the performance time. A successful block was defined as the insertion of the needle into the target nerve. Average performance times were compared using analysis of variance. P<0.05 was considered significant. Data presented as mean (standard deviation). All blocks were successful. There were significant differences in performance times between co-authors to perform the manual blocks, either superficial (P=0.001) or profound (P=0.0001); no statistical difference between co-authors was noted for the robot-assisted blocks. Linear regression indicated that the average decrease in time between consecutive trials for robot-assisted blocks of 1.8 (1.6) s was significantly (P=0.007) greater than the decrease for manual blocks of 0.3 (0.3) s. Robot assistance of nerve blocks allows for faster learning of needle guidance over manual positioning and reduces inter-subject performance variability. © The Author [2014]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Dynamics of Hippocampal Protein Expression During Long-term Spatial Memory Formation*

PubMed Central

Borovok, Natalia; Nesher, Elimelech; Levin, Yishai; Reichenstein, Michal; Pinhasov, Albert

2016-01-01

Spatial memory depends on the hippocampus, which is particularly vulnerable to aging. This vulnerability has implications for the impairment of navigation capacities in older people, who may show a marked drop in performance of spatial tasks with advancing age. Contemporary understanding of long-term memory formation relies on molecular mechanisms underlying long-term synaptic plasticity. With memory acquisition, activity-dependent changes occurring in synapses initiate multiple signal transduction pathways enhancing protein turnover. This enhancement facilitates de novo synthesis of plasticity related proteins, crucial factors for establishing persistent long-term synaptic plasticity and forming memory engrams. Extensive studies have been performed to elucidate molecular mechanisms of memory traces formation; however, the identity of plasticity related proteins is still evasive. In this study, we investigated protein turnover in mouse hippocampus during long-term spatial memory formation using the reference memory version of radial arm maze (RAM) paradigm. We identified 1592 proteins, which exhibited a complex picture of expression changes during spatial memory formation. Variable linear decomposition reduced significantly data dimensionality and enriched three principal factors responsible for variance of memory-related protein levels at (1) the initial phase of memory acquisition (165 proteins), (2) during the steep learning improvement (148 proteins), and (3) the final phase of the learning curve (123 proteins). Gene ontology and signaling pathways analysis revealed a clear correlation between memory improvement and learning phase-curbed expression profiles of proteins belonging to specific functional categories. We found differential enrichment of (1) neurotrophic factors signaling pathways, proteins regulating synaptic transmission, and actin microfilament during the first day of the learning curve; (2) transcription and translation machinery, protein trafficking, enhancement of metabolic activity, and Wnt signaling pathway during the steep phase of memory formation; and (3) cytoskeleton organization proteins. Taken together, this study clearly demonstrates dynamic assembly and disassembly of protein-protein interaction networks depending on the stage of memory formation engrams. PMID:26598641

Last-position elimination-based learning automata.

PubMed

Zhang, Junqi; Wang, Cheng; Zhou, MengChu

2014-12-01

An update scheme of the state probability vector of actions is critical for learning automata (LA). The most popular is the pursuit scheme that pursues the estimated optimal action and penalizes others. This paper proposes a reverse philosophy that leads to last-position elimination-based learning automata (LELA). The action graded last in terms of the estimated performance is penalized by decreasing its state probability and is eliminated when its state probability becomes zero. All active actions, that is, actions with nonzero state probability, equally share the penalized state probability from the last-position action at each iteration. The proposed LELA is characterized by the relaxed convergence condition for the optimal action, the accelerated step size of the state probability update scheme for the estimated optimal action, and the enriched sampling for the estimated nonoptimal actions. The proof of the ϵ-optimal property for the proposed algorithm is presented. Last-position elimination is a widespread philosophy in the real world and has proved to be also helpful for the update scheme of the learning automaton via the simulations of well-known benchmark environments. In the simulations, two versions of the LELA, using different selection strategies of the last action, are compared with the classical pursuit algorithms Discretized Pursuit Reward-Inaction (DP(RI)) and Discretized Generalized Pursuit Algorithm (DGPA). Simulation results show that the proposed schemes achieve significantly faster convergence and higher accuracy than the classical ones. Specifically, the proposed schemes reduce the interval to find the best parameter for a specific environment in the classical pursuit algorithms. Thus, they can have their parameter tuning easier to perform and can save much more time when applied to a practical case. Furthermore, the convergence curves and the corresponding variance coefficient curves of the contenders are illustrated to characterize their essential differences and verify the analysis results of the proposed algorithms.

[Initiating a Robotic Program for Abdominal Surgery - Experiences from a Centre in Germany].

PubMed

Brunner, Maximilian; Matzel, Klaus; Aladashvili, Archil; Krautz, Christian; Grützmann, Robert; Croner, Roland

2018-05-18

Robotic systems are becoming increasingly important in abdominal surgery. We describe the implementation of a robotic program at a German centre for abdominal surgery, with focus on feasibility, safety, patient selection, learning curves, financial aspects and the lessons learned. This retrospective analysis covered data on patient demographics, intra- and postoperative parameters, oncological results and costs of all robotic-assisted abdominal operations performed at our institution between August 2012 to December 2016. It was also evaluated how possible factors for preoperative patient selection might influence intra- or postoperative outcome and learning parameters. 81 operations were performed - mostly colorectal resections (n = 35), ventral mesh rectopexy (n = 23) and liver resections (n = 18). The conversion rate was 7%. All oncological patients underwent R0 resection. Mean postoperative hospitalisation was 8.8 days; mean morbidity was 24%, with major complications (Clavien-Dindo > II) in 7%; mortality was 0%. BMI above 33.5 kg/m 2 was associated with significantly higher morbidity (p = 0.024) and rate of major complications (p = 0.046), as well as a significantly longer hospitalisation (p = 0.009). Patients older than 65 years had significantly higher morbidity (p = 0.025). With increasing numbers of operations, time of surgery decreased (p = 0.001). The average cost of a robot-assisted operation, including hospital stay, was 15,221 €. The costs of robotic sigmoid resections or liver resections were higher (compared to the open approach: 106.8 and 62.8% higher, respectively, compared to the laparoscopic approach 93.5 and 66.5% higher, respectively). Robotic surgery is a safe approach. A crucial factor in the successful and safe performance of robotic assisted operations is proper patient selection, especially during the implementation period. The inevitable learning curve and the higher costs compared to open and laparocopic surgery must be respected and specialisation of the whole team is necessary. Georg Thieme Verlag KG Stuttgart · New York.

Habitat suitability index curves for paddlefish, developed by the delphi technique

USGS Publications Warehouse

Crance, John H.

1987-01-01

A Delphi exercise conducted with a panel of 11 experts on paddlefish (Polyodon spathula) and an evaluator resulted in 14 riverine habitat suitability index curves associating various life stages or activities of paddlefish with four variables: velocity, depth, substrate type, and temperature. The panel reached a consensus on six of the curves and eight to 10 panelists agreed on the others. Several panelists reported that they found the Delphi exercise to be a good learning experience, and they believed the technique is an appropriate interim method for developing suitability index curves when available data are inadequate for more conventional statistical analyses. Documentation of good paddlefish spawning habitat was the data need most commonly identified by the panelists.

Using the weighted area under the net benefit curve for decision curve analysis.

PubMed

Talluri, Rajesh; Shete, Sanjay

2016-07-18

Risk prediction models have been proposed for various diseases and are being improved as new predictors are identified. A major challenge is to determine whether the newly discovered predictors improve risk prediction. Decision curve analysis has been proposed as an alternative to the area under the curve and net reclassification index to evaluate the performance of prediction models in clinical scenarios. The decision curve computed using the net benefit can evaluate the predictive performance of risk models at a given or range of threshold probabilities. However, when the decision curves for 2 competing models cross in the range of interest, it is difficult to identify the best model as there is no readily available summary measure for evaluating the predictive performance. The key deterrent for using simple measures such as the area under the net benefit curve is the assumption that the threshold probabilities are uniformly distributed among patients. We propose a novel measure for performing decision curve analysis. The approach estimates the distribution of threshold probabilities without the need of additional data. Using the estimated distribution of threshold probabilities, the weighted area under the net benefit curve serves as the summary measure to compare risk prediction models in a range of interest. We compared 3 different approaches, the standard method, the area under the net benefit curve, and the weighted area under the net benefit curve. Type 1 error and power comparisons demonstrate that the weighted area under the net benefit curve has higher power compared to the other methods. Several simulation studies are presented to demonstrate the improvement in model comparison using the weighted area under the net benefit curve compared to the standard method. The proposed measure improves decision curve analysis by using the weighted area under the curve and thereby improves the power of the decision curve analysis to compare risk prediction models in a clinical scenario.

Analysis and Recognition of Curve Type as The Basis of Object Recognition in Image

NASA Astrophysics Data System (ADS)

Nugraha, Nurma; Madenda, Sarifuddin; Indarti, Dina; Dewi Agushinta, R.; Ernastuti

2016-06-01

An object in an image when analyzed further will show the characteristics that distinguish one object with another object in an image. Characteristics that are used in object recognition in an image can be a color, shape, pattern, texture and spatial information that can be used to represent objects in the digital image. The method has recently been developed for image feature extraction on objects that share characteristics curve analysis (simple curve) and use the search feature of chain code object. This study will develop an algorithm analysis and the recognition of the type of curve as the basis for object recognition in images, with proposing addition of complex curve characteristics with maximum four branches that will be used for the process of object recognition in images. Definition of complex curve is the curve that has a point of intersection. By using some of the image of the edge detection, the algorithm was able to do the analysis and recognition of complex curve shape well.

Evaluation of PCR and high-resolution melt curve analysis for differentiation of Salmonella isolates.

PubMed

Saeidabadi, Mohammad Sadegh; Nili, Hassan; Dadras, Habibollah; Sharifiyazdi, Hassan; Connolly, Joanne; Valcanis, Mary; Raidal, Shane; Ghorashi, Seyed Ali

2017-06-01

Consumption of poultry products contaminated with Salmonella is one of the major causes of foodborne diseases worldwide and therefore detection and differentiation of Salmonella spp. in poultry is important. In this study, oligonucleotide primers were designed from hemD gene and a PCR followed by high-resolution melt (HRM) curve analysis was developed for rapid differentiation of Salmonella isolates. Amplicons of 228 bp were generated from 16 different Salmonella reference strains and from 65 clinical field isolates mainly from poultry farms. HRM curve analysis of the amplicons differentiated Salmonella isolates and analysis of the nucleotide sequence of the amplicons from selected isolates revealed that each melting curve profile was related to a unique DNA sequence. The relationship between reference strains and tested specimens was also evaluated using a mathematical model without visual interpretation of HRM curves. In addition, the potential of the PCR-HRM curve analysis was evaluated for genotyping of additional Salmonella isolates from different avian species. The findings indicate that PCR followed by HRM curve analysis provides a rapid and robust technique for genotyping of Salmonella isolates to determine the serovar/serotype.

Nonlinear dynamical modes of climate variability: from curves to manifolds

NASA Astrophysics Data System (ADS)

Gavrilov, Andrey; Mukhin, Dmitry; Loskutov, Evgeny; Feigin, Alexander

2016-04-01

The necessity of efficient dimensionality reduction methods capturing dynamical properties of the system from observed data is evident. Recent study shows that nonlinear dynamical mode (NDM) expansion is able to solve this problem and provide adequate phase variables in climate data analysis [1]. A single NDM is logical extension of linear spatio-temporal structure (like empirical orthogonal function pattern): it is constructed as nonlinear transformation of hidden scalar time series to the space of observed variables, i. e. projection of observed dataset onto a nonlinear curve. Both the hidden time series and the parameters of the curve are learned simultaneously using Bayesian approach. The only prior information about the hidden signal is the assumption of its smoothness. The optimal nonlinearity degree and smoothness are found using Bayesian evidence technique. In this work we do further extension and look for vector hidden signals instead of scalar with the same smoothness restriction. As a result we resolve multidimensional manifolds instead of sum of curves. The dimension of the hidden manifold is optimized using also Bayesian evidence. The efficiency of the extension is demonstrated on model examples. Results of application to climate data are demonstrated and discussed. The study is supported by Government of Russian Federation (agreement #14.Z50.31.0033 with the Institute of Applied Physics of RAS). 1. Mukhin, D., Gavrilov, A., Feigin, A., Loskutov, E., & Kurths, J. (2015). Principal nonlinear dynamical modes of climate variability. Scientific Reports, 5, 15510. http://doi.org/10.1038/srep15510

Automatic feature learning using multichannel ROI based on deep structured algorithms for computerized lung cancer diagnosis.

PubMed

Sun, Wenqing; Zheng, Bin; Qian, Wei

2017-10-01

This study aimed to analyze the ability of extracting automatically generated features using deep structured algorithms in lung nodule CT image diagnosis, and compare its performance with traditional computer aided diagnosis (CADx) systems using hand-crafted features. All of the 1018 cases were acquired from Lung Image Database Consortium (LIDC) public lung cancer database. The nodules were segmented according to four radiologists' markings, and 13,668 samples were generated by rotating every slice of nodule images. Three multichannel ROI based deep structured algorithms were designed and implemented in this study: convolutional neural network (CNN), deep belief network (DBN), and stacked denoising autoencoder (SDAE). For the comparison purpose, we also implemented a CADx system using hand-crafted features including density features, texture features and morphological features. The performance of every scheme was evaluated by using a 10-fold cross-validation method and an assessment index of the area under the receiver operating characteristic curve (AUC). The observed highest area under the curve (AUC) was 0.899±0.018 achieved by CNN, which was significantly higher than traditional CADx with the AUC=0.848±0.026. The results from DBN was also slightly higher than CADx, while SDAE was slightly lower. By visualizing the automatic generated features, we found some meaningful detectors like curvy stroke detectors from deep structured schemes. The study results showed the deep structured algorithms with automatically generated features can achieve desirable performance in lung nodule diagnosis. With well-tuned parameters and large enough dataset, the deep learning algorithms can have better performance than current popular CADx. We believe the deep learning algorithms with similar data preprocessing procedure can be used in other medical image analysis areas as well. Copyright © 2017. Published by Elsevier Ltd.

Changes in teachers' involvement versus rejection and links with academic motivation during the first year of secondary education: a multilevel growth curve analysis.

PubMed

Maulana, Ridwan; Opdenakker, Marie-Christine; Stroet, Kim; Bosker, Roel

2013-09-01

Research consistently shows that the learning environment plays an important role for early adolescents' learning and outcomes and suggests that good teacher-student relationships can serve as a protective factor for maintaining young adolescents' interest and active engagement in learning. However, less is known about the dynamic nature of teacher-student relationships and how they link with academic motivation development. Furthermore, little is known about the nature and the effects of teacher-student relationships in a cross-national context. The present study investigated changes in two components of teacher-student relationships (teachers' involvement vs. rejection) and examined links with students' academic motivation during the first grade of secondary school. Ten Dutch and ten Indonesian teachers (65 % female) from 24 classes were videoed 12 times across the school year, and four videos for each class were selected randomly and coded on teachers' involvement versus rejection. A total of 713 students (52 % girls) completed four-wave measures of their academic motivation after each video observation. Multilevel growth curve modeling revealed that the teacher's involvement changed in a curvilinear way and decreased across the first year of secondary education, while changes in the teacher's rejection did not follow a linear time function. Academic motivation changed in an undesirable way: controlled motivation increased, while autonomous motivation decreased over time. Teachers' involvement had a unique contribution in preventing high levels of controlled motivation in both countries. Findings suggest that teacher-student relationships (teachers' involvement) play an essential role in early adolescents' motivation regardless of the nations and should be a priority for schools.

Using Machine-Learned Bayesian Belief Networks to Predict Perioperative Risk of Clostridium Difficile Infection Following Colon Surgery

PubMed Central

Bilchik, Anton; Eberhardt, John; Kalina, Philip; Nissan, Aviram; Johnson, Eric; Avital, Itzhak; Stojadinovic, Alexander

2012-01-01

Background Clostridium difficile (C-Diff) infection following colorectal resection is an increasing source of morbidity and mortality. Objective We sought to determine if machine-learned Bayesian belief networks (ml-BBNs) could preoperatively provide clinicians with postoperative estimates of C-Diff risk. Methods We performed a retrospective modeling of the Nationwide Inpatient Sample (NIS) national registry dataset with independent set validation. The NIS registries for 2005 and 2006 were used for initial model training, and the data from 2007 were used for testing and validation. International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes were used to identify subjects undergoing colon resection and postoperative C-Diff development. The ml-BBNs were trained using a stepwise process. Receiver operating characteristic (ROC) curve analysis was conducted and area under the curve (AUC), positive predictive value (PPV), and negative predictive value (NPV) were calculated. Results From over 24 million admissions, 170,363 undergoing colon resection met the inclusion criteria. Overall, 1.7% developed postoperative C-Diff. Using the ml-BBN to estimate C-Diff risk, model AUC is 0.75. Using only known a priori features, AUC is 0.74. The model has two configurations: a high sensitivity and a high specificity configuration. Sensitivity, specificity, PPV, and NPV are 81.0%, 50.1%, 2.6%, and 99.4% for high sensitivity and 55.4%, 81.3%, 3.5%, and 99.1% for high specificity. C-Diff has 4 first-degree associates that influence the probability of C-Diff development: weight loss, tumor metastases, inflammation/infections, and disease severity. Conclusions Machine-learned BBNs can produce robust estimates of postoperative C-Diff infection, allowing clinicians to identify high-risk patients and potentially implement measures to reduce its incidence or morbidity. PMID:23611947

Early Detection of Ureteropelvic Junction Obstruction Using Signal Analysis and Machine Learning: A Dynamic Solution to a Dynamic Problem.

PubMed

Blum, Emily S; Porras, Antonio R; Biggs, Elijah; Tabrizi, Pooneh R; Sussman, Rachael D; Sprague, Bruce M; Shalaby-Rana, Eglal; Majd, Massoud; Pohl, Hans G; Linguraru, Marius George

2017-10-21

We sought to define features that describe the dynamic information in diuresis renograms for the early detection of clinically significant hydronephrosis caused by ureteropelvic junction obstruction. We studied the diuresis renogram of 55 patients with a mean ± SD age of 75 ± 66 days who had congenital hydronephrosis at initial presentation. Five patients had bilaterally affected kidneys for a total of 60 diuresis renograms. Surgery was performed on 35 kidneys. We extracted 45 features based on curve shape and wavelet analysis from the drainage curves recorded after furosemide administration. The optimal features were selected as the combination that maximized the ROC AUC obtained from a linear support vector machine classifier trained to classify patients as with or without obstruction. Using these optimal features we performed leave 1 out cross validation to estimate the accuracy, sensitivity and specificity of our framework. Results were compared to those obtained using post-diuresis drainage half-time and the percent of clearance after 30 minutes. Our framework had 93% accuracy, including 91% sensitivity and 96% specificity, to predict surgical cases. This was a significant improvement over the same accuracy of 82%, including 71% sensitivity and 96% specificity obtained from half-time and 30-minute clearance using the optimal thresholds of 24.57 minutes and 55.77%, respectively. Our machine learning framework significantly improved the diagnostic accuracy of clinically significant hydronephrosis compared to half-time and 30-minute clearance. This aids in the clinical decision making process by offering a tool for earlier detection of severe cases and it has the potential to reduce the number of diuresis renograms required for diagnosis. Copyright © 2018 American Urological Association Education and Research, Inc. Published by Elsevier Inc. All rights reserved.

Are minimally invasive procedures harder to acquire than conventional surgical procedures?

PubMed

Hiemstra, Ellen; Kolkman, Wendela; le Cessie, Saskia; Jansen, Frank Willem

2011-01-01

It is frequently suggested that minimally invasive surgery (MIS) is harder to acquire than conventional surgery. To test this hypothesis, residents' learning curves of both surgical skills are compared. Residents had to be assessed using a general global rating scale of the OSATS (Objective Structured Assessment of Technical Skills) for every procedure they performed as primary surgeon during a 3-month clinical rotation in gynecological surgery. Nine postgraduate-year-4 residents collected a total of 319 OSATS during the 2 years and 3 months investigation period. These assessments concerned 129 MIS (laparoscopic and hysteroscopic) and 190 conventional (open abdominal and vaginal) procedures. Learning curves (in this study defined as OSATS score plotted against procedure-specific caseload) for MIS and conventional surgery were compared using a linear mixed model. The MIS curve revealed to be steeper than the conventional curve (1.77 vs. 0.75 OSATS points per assessed procedure; 95% CI 1.19-2.35 vs. 0.15-1.35, p < 0.01). Basic MIS procedures do not seem harder to acquire during residency than conventional surgical procedures. This may have resulted from the incorporation of structured MIS training programs in residency. Hopefully, this will lead to a more successful implementation of the advanced MIS procedures. Copyright © 2010 S. Karger AG, Basel.

Deep Processing Strategies and Critical Thinking: Developmental Trajectories Using Latent Growth Analyses

ERIC Educational Resources Information Center

Phan, Huy P.

2011-01-01

The author explored the developmental courses of deep learning approach and critical thinking over a 2-year period. Latent growth curve modeling (LGM) procedures were used to test and trace the trajectories of both theoretical frameworks over time. Participants were 264 (119 women, 145 men) university undergraduates. The Deep Learning subscale of…

A Study of Research and Development Contract Requirements and Their Growth.

DTIC Science & Technology

1979-05-01

Bruner , Jerome S.; Jacqueline J. Goodnow; and George A. Austin. A Study of Thinking. New York, N.Y.: John Wiley and Sons, Inc., 1956. 5. Busek...Interest Continued ... 34 A Learning Analogy ............................ 42 CHAPTER THREE - THE CONCEPTUAL MODEL .............. 49 System...Definitions in One Project ......... ............. Table Eight - Requirement Counting Learning Curve Results ....................... 100 Table Nine - Selected

Patterns and Effectiveness of Mobile Device Usage by Japanese Undergraduates for L2 Acquisition Purposes

ERIC Educational Resources Information Center

Pagel, James W.; Lambacher, Stephen G.

2014-01-01

Mobile technologies, such as smartphones and tablets, are rapidly gaining popularity as an effective means to enhance foreign language learning. However, does the incorporation of these mobile devices really benefit the learner or simply satisfy the instructor's need to be innovative and ahead of the learning curve? The present study seeks to…

Learning Alternatives and Strategies for Students Who Are Struggling

ERIC Educational Resources Information Center

Johnston, Don C.

2008-01-01

Much of what happens in the learning process focuses on teaching to the average student, but the bell curve has deflated with fewer students in the middle, making the educational dilemma more about "how to connect" with students regardless of diverse abilities and various backgrounds. According to the "Twenty-Fourth Annual Report to Congress" by…

Improving Critical Thinking Skills Using Learning Model Logan Avenue Problem Solving (LAPS)-Heuristic

ERIC Educational Resources Information Center

Anggrianto, Desi; Churiyah, Madziatul; Arief, Mohammad

2016-01-01

This research was conducted in order to know the effect of Logan Avenue Problem Solving (LAPS)-Heuristic learning model towards critical thinking skills of students of class X Office Administration (APK) in SMK Negeri 1 Ngawi, East Java, Indonesia on material curve and equilibrium of demand and supply, subject Introduction to Economics and…

Space Human Factors: Research to Application

NASA Technical Reports Server (NTRS)

Woolford, Barbara

2008-01-01

Human Factors has been instrumental in preventing potential on-orbit hazards and increasing overall crew safety. Poor performance & operational learning curves on-orbit are mitigated. Human-centered design is applied to optimize design and minimize potentially hazardous conditions, especially with larger crew sizes and habitat constraints. Lunar and Mars requirements and design developments are enhanced, based on ISS Lessons Learned.

Development Process and Technical Aspects of Laparoscopic Hepatectomy: Learning Curve Based on 15 Years of Experience.

PubMed

Komatsu, Shohei; Scatton, Olivier; Goumard, Claire; Sepulveda, Ailton; Brustia, Raffaele; Perdigao, Fabiano; Soubrane, Olivier

2017-05-01

Laparoscopic hepatectomy continues to be a challenging operation associated with a steep learning curve. This study aimed to evaluate the learning process during 15 years of experience with laparoscopic hepatectomy and to identify approaches to standardization of this procedure. Prospectively collected data of 317 consecutive laparoscopic hepatectomies performed from January 2000 to December 2014 were reviewed retrospectively. The operative procedures were classified into 4 categories (minor hepatectomy, left lateral sectionectomy [LLS], left hepatectomy, and right hepatectomy), and indications were classified into 5 categories (benign-borderline tumor, living donor, metastatic liver tumor, biliary malignancy, and hepatocellular carcinoma). During the first 10 years, the procedures were limited mainly to minor hepatectomy and LLS, and the indications were limited to benign-borderline tumor and living donor. Implementation of major hepatectomy rapidly increased the proportion of malignant tumors, especially hepatocellular carcinoma, starting from 2011. Conversion rates decreased with experience for LLS (13.3% vs 3.4%; p = 0.054) and left hepatectomy (50.0% vs 15.0%; p = 0.012), but not for right hepatectomy (41.4% vs 35.7%; p = 0.661). Our 15-year experience clearly demonstrates the stepwise procedural evolution from LLS through left hepatectomy to right hepatectomy, as well as the trend in indications from benign-borderline tumor/living donor to malignant tumors. In contrast to LLS and left hepatectomy, a learning curve was not observed for right hepatectomy. The ongoing development process can contribute to faster standardization necessary for future advances in laparoscopic hepatectomy. Copyright © 2017 American College of Surgeons. Published by Elsevier Inc. All rights reserved.

It Pays to Go Off-Track: Practicing with Error-Augmenting Haptic Feedback Facilitates Learning of a Curve-Tracing Task

PubMed Central

Williams, Camille K.; Tremblay, Luc; Carnahan, Heather

2016-01-01

Researchers in the domain of haptic training are now entering the long-standing debate regarding whether or not it is best to learn a skill by experiencing errors. Haptic training paradigms provide fertile ground for exploring how various theories about feedback, errors and physical guidance intersect during motor learning. Our objective was to determine how error minimizing, error augmenting and no haptic feedback while learning a self-paced curve-tracing task impact performance on delayed (1 day) retention and transfer tests, which indicate learning. We assessed performance using movement time and tracing error to calculate a measure of overall performance – the speed accuracy cost function. Our results showed that despite exhibiting the worst performance during skill acquisition, the error augmentation group had significantly better accuracy (but not overall performance) than the error minimization group on delayed retention and transfer tests. The control group’s performance fell between that of the two experimental groups but was not significantly different from either on the delayed retention test. We propose that the nature of the task (requiring online feedback to guide performance) coupled with the error augmentation group’s frequent off-target experience and rich experience of error-correction promoted information processing related to error-detection and error-correction that are essential for motor learning. PMID:28082937

Machine learning models in breast cancer survival prediction.

PubMed

Montazeri, Mitra; Montazeri, Mohadeseh; Montazeri, Mahdieh; Beigzadeh, Amin

2016-01-01

Breast cancer is one of the most common cancers with a high mortality rate among women. With the early diagnosis of breast cancer survival will increase from 56% to more than 86%. Therefore, an accurate and reliable system is necessary for the early diagnosis of this cancer. The proposed model is the combination of rules and different machine learning techniques. Machine learning models can help physicians to reduce the number of false decisions. They try to exploit patterns and relationships among a large number of cases and predict the outcome of a disease using historical cases stored in datasets. The objective of this study is to propose a rule-based classification method with machine learning techniques for the prediction of different types of Breast cancer survival. We use a dataset with eight attributes that include the records of 900 patients in which 876 patients (97.3%) and 24 (2.7%) patients were females and males respectively. Naive Bayes (NB), Trees Random Forest (TRF), 1-Nearest Neighbor (1NN), AdaBoost (AD), Support Vector Machine (SVM), RBF Network (RBFN), and Multilayer Perceptron (MLP) machine learning techniques with 10-cross fold technique were used with the proposed model for the prediction of breast cancer survival. The performance of machine learning techniques were evaluated with accuracy, precision, sensitivity, specificity, and area under ROC curve. Out of 900 patients, 803 patients and 97 patients were alive and dead, respectively. In this study, Trees Random Forest (TRF) technique showed better results in comparison to other techniques (NB, 1NN, AD, SVM and RBFN, MLP). The accuracy, sensitivity and the area under ROC curve of TRF are 96%, 96%, 93%, respectively. However, 1NN machine learning technique provided poor performance (accuracy 91%, sensitivity 91% and area under ROC curve 78%). This study demonstrates that Trees Random Forest model (TRF) which is a rule-based classification model was the best model with the highest level of accuracy. Therefore, this model is recommended as a useful tool for breast cancer survival prediction as well as medical decision making.

Identifying HIV associated neurocognitive disorder using large-scale Granger causality analysis on resting-state functional MRI

NASA Astrophysics Data System (ADS)

DSouza, Adora M.; Abidin, Anas Z.; Leistritz, Lutz; Wismüller, Axel

2017-02-01

We investigate the applicability of large-scale Granger Causality (lsGC) for extracting a measure of multivariate information flow between pairs of regional brain activities from resting-state functional MRI (fMRI) and test the effectiveness of these measures for predicting a disease state. Such pairwise multivariate measures of interaction provide high-dimensional representations of connectivity profiles for each subject and are used in a machine learning task to distinguish between healthy controls and individuals presenting with symptoms of HIV Associated Neurocognitive Disorder (HAND). Cognitive impairment in several domains can occur as a result of HIV infection of the central nervous system. The current paradigm for assessing such impairment is through neuropsychological testing. With fMRI data analysis, we aim at non-invasively capturing differences in brain connectivity patterns between healthy subjects and subjects presenting with symptoms of HAND. To classify the extracted interaction patterns among brain regions, we use a prototype-based learning algorithm called Generalized Matrix Learning Vector Quantization (GMLVQ). Our approach to characterize connectivity using lsGC followed by GMLVQ for subsequent classification yields good prediction results with an accuracy of 87% and an area under the ROC curve (AUC) of up to 0.90. We obtain a statistically significant improvement (p<0.01) over a conventional Granger causality approach (accuracy = 0.76, AUC = 0.74). High accuracy and AUC values using our multivariate method to connectivity analysis suggests that our approach is able to better capture changes in interaction patterns between different brain regions when compared to conventional Granger causality analysis known from the literature.

Determining the influence of past development experience on the cost of strategic ballistic missile development

NASA Astrophysics Data System (ADS)

Whitney, Dwight E.

The influence of learning in the form of past relevant experience was examined in data collected for strategic ballistic missiles developed by the United States. A total of twenty-four new missiles were developed and entered service between 1954 and 1990. Missile development costs were collected and analyzed by regression analysis using the learning curve model with factors for past experience and other relevant cost estimating relationships. The purpose of the study was to determine whether prior development experience was a factor in the development cost of these like systems. Of the twenty-four missiles in the population, development costs for twelve of the missiles were collected from the literature. Since the costs were found to be segmented by military service, a discrete input variable for military service was used as one of the cost estimating relationships. Because there were only two US Navy samples, too few to analyze for segmentation and learning rate, they were excluded from the final analysis. The final analysis was on a sample of ten out of eighteen US Army and US Air Force missiles within the population. The result of the analysis found past experience to be a statistically significant factor in describing the development cost of the US Army and US Air Force missiles. The influence equated to a 0.86 progress ratio, indicating prior development experience had a positive (cost-reducing) influence on their development cost. Based on the result, it was concluded that prior development experience was a factor in the development cost of these systems.

Classification of breast cancer in ultrasound imaging using a generic deep learning analysis software: a pilot study.

PubMed

Becker, Anton S; Mueller, Michael; Stoffel, Elina; Marcon, Magda; Ghafoor, Soleen; Boss, Andreas

2018-02-01

To train a generic deep learning software (DLS) to classify breast cancer on ultrasound images and to compare its performance to human readers with variable breast imaging experience. In this retrospective study, all breast ultrasound examinations from January 1, 2014 to December 31, 2014 at our institution were reviewed. Patients with post-surgical scars, initially indeterminate, or malignant lesions with histological diagnoses or 2-year follow-up were included. The DLS was trained with 70% of the images, and the remaining 30% were used to validate the performance. Three readers with variable expertise also evaluated the validation set (radiologist, resident, medical student). Diagnostic accuracy was assessed with a receiver operating characteristic analysis. 82 patients with malignant and 550 with benign lesions were included. Time needed for training was 7 min (DLS). Evaluation time for the test data set were 3.7 s (DLS) and 28, 22 and 25 min for human readers (decreasing experience). Receiver operating characteristic analysis revealed non-significant differences (p-values 0.45-0.47) in the area under the curve of 0.84 (DLS), 0.88 (experienced and intermediate readers) and 0.79 (inexperienced reader). DLS may aid diagnosing cancer on breast ultrasound images with an accuracy comparable to radiologists, and learns better and faster than a human reader with no prior experience. Further clinical trials with dedicated algorithms are warranted. Advances in knowledge: DLS can be trained classify cancer on breast ultrasound images high accuracy even with comparably few training cases. The fast evaluation speed makes real-time image analysis feasible.

Is there a learning curve for the TVT-O procedure? A prospective single-surgeon study of 372 consecutive cases.

PubMed

Serati, Maurizio; Bogani, Giorgio; Braga, Andrea; Sorice, Paola; Salvatore, Stefano; Uccella, Stefano; Ghezzi, Fabio

2015-03-01

To evaluate for the first time in the literature the learning curve of Inside-out transobturator tape (TVT-O™). A prospective observational study was conducted in a tertiary reference center. Consecutive women treated by TVT-O™ performed by one surgeon were included. Data regarding subjective, objective cure rates, and adverse events were collected. Trends, over the number of procedures, were estimated using assay analyses. Number of procedures and variables were interpolating in standard curves using linear lines. Three hundred and seventy two procedures were included. Postoperative pain levels decreased with the increase in the level of expertise (pain levels: 1-day: from 6.6 (±3.3) to 4.3 (±3.1); 95%CI: -0.01603 to 0.001235, p=0.04; 2-day: from 5.6 (±4.1) to 3.6 (±3.7); 95%CI: -0.02092 to -0.002497, p=0.01; 12-month: from 0.1 (±0.7) to 0 (±0); 95%CI: -0.001814 to 0.05019, p=0.07). Overall, objective cure rate was achieved in 93.5% of patients. Additionally, 88.2% and 88.7% patients reported "much better" feeling at PGI-I scale and 80% reduction in UDI score, respectively. We observed, that delta ICIQ-sf (from 12 (±8.7) to 14 (±6.0); p=0.04) and delta-UDI (from 91% to 97%; p=0.04) improved over the time. TVT-O procedure offers excellent outcomes with high objective and subjective cure rates and low complications rate, even at the beginning of the surgeon's learning curve. However, a high experience of the surgeon could significantly improve the subjective cure rate and could reduce postoperative the groin pain. Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.

Learning the ideal observer for SKE detection tasks by use of convolutional neural networks (Cum Laude Poster Award)

NASA Astrophysics Data System (ADS)

Zhou, Weimin; Anastasio, Mark A.

2018-03-01

It has been advocated that task-based measures of image quality (IQ) should be employed to evaluate and optimize imaging systems. Task-based measures of IQ quantify the performance of an observer on a medically relevant task. The Bayesian Ideal Observer (IO), which employs complete statistical information of the object and noise, achieves the upper limit of the performance for a binary signal classification task. However, computing the IO performance is generally analytically intractable and can be computationally burdensome when Markov-chain Monte Carlo (MCMC) techniques are employed. In this paper, supervised learning with convolutional neural networks (CNNs) is employed to approximate the IO test statistics for a signal-known-exactly and background-known-exactly (SKE/BKE) binary detection task. The receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC) are compared to those produced by the analytically computed IO. The advantages of the proposed supervised learning approach for approximating the IO are demonstrated.

State - Level Regulation's Effectiveness in Addressing Global Climate Change and Promoting Solar Energy Deployment

NASA Astrophysics Data System (ADS)

Peterman, Carla Joy

Paper 1, Local Solutions to Global Problems: Climate Change Policies and Regulatory Jurisdiction, considers the efficacy of various types of environmental regulations when they are applied locally to pollutants whose damages extend beyond the jurisdiction of the local regulators. Local regulations of a global pollutant may be ineffective if producers and consumers can avoid them by transacting outside the reach of the local regulator. In many cases, this may involve the physical relocation of the economic activity, a problem often referred to as "leakage." This paper highlights another way in which local policies can be circumvented: through the shuffling of who buys from whom. The paper maintains that the problems of reshuffling are exacerbated when the options for compliance with the regulations are more flexible. Numerical analyses is presented demonstrating that several proposed policies to limit greenhouse gas emissions from the California electricity sector may have very little effect on carbon emissions if they are applied only within that state. Paper 1 concludes that although local subsidies for energy efficiency, renewable electricity, and transportation biofuels constitute attempts to pick technology winners, they may be the only mechanisms that local jurisdictions, acting alone, have at their disposal to address climate change. Paper 2, Pass-Through of Solar PV Incentives to Consumers: The Early Years of California's Solar PV Incentives, examines the pass through of incentives to California solar PV system owners. The full post-subsidy price consumers pay for solar power is a key metric of the success of solar PV incentive programs and of overall PV market performance. This study examines the early years of California's most recent wave of distributed solar PV incentives (2000-2008) to determine the pass-through of incentives. Examination of this period is both intellectually and pragmatically important due to the high level of incentives provided and subsequent high cost to ratepayers; policymakers' expectations that price declines accrue to consumers; and market structure characteristics that might contribute to incomplete pass-through. This analysis shows that incentive passthrough in the California residential solar PV programs was incomplete. Consumer prices declined 54 cents for every additional dollar of incentive received. A large share of the incentive is captured by the solar PV contractor or other actors in the solar PV supply chain. The finding of incomplete pass-through is persistent across specifications. The analysis also identifies a lower degree of incentive pass-through for consumers in the highest income zip codes. Whether expectations of incentives' pass-through align with reality is critically important in the beginning years of emerging clean energy technology programs since this can affect the likelihood of future government investments and public support. Given the often-held policy assumption that consumer prices are declining in response to incentives, it is useful for policymakers to understand the circumstances under which such an assumption may not hold. Paper 3, Testing the Boundaries of the Solar Photovoltaic Learning System, tests how the choice of experience curves' geographic and technology assumptions affect solar PV experience curve results. Historically, solar PV experience curves have assumed one experience curve represents both module and non-module learning and that this learning happens at a global scale. These assumptions may be inaccurate for solar PV since the learning system, and technology and geographic boundaries, are likely different between PV modules and non-module components. Using 2004 to 2008 PV system price data from 13 states, and a longer time series of PV price data for California, some evidence is found that cumulative capacity at the state level is a better predictor of non-module costs than U.S. or global capacity. This paper explores, but is unable to significantly determine, how knowledge spillovers from neighboring states can influence a state's non-module costs. Given data limitations, and limitations to the two-factor experience model methodology itself, it is not possible to conclusively determine the correct geographic boundary for the non-module learning system. Throughout the paper ways in which the experience curve model and data can be augmented to achieve a better estimation are discussed. 2.

A Cognitive Machine Learning Algorithm for Cardiac Imaging: A Pilot Study for Differentiating Constrictive Pericarditis from Restrictive Cardiomyopathy

PubMed Central

Sengupta, Partho P.; Huang, Yen-Min; Bansal, Manish; Ashrafi, Ali; Fisher, Matt; Shameer, Khader; Gall, Walt; Dudley, Joel T

2016-01-01

Background Associating a patient’s profile with the memories of prototypical patients built through previous repeat clinical experience is a key process in clinical judgment. We hypothesized that a similar process using a cognitive computing tool would be well suited for learning and recalling multidimensional attributes of speckle tracking echocardiography (STE) data sets derived from patients with known constrictive pericarditis (CP) and restrictive cardiomyopathy (RCM). Methods and Results Clinical and echocardiographic data of 50 patients with CP and 44 with RCM were used for developing an associative memory classifier (AMC) based machine learning algorithm. The STE data was normalized in reference to 47 controls with no structural heart disease, and the diagnostic area under the receiver operating characteristic curve (AUC) of the AMC was evaluated for differentiating CP from RCM. Using only STE variables, AMC achieved a diagnostic AUC of 89·2%, which improved to 96·2% with addition of 4 echocardiographic variables. In comparison, the AUC of early diastolic mitral annular velocity and left ventricular longitudinal strain were 82.1% and 63·7%, respectively. Furthermore, AMC demonstrated greater accuracy and shorter learning curves than other machine learning approaches with accuracy asymptotically approaching 90% after a training fraction of 0·3 and remaining flat at higher training fractions. Conclusions This study demonstrates feasibility of a cognitive machine learning approach for learning and recalling patterns observed during echocardiographic evaluations. Incorporation of machine learning algorithms in cardiac imaging may aid standardized assessments and support the quality of interpretations, particularly for novice readers with limited experience. PMID:27266599

Computer-based learning of spelling skills in children with and without dyslexia.

PubMed

Kast, Monika; Baschera, Gian-Marco; Gross, Markus; Jäncke, Lutz; Meyer, Martin

2011-12-01

Our spelling training software recodes words into multisensory representations comprising visual and auditory codes. These codes represent information about letters and syllables of a word. An enhanced version, developed for this study, contains an additional phonological code and an improved word selection controller relying on a phoneme-based student model. We investigated the spelling behavior of children by means of learning curves based on log-file data of the previous and the enhanced software version. First, we compared the learning progress of children with dyslexia working either with the previous software (n = 28) or the adapted version (n = 37). Second, we investigated the spelling behavior of children with dyslexia (n = 37) and matched children without dyslexia (n = 25). To gain deeper insight into which factors are relevant for acquiring spelling skills, we analyzed the influence of cognitive abilities, such as attention functions and verbal memory skills, on the learning behavior. All investigations of the learning process are based on learning curve analyses of the collected log-file data. The results evidenced that those children with dyslexia benefit significantly from the additional phonological cue and the corresponding phoneme-based student model. Actually, children with dyslexia improve their spelling skills to the same extent as children without dyslexia and were able to memorize phoneme to grapheme correspondence when given the correct support and adequate training. In addition, children with low attention functions benefit from the structured learning environment. Generally, our data showed that memory sources are supportive cognitive functions for acquiring spelling skills and for using the information cues of a multi-modal learning environment.

How to start a minimal access mitral valve program.

PubMed

Hunter, Steven

2013-11-01

The seven pillars of governance established by the National Health Service in the United Kingdom provide a useful framework for the process of introducing new procedures to a hospital. Drawing from local experience, the author present guidance for institutions considering establishing a minimal access mitral valve program. The seven pillars of governance apply to the practice of minimally invasive mitral valve surgery, based on the principle of patient-centred practice. The author delineate the benefits of minimally invasive mitral valve surgery in terms of: "clinical effectiveness", including reduced length of hospital stay, "risk management effectiveness", including conversion to sternotomy and aortic dissection, "patient experience" including improved cosmesis and quicker recovery, and the effectiveness of communication, resources and strategies in the implementation of minimally invasive mitral valve surgery. Finally, the author have identified seven learning curves experienced by surgeons involved in introducing a minimal access mitral valve program. The learning curves are defined as: techniques of mitral valve repair, Transoesophageal Echocardiography-guided cannulation, incisions, instruments, visualization, aortic occlusion and cardiopulmonary bypass strategies. From local experience, the author provide advice on how to reduce the learning curves, such as practising with the specialised instruments and visualization techniques during sternotomy cases. Underpinning the NHS pillars are the principles of systems awareness, teamwork, communication, ownership and leadership, all of which are paramount to performing any surgery but more so with minimal access surgery, as will be highlighted throughout this paper.

Anesthesiologists' learning curves for bedside qualitative ultrasound assessment of gastric content: a cohort study.

PubMed

Arzola, Cristian; Carvalho, Jose C A; Cubillos, Javier; Ye, Xiang Y; Perlas, Anahi

2013-08-01

Focused assessment of the gastric antrum by ultrasound is a feasible tool to evaluate the quality of the stomach content. We aimed to determine the amount of training an anesthesiologist would need to achieve competence in the bedside ultrasound technique for qualitative assessment of gastric content. Six anesthesiologists underwent a teaching intervention followed by a formative assessment; then learning curves were constructed. Participants received didactic teaching (reading material, picture library, and lecture) and an interactive hands-on workshop on live models directed by an expert sonographer. The participants were instructed on how to perform a systematic qualitative assessment to diagnose one of three distinct categories of gastric content (empty, clear fluid, solid) in healthy volunteers. Individual learning curves were constructed using the cumulative sum method, and competence was defined as a 90% success rate in a series of ultrasound examinations. A predictive model was further developed based on the entire cohort performance to determine the number of cases required to achieve a 95% success rate. Each anesthesiologist performed 30 ultrasound examinations (a total of 180 assessments), and three of the six participants achieved competence. The average number of cases required to achieve 90% and 95% success rates was estimated to be 24 and 33, respectively. With appropriate training and supervision, it is estimated that anesthesiologists will achieve a 95% success rate in bedside qualitative ultrasound assessment after performing approximately 33 examinations.

A theory for how sensorimotor skills are learned and retained in noisy and nonstationary neural circuits

PubMed Central

Ajemian, Robert; D’Ausilio, Alessandro; Moorman, Helene; Bizzi, Emilio

2013-01-01

During the process of skill learning, synaptic connections in our brains are modified to form motor memories of learned sensorimotor acts. The more plastic the adult brain is, the easier it is to learn new skills or adapt to neurological injury. However, if the brain is too plastic and the pattern of synaptic connectivity is constantly changing, new memories will overwrite old memories, and learning becomes unstable. This trade-off is known as the stability–plasticity dilemma. Here a theory of sensorimotor learning and memory is developed whereby synaptic strengths are perpetually fluctuating without causing instability in motor memory recall, as long as the underlying neural networks are sufficiently noisy and massively redundant. The theory implies two distinct stages of learning—preasymptotic and postasymptotic—because once the error drops to a level comparable to that of the noise-induced error, further error reduction requires altered network dynamics. A key behavioral prediction derived from this analysis is tested in a visuomotor adaptation experiment, and the resultant learning curves are modeled with a nonstationary neural network. Next, the theory is used to model two-photon microscopy data that show, in animals, high rates of dendritic spine turnover, even in the absence of overt behavioral learning. Finally, the theory predicts enhanced task selectivity in the responses of individual motor cortical neurons as the level of task expertise increases. From these considerations, a unique interpretation of sensorimotor memory is proposed—memories are defined not by fixed patterns of synaptic weights but, rather, by nonstationary synaptic patterns that fluctuate coherently. PMID:24324147

A polymerase chain reaction-coupled high-resolution melting curve analytical approach for the monitoring of monospecificity of avian Eimeria species.

PubMed

Kirkpatrick, Naomi C; Blacker, Hayley P; Woods, Wayne G; Gasser, Robin B; Noormohammadi, Amir H

2009-02-01

Coccidiosis is a significant disease of poultry caused by different species of Eimeria. Differentiation of Eimeria species is important for the quality control of the live attenuated Eimeria vaccines derived from monospecific lines of Eimeria spp. In this study, high-resolution melting (HRM) curve analysis of the amplicons generated from the second internal transcribed spacer of nuclear ribosomal DNA (ITS-2) was used to distinguish between seven pathogenic Eimeria species of chickens, and the results were compared with those obtained from the previously described technique, capillary electrophoresis. Using a series of known monospecific lines of Eimeria species, HRM curve analysis was shown to distinguish between Eimeria acervulina, Eimeria brunetti, Eimeria maxima, Eimeria mitis, Eimeria necatrix, Eimeria praecox and Eimeria tenella. Computerized analysis of the HRM curves and capillary electrophoresis profiles could detect the dominant species in several specimens containing different ratios of E. necatrix and E. maxima and of E. tenella and E. acervulina. The HRM curve analysis identified all of the mixtures as "variation" to the reference species, and also identified the minor species in some mixtures. Computerized HRM curve analysis also detected impurities in 21 possible different combinations of the seven Eimeria species. The PCR-based HRM curve analysis of the ITS-2 provides a powerful tool for the detection and identification of pure Eimeria species. The HRM curve analysis could also be used as a rapid tool in the quality assurance of Eimeria vaccine production to confirm the purity of the monospecific cell lines. The HRM curve analysis is rapid and reliable and can be performed in a single test tube in less than 3 h.

Learning Curves: Making Quality Online Health Information Available at a Fitness Center.

PubMed

Dobbins, Montie T; Tarver, Talicia; Adams, Mararia; Jones, Dixie A

2012-01-01

Meeting consumer health information needs can be a challenge. Research suggests that women seek health information from a variety of resources, including the Internet. In an effort to make women aware of reliable health information sources, the Louisiana State University Health Sciences Center - Shreveport Medical Library engaged in a partnership with a franchise location of Curves International, Inc. This article will discuss the project, its goals and its challenges.

RS/1 in the Clinical Environment

PubMed Central

Kush, Thomas

1980-01-01

This paper describes the design of RS/1,™ the Research System, and its use in clinical patient studies. RS/1 is an interactive computer software system developed by the Medical Systems Group at BBN. Investigators and technicians who have never before used computers can learn RS/1 with a few hours of training. It uses familiar and intuitive concepts for data handling and data analysis, such as the “automated notebook” format of data storage, the direct use of graphs in curve-fitting, and a simple command language. Its versatility has made RS/1 useful in clinical research contexts, especially for studies involving patient care data.

Physiological Sensing Now Open to the World: New Resources Are Allowing Us to Learn, Experiment, and Create Imaginative Solutions for Biomedical Applications.

PubMed

da Silva, Hugo Placido

2018-01-01

With the advent of low-cost computing platforms, such as Arduino (http://www.arduino.cc) and Raspberry Pi (http://www.raspberrypi.org), it has become clear that lowering the cost barrier and shortening the learning curve, with the backing of a motivated community, would play a transformational role in the way people learn, experiment, and create imaginative solutions to outstanding problems that can benefit from embedded systems.

Analyzing Student Learning Outcomes: Usefulness of Logistic and Cox Regression Models. IR Applications, Volume 5

ERIC Educational Resources Information Center

Chen, Chau-Kuang

2005-01-01

Logistic and Cox regression methods are practical tools used to model the relationships between certain student learning outcomes and their relevant explanatory variables. The logistic regression model fits an S-shaped curve into a binary outcome with data points of zero and one. The Cox regression model allows investigators to study the duration…

Subjective Estimates of Job Performance after Job Preview: Determinants of Anticipated Learning Curves

ERIC Educational Resources Information Center

Ackerman, Phillip L.; Shapiro, Stacey; Beier, Margaret E.

2011-01-01

When people choose a particular occupation, they presumably make an implicit judgment that they will perform well on a job at some point in the future, typically after extensive education and/or on-the-job experience. Research on learning and skill acquisition has pointed to a power law of practice, where large gains in performance come early in…

Decision curve analysis revisited: overall net benefit, relationships to ROC curve analysis, and application to case-control studies.

PubMed

Rousson, Valentin; Zumbrunn, Thomas

2011-06-22

Decision curve analysis has been introduced as a method to evaluate prediction models in terms of their clinical consequences if used for a binary classification of subjects into a group who should and into a group who should not be treated. The key concept for this type of evaluation is the "net benefit", a concept borrowed from utility theory. We recall the foundations of decision curve analysis and discuss some new aspects. First, we stress the formal distinction between the net benefit for the treated and for the untreated and define the concept of the "overall net benefit". Next, we revisit the important distinction between the concept of accuracy, as typically assessed using the Youden index and a receiver operating characteristic (ROC) analysis, and the concept of utility of a prediction model, as assessed using decision curve analysis. Finally, we provide an explicit implementation of decision curve analysis to be applied in the context of case-control studies. We show that the overall net benefit, which combines the net benefit for the treated and the untreated, is a natural alternative to the benefit achieved by a model, being invariant with respect to the coding of the outcome, and conveying a more comprehensive picture of the situation. Further, within the framework of decision curve analysis, we illustrate the important difference between the accuracy and the utility of a model, demonstrating how poor an accurate model may be in terms of its net benefit. Eventually, we expose that the application of decision curve analysis to case-control studies, where an accurate estimate of the true prevalence of a disease cannot be obtained from the data, is achieved with a few modifications to the original calculation procedure. We present several interrelated extensions to decision curve analysis that will both facilitate its interpretation and broaden its potential area of application.

Decision curve analysis revisited: overall net benefit, relationships to ROC curve analysis, and application to case-control studies

PubMed Central

2011-01-01

Background Decision curve analysis has been introduced as a method to evaluate prediction models in terms of their clinical consequences if used for a binary classification of subjects into a group who should and into a group who should not be treated. The key concept for this type of evaluation is the "net benefit", a concept borrowed from utility theory. Methods We recall the foundations of decision curve analysis and discuss some new aspects. First, we stress the formal distinction between the net benefit for the treated and for the untreated and define the concept of the "overall net benefit". Next, we revisit the important distinction between the concept of accuracy, as typically assessed using the Youden index and a receiver operating characteristic (ROC) analysis, and the concept of utility of a prediction model, as assessed using decision curve analysis. Finally, we provide an explicit implementation of decision curve analysis to be applied in the context of case-control studies. Results We show that the overall net benefit, which combines the net benefit for the treated and the untreated, is a natural alternative to the benefit achieved by a model, being invariant with respect to the coding of the outcome, and conveying a more comprehensive picture of the situation. Further, within the framework of decision curve analysis, we illustrate the important difference between the accuracy and the utility of a model, demonstrating how poor an accurate model may be in terms of its net benefit. Eventually, we expose that the application of decision curve analysis to case-control studies, where an accurate estimate of the true prevalence of a disease cannot be obtained from the data, is achieved with a few modifications to the original calculation procedure. Conclusions We present several interrelated extensions to decision curve analysis that will both facilitate its interpretation and broaden its potential area of application. PMID:21696604

Computer-aided diagnosis of prostate cancer using a deep convolutional neural network from multiparametric MRI.

PubMed

Song, Yang; Zhang, Yu-Dong; Yan, Xu; Liu, Hui; Zhou, Minxiong; Hu, Bingwen; Yang, Guang

2018-04-16

Deep learning is the most promising methodology for automatic computer-aided diagnosis of prostate cancer (PCa) with multiparametric MRI (mp-MRI). To develop an automatic approach based on deep convolutional neural network (DCNN) to classify PCa and noncancerous tissues (NC) with mp-MRI. Retrospective. In all, 195 patients with localized PCa were collected from a PROSTATEx database. In total, 159/17/19 patients with 444/48/55 observations (215/23/23 PCas and 229/25/32 NCs) were randomly selected for training/validation/testing, respectively. T 2 -weighted, diffusion-weighted, and apparent diffusion coefficient images. A radiologist manually labeled the regions of interest of PCas and NCs and estimated the Prostate Imaging Reporting and Data System (PI-RADS) scores for each region. Inspired by VGG-Net, we designed a patch-based DCNN model to distinguish between PCa and NCs based on a combination of mp-MRI data. Additionally, an enhanced prediction method was used to improve the prediction accuracy. The performance of DCNN prediction was tested using a receiver operating characteristic (ROC) curve, and the area under the ROC curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Moreover, the predicted result was compared with the PI-RADS score to evaluate its clinical value using decision curve analysis. Two-sided Wilcoxon signed-rank test with statistical significance set at 0.05. The DCNN produced excellent diagnostic performance in distinguishing between PCa and NC for testing datasets with an AUC of 0.944 (95% confidence interval: 0.876-0.994), sensitivity of 87.0%, specificity of 90.6%, PPV of 87.0%, and NPV of 90.6%. The decision curve analysis revealed that the joint model of PI-RADS and DCNN provided additional net benefits compared with the DCNN model and the PI-RADS scheme. The proposed DCNN-based model with enhanced prediction yielded high performance in statistical analysis, suggesting that DCNN could be used in computer-aided diagnosis (CAD) for PCa classification. 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018. © 2018 International Society for Magnetic Resonance in Medicine.

Toward a real-time system for temporal enhanced ultrasound-guided prostate biopsy.

PubMed

Azizi, Shekoofeh; Van Woudenberg, Nathan; Sojoudi, Samira; Li, Ming; Xu, Sheng; Abu Anas, Emran M; Yan, Pingkun; Tahmasebi, Amir; Kwak, Jin Tae; Turkbey, Baris; Choyke, Peter; Pinto, Peter; Wood, Bradford; Mousavi, Parvin; Abolmaesumi, Purang

2018-03-27

We have previously proposed temporal enhanced ultrasound (TeUS) as a new paradigm for tissue characterization. TeUS is based on analyzing a sequence of ultrasound data with deep learning and has been demonstrated to be successful for detection of cancer in ultrasound-guided prostate biopsy. Our aim is to enable the dissemination of this technology to the community for large-scale clinical validation. In this paper, we present a unified software framework demonstrating near-real-time analysis of ultrasound data stream using a deep learning solution. The system integrates ultrasound imaging hardware, visualization and a deep learning back-end to build an accessible, flexible and robust platform. A client-server approach is used in order to run computationally expensive algorithms in parallel. We demonstrate the efficacy of the framework using two applications as case studies. First, we show that prostate cancer detection using near-real-time analysis of RF and B-mode TeUS data and deep learning is feasible. Second, we present real-time segmentation of ultrasound prostate data using an integrated deep learning solution. The system is evaluated for cancer detection accuracy on ultrasound data obtained from a large clinical study with 255 biopsy cores from 157 subjects. It is further assessed with an independent dataset with 21 biopsy targets from six subjects. In the first study, we achieve area under the curve, sensitivity, specificity and accuracy of 0.94, 0.77, 0.94 and 0.92, respectively, for the detection of prostate cancer. In the second study, we achieve an AUC of 0.85. Our results suggest that TeUS-guided biopsy can be potentially effective for the detection of prostate cancer.

Machine learning for prediction of all-cause mortality in patients with suspected coronary artery disease: a 5-year multicentre prospective registry analysis

PubMed Central

Motwani, Manish; Dey, Damini; Berman, Daniel S.; Germano, Guido; Achenbach, Stephan; Al-Mallah, Mouaz H.; Andreini, Daniele; Budoff, Matthew J.; Cademartiri, Filippo; Callister, Tracy Q.; Chang, Hyuk-Jae; Chinnaiyan, Kavitha; Chow, Benjamin J.W.; Cury, Ricardo C.; Delago, Augustin; Gomez, Millie; Gransar, Heidi; Hadamitzky, Martin; Hausleiter, Joerg; Hindoyan, Niree; Feuchtner, Gudrun; Kaufmann, Philipp A.; Kim, Yong-Jin; Leipsic, Jonathon; Lin, Fay Y.; Maffei, Erica; Marques, Hugo; Pontone, Gianluca; Raff, Gilbert; Rubinshtein, Ronen; Shaw, Leslee J.; Stehli, Julia; Villines, Todd C.; Dunning, Allison; Min, James K.; Slomka, Piotr J.

2017-01-01

Aims Traditional prognostic risk assessment in patients undergoing non-invasive imaging is based upon a limited selection of clinical and imaging findings. Machine learning (ML) can consider a greater number and complexity of variables. Therefore, we investigated the feasibility and accuracy of ML to predict 5-year all-cause mortality (ACM) in patients undergoing coronary computed tomographic angiography (CCTA), and compared the performance to existing clinical or CCTA metrics. Methods and results The analysis included 10 030 patients with suspected coronary artery disease and 5-year follow-up from the COronary CT Angiography EvaluatioN For Clinical Outcomes: An InteRnational Multicenter registry. All patients underwent CCTA as their standard of care. Twenty-five clinical and 44 CCTA parameters were evaluated, including segment stenosis score (SSS), segment involvement score (SIS), modified Duke index (DI), number of segments with non-calcified, mixed or calcified plaques, age, sex, gender, standard cardiovascular risk factors, and Framingham risk score (FRS). Machine learning involved automated feature selection by information gain ranking, model building with a boosted ensemble algorithm, and 10-fold stratified cross-validation. Seven hundred and forty-five patients died during 5-year follow-up. Machine learning exhibited a higher area-under-curve compared with the FRS or CCTA severity scores alone (SSS, SIS, DI) for predicting all-cause mortality (ML: 0.79 vs. FRS: 0.61, SSS: 0.64, SIS: 0.64, DI: 0.62; P< 0.001). Conclusions Machine learning combining clinical and CCTA data was found to predict 5-year ACM significantly better than existing clinical or CCTA metrics alone. PMID:27252451

Machine learning for prediction of all-cause mortality in patients with suspected coronary artery disease: a 5-year multicentre prospective registry analysis.

PubMed

Motwani, Manish; Dey, Damini; Berman, Daniel S; Germano, Guido; Achenbach, Stephan; Al-Mallah, Mouaz H; Andreini, Daniele; Budoff, Matthew J; Cademartiri, Filippo; Callister, Tracy Q; Chang, Hyuk-Jae; Chinnaiyan, Kavitha; Chow, Benjamin J W; Cury, Ricardo C; Delago, Augustin; Gomez, Millie; Gransar, Heidi; Hadamitzky, Martin; Hausleiter, Joerg; Hindoyan, Niree; Feuchtner, Gudrun; Kaufmann, Philipp A; Kim, Yong-Jin; Leipsic, Jonathon; Lin, Fay Y; Maffei, Erica; Marques, Hugo; Pontone, Gianluca; Raff, Gilbert; Rubinshtein, Ronen; Shaw, Leslee J; Stehli, Julia; Villines, Todd C; Dunning, Allison; Min, James K; Slomka, Piotr J

2017-02-14

Traditional prognostic risk assessment in patients undergoing non-invasive imaging is based upon a limited selection of clinical and imaging findings. Machine learning (ML) can consider a greater number and complexity of variables. Therefore, we investigated the feasibility and accuracy of ML to predict 5-year all-cause mortality (ACM) in patients undergoing coronary computed tomographic angiography (CCTA), and compared the performance to existing clinical or CCTA metrics. The analysis included 10 030 patients with suspected coronary artery disease and 5-year follow-up from the COronary CT Angiography EvaluatioN For Clinical Outcomes: An InteRnational Multicenter registry. All patients underwent CCTA as their standard of care. Twenty-five clinical and 44 CCTA parameters were evaluated, including segment stenosis score (SSS), segment involvement score (SIS), modified Duke index (DI), number of segments with non-calcified, mixed or calcified plaques, age, sex, gender, standard cardiovascular risk factors, and Framingham risk score (FRS). Machine learning involved automated feature selection by information gain ranking, model building with a boosted ensemble algorithm, and 10-fold stratified cross-validation. Seven hundred and forty-five patients died during 5-year follow-up. Machine learning exhibited a higher area-under-curve compared with the FRS or CCTA severity scores alone (SSS, SIS, DI) for predicting all-cause mortality (ML: 0.79 vs. FRS: 0.61, SSS: 0.64, SIS: 0.64, DI: 0.62; P< 0.001). Machine learning combining clinical and CCTA data was found to predict 5-year ACM significantly better than existing clinical or CCTA metrics alone. Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2016. For permissions please email: journals.permissions@oup.com.

Combining classifiers using their receiver operating characteristics and maximum likelihood estimation.

PubMed

Haker, Steven; Wells, William M; Warfield, Simon K; Talos, Ion-Florin; Bhagwat, Jui G; Goldberg-Zimring, Daniel; Mian, Asim; Ohno-Machado, Lucila; Zou, Kelly H

2005-01-01

In any medical domain, it is common to have more than one test (classifier) to diagnose a disease. In image analysis, for example, there is often more than one reader or more than one algorithm applied to a certain data set. Combining of classifiers is often helpful, but determining the way in which classifiers should be combined is not trivial. Standard strategies are based on learning classifier combination functions from data. We describe a simple strategy to combine results from classifiers that have not been applied to a common data set, and therefore can not undergo this type of joint training. The strategy, which assumes conditional independence of classifiers, is based on the calculation of a combined Receiver Operating Characteristic (ROC) curve, using maximum likelihood analysis to determine a combination rule for each ROC operating point. We offer some insights into the use of ROC analysis in the field of medical imaging.

Combining Classifiers Using Their Receiver Operating Characteristics and Maximum Likelihood Estimation*

PubMed Central

Haker, Steven; Wells, William M.; Warfield, Simon K.; Talos, Ion-Florin; Bhagwat, Jui G.; Goldberg-Zimring, Daniel; Mian, Asim; Ohno-Machado, Lucila; Zou, Kelly H.

2010-01-01

In any medical domain, it is common to have more than one test (classifier) to diagnose a disease. In image analysis, for example, there is often more than one reader or more than one algorithm applied to a certain data set. Combining of classifiers is often helpful, but determining the way in which classifiers should be combined is not trivial. Standard strategies are based on learning classifier combination functions from data. We describe a simple strategy to combine results from classifiers that have not been applied to a common data set, and therefore can not undergo this type of joint training. The strategy, which assumes conditional independence of classifiers, is based on the calculation of a combined Receiver Operating Characteristic (ROC) curve, using maximum likelihood analysis to determine a combination rule for each ROC operating point. We offer some insights into the use of ROC analysis in the field of medical imaging. PMID:16685884

Out-of-Sample Extrapolation utilizing Semi-Supervised Manifold Learning (OSE-SSL): Content Based Image Retrieval for Histopathology Images

PubMed Central

Sparks, Rachel; Madabhushi, Anant

2016-01-01

Content-based image retrieval (CBIR) retrieves database images most similar to the query image by (1) extracting quantitative image descriptors and (2) calculating similarity between database and query image descriptors. Recently, manifold learning (ML) has been used to perform CBIR in a low dimensional representation of the high dimensional image descriptor space to avoid the curse of dimensionality. ML schemes are computationally expensive, requiring an eigenvalue decomposition (EVD) for every new query image to learn its low dimensional representation. We present out-of-sample extrapolation utilizing semi-supervised ML (OSE-SSL) to learn the low dimensional representation without recomputing the EVD for each query image. OSE-SSL incorporates semantic information, partial class label, into a ML scheme such that the low dimensional representation co-localizes semantically similar images. In the context of prostate histopathology, gland morphology is an integral component of the Gleason score which enables discrimination between prostate cancer aggressiveness. Images are represented by shape features extracted from the prostate gland. CBIR with OSE-SSL for prostate histology obtained from 58 patient studies, yielded an area under the precision recall curve (AUPRC) of 0.53 ± 0.03 comparatively a CBIR with Principal Component Analysis (PCA) to learn a low dimensional space yielded an AUPRC of 0.44 ± 0.01. PMID:27264985

What is the learning curve for the anterior approach for total hip arthroplasty?

PubMed

de Steiger, Richard Noel; Lorimer, Michelle; Solomon, Michael

2015-12-01

There are many factors that may affect the learning curve for total hip arthroplasty (THA) and surgical approach is one of these. There has been renewed interest in the direct anterior approach for THA with variable outcomes reported, but few studies have documented a surgeon's individual learning curve when using this approach. (1) What was the revision rate for all surgeons adopting the anterior approach for placement of a particular implant? (2) What was the revision rate for surgeons who performed > 100 cases in this fashion? (3) Is there a minimum number of cases required to complete a learning curve for this procedure? The Australian Orthopaedic Association National Joint Replacement Registry prospectively collects data on all primary and revision joint arthroplasty surgery. We analyzed all conventional THAs performed up to December 31, 2013, with a primary diagnosis of osteoarthritis using a specific implant combination and secondarily those associated with surgeons performing more than 100 procedures. Ninety-five percent of these procedures were performed through the direct anterior approach. Procedures using this combination were ordered from earliest (first procedure date) to latest (last procedure date) for each individual surgeon. Using the order number for each surgeon, five operation groups were defined: one to 15 operations, 16 to 30 operations, 31 to 50 operations, 51 to 100 operations, and > 100 operations. The primary outcome measure was time to first revision using Kaplan-Meier estimates of survivorship. Sixty-eight surgeons performed 5499 THAs using the specified implant combination. The cumulative percent revision at 4 years for all 68 surgeons was 3% (95% confidence interval [CI], 2.5-3.8). For surgeons who had performed over 100 operations, the cumulative revision rate was 3% (95% CI, 2.0-3.5). It was not until surgeons had performed over 50 operations that there was no difference in the cumulative percent revision compared with over 100 operations. The cumulative percent revision for surgeons performing 51 to 100 operations at 4 years was 3% (95% CI, 1.5-5.4) and over 100 operations 2% (95% CI, 1.2-2.7; hazard ratio, 1.40 [95% CI, 0.7-2.7]; p = 0.33). There is a learning curve for the anterior approach for THA even when using a prosthesis combination specifically marketed for that approach. We found that 50 or more procedures need to be performed by a surgeon before the rate of revision is no different from performing 100 or more procedures. Surgeons should be aware of this initial higher rate of revision when deciding which approach delivers the best outcome for their patients.

Learning curves in abdominal wall reconstruction with components separation: one step closer toward improving outcomes and reducing complications.

PubMed

Hultman, Charles Scott; Clayton, John L; Kittinger, Benjamin J; Tong, Winnie M

2014-01-01

Learning curves are characterized by incremental improvement of a process, through repetition and reduction in variability, but can be disrupted with the emergence of new techniques and technologies. Abdominal wall reconstruction continues to evolve, with the introduction of components separation in the 1990s and biologic mesh in the 2000s. As such, attempts at innovation may impact the success of reconstructive outcomes and yield a changing set of complications. The purpose of this project was to describe the paradigm shift that has occurred in abdominal wall reconstruction during the past 10 years, focusing on the incorporation of new materials and methods. We reviewed 150 consecutive patients who underwent abdominal wall reconstruction of midline defects with components separation, from 2000 to 2010. Both univariate and multivariate logistic regression analyses were performed to identify risk factors for complications. Patients were stratified into the following periods: early (2000-2003), middle (2004-2006), and late (2007-2010). From 2000 to 2010, we performed 150 abdominal wall reconstructions with components separation [mean age, 50.2 years; body mass index (BMI), 30.4; size of defect, 357 cm; length of stay, 9.6 days; follow-up, 4.4 years]. Primary fascial closure was performed in 120 patients. Mesh was used in 114 patients in the following locations: overlay (n = 28), inlay (n = 30), underlay (n = 54), and unknown (n = 2). Complications occurred in a bimodal distribution, highest in 2001 (introduction of biologic mesh) and 2008 (conversion from underlay to overlay location). Age, sex, history of smoking, defect size, and length of stay were not associated with incidence of complications. Unadjusted risk factors for seroma (16.8%) were elevated BMI, of previous hernia repairs, use of overlay mesh, and late portion of the learning curve, with logistic regression supporting only late portion of the learning curve [odds ratio (OR), 4.3; 95% confidence interval (CI), 1.0-18.6] and BMI (OR, 1.17; 95% CI, 1.06-1.29). The only unadjusted risk factor for recurrence was location of mesh. Logistic regression, comparing underlay, inlay, and overlay mesh to no mesh, revealed that the use of underlay mesh predicted recurrence (OR, 3.0; 95% CI, 1.04-8.64). All P values were less than 0.05. The overall learning curve for a specific procedure, such as abdominal wall reconstruction, can be quite volatile, especially as innovative techniques and new technologies are introduced and incorporated into the surgeon's practice. Our current practice includes primary repair myofascial flap of the components separation and the use of biologic mesh as an overlay graft, anchored to the external oblique. This process of outcome improvement is not gradual but is often punctuated by periods of failure and redemption.

Hoarding behaviors in children with learning disabilities.

PubMed

Testa, Renée; Pantelis, Christos; Fontenelle, Leonardo F

2011-05-01

Our objective was to describe the prevalence, comorbidity, and neuropsychological profiles of children with hoarding and learning disabilities. From 61 children with learning disabilities, 16.4% exhibited hoarding as a major clinical issue. Although children with learning disabilities and hoarding displayed greater rates of obsessive-compulsive disorder (30%) as compared to those with learning disabilities without hoarding (5.9%), the majority of patients belonging to the former group did not display obsessive-compulsive disorder diagnosis. When learning disability patients with hoarding were compared to age-, sex-, and IQ-matched learning disability subjects without hoarding, hoarders exhibited a slower learning curve on word list-learning task. In conclusion, salient hoarding behaviors were found to be relatively common in a sample of children with learning disabilities and not necessarily associated with obsessive-compulsive disorder, supporting its nosological independence. It is unclear whether underlying cognitive features may play a major role in the development of hoarding behaviors in children with learning disabilities.

Using MetaboAnalyst 3.0 for Comprehensive Metabolomics Data Analysis.

PubMed

Xia, Jianguo; Wishart, David S

2016-09-07

MetaboAnalyst (http://www.metaboanalyst.ca) is a comprehensive Web application for metabolomic data analysis and interpretation. MetaboAnalyst handles most of the common metabolomic data types from most kinds of metabolomics platforms (MS and NMR) for most kinds of metabolomics experiments (targeted, untargeted, quantitative). In addition to providing a variety of data processing and normalization procedures, MetaboAnalyst also supports a number of data analysis and data visualization tasks using a range of univariate, multivariate methods such as PCA (principal component analysis), PLS-DA (partial least squares discriminant analysis), heatmap clustering and machine learning methods. MetaboAnalyst also offers a variety of tools for metabolomic data interpretation including MSEA (metabolite set enrichment analysis), MetPA (metabolite pathway analysis), and biomarker selection via ROC (receiver operating characteristic) curve analysis, as well as time series and power analysis. This unit provides an overview of the main functional modules and the general workflow of the latest version of MetaboAnalyst (MetaboAnalyst 3.0), followed by eight detailed protocols. © 2016 by John Wiley & Sons, Inc. Copyright © 2016 John Wiley & Sons, Inc.

Methods of Technological Forecasting,

DTIC Science & Technology

1977-05-01

Trend Extrapolation Progress Curve Analogy Trend Correlation Substitution Analysis or Substitution Growth Curves Envelope Curve Advances in the State of...the Art Technological Mapping Contextual Mapping Matrix Input-Output Analysis Mathematical Models Simulation Models Dynamic Modelling. CHAPTER IV...Generation Interaction between Needs and Possibilities Map of the Technological Future — (‘ross- Impact Matri x Discovery Matrix Morphological Analysis

Impact of a learning curve on the survivorship of 4802 cementless total hip arthroplasties.

PubMed

Magill, P; Blaney, J; Hill, J C; Bonnin, M P; Beverland, D E

2016-12-01

Our aim was to report survivorship data and lessons learned with the Corail/Pinnacle cementless total hip arthroplasty (THA) system. Between August 2005 and March 2015, a total of 4802 primary cementless Corail/Pinnacle THAs were performed in 4309 patients. In March 2016, we reviewed these hips from a prospectively maintained database. A total of 80 hips (1.67%) have been revised which is equivalent to a cumulative risk of revision of 2.5% at ten years. The rate of revision was not significantly higher in patients aged ≥ 70 years (p = 0.93). The leading indications for revision were instability (n = 22, 0.46%), infection (n = 20, 0.42%), aseptic femoral loosening (n = 15, 0.31%) and femoral fracture (n = 6, 0.12%). There were changes in the surgical technique with respect to the Corail femoral component during the ten-year period involving a change to collared components and a trend towards larger size. These resulted in a decrease in the rate of iatrogenic femoral fracture and a decrease in the rate of aseptic loosening. The rate of revision in this series is comparable with the best performing THAs in registry data. Most revisions were not directly related to the implants. Despite extensive previous experience with cemented femoral components, the senior author noted a learning curve requiring increased focus on primary stability. The number of revisions related to the femoral component is reducing. Any new technology has a learning curve that may be independent of surgical experience. Cite this article: Bone Joint J 2016;98-B:1589-96. ©2016 The British Editorial Society of Bone & Joint Surgery.

Trainee competence in thoracoscopic esophagectomy in the prone position: evaluation using cumulative sum techniques.

PubMed

Oshikiri, Taro; Yasuda, Takashi; Yamamoto, Masashi; Kanaji, Shingo; Yamashita, Kimihiro; Matsuda, Takeru; Sumi, Yasuo; Nakamura, Tetsu; Fujino, Yasuhiro; Tominaga, Masahiro; Suzuki, Satoshi; Kakeji, Yoshihiro

2016-09-01

Minimally invasive esophagectomy (MIE) has less morbidity than the open approach. In particular, thoracoscopic esophagectomy in the prone position (TEP) has been performed worldwide. Using the cumulative sum control chart (CUSUM) method, this study aimed to confirm whether a trainee surgeon who learned established standards would become skilled in TEP with a shorter learning curve than that of the mentoring surgeon. Surgeon A performed TEP in 100 patients; the first 22 patients comprised period 1. His learning curve, defined based on the operation time (OT) of the thoracic procedure, was evaluated using the CUSUM method, and short-term outcomes were assessed. Another 22 patients underwent TEP performed by surgeon B, with outcomes compared to those of surgeon A's period 1. Using the CUSUM chart, the peak point of the thoracic procedure OT occurred at the 44th case in surgeon A's experience of 100 cases. With surgeon A's first 22 cases (period 1), the peak point of the thoracic procedure OT could not be confirmed and graph is expanding soaring at CUSUM chart. The CUSUM chart of surgeon B's experience of 22 cases clearly indicated that the peak point of the thoracic procedure OT occurred at the 17th case. The rate of recurrent laryngeal nerve palsy for surgeon B (9 %) was significantly lower than for surgeon A in period 1 (36 %) (p = 0.0266). There is some possibility for a trainee surgeon to attain the required basic skills to perform TEP in a relatively short period of time using a standardized procedure developed by a mentoring surgeon. The CUSUM method should be useful in evaluating trainee competence during an initial series of procedures, by assessing the learning curve defined by OT.

Photometric classification of type Ia supernovae in the SuperNova Legacy Survey with supervised learning

DOE Office of Scientific and Technical Information (OSTI.GOV)

Möller, A.; Ruhlmann-Kleider, V.; Leloup, C.

In the era of large astronomical surveys, photometric classification of supernovae (SNe) has become an important research field due to limited spectroscopic resources for candidate follow-up and classification. In this work, we present a method to photometrically classify type Ia supernovae based on machine learning with redshifts that are derived from the SN light-curves. This method is implemented on real data from the SNLS deferred pipeline, a purely photometric pipeline that identifies SNe Ia at high-redshifts (0.2 < z < 1.1). Our method consists of two stages: feature extraction (obtaining the SN redshift from photometry and estimating light-curve shape parameters)more » and machine learning classification. We study the performance of different algorithms such as Random Forest and Boosted Decision Trees. We evaluate the performance using SN simulations and real data from the first 3 years of the Supernova Legacy Survey (SNLS), which contains large spectroscopically and photometrically classified type Ia samples. Using the Area Under the Curve (AUC) metric, where perfect classification is given by 1, we find that our best-performing classifier (Extreme Gradient Boosting Decision Tree) has an AUC of 0.98.We show that it is possible to obtain a large photometrically selected type Ia SN sample with an estimated contamination of less than 5%. When applied to data from the first three years of SNLS, we obtain 529 events. We investigate the differences between classifying simulated SNe, and real SN survey data. In particular, we find that applying a thorough set of selection cuts to the SN sample is essential for good classification. This work demonstrates for the first time the feasibility of machine learning classification in a high- z SN survey with application to real SN data.« less

Variation of safety indices during in the learning curve for color Doppler assessment of the fetal heart at 11+0 to 13+6 weeks' gestation.

PubMed

Nemescu, Dragos; Berescu, Anca; Rotariu, Cristian

2015-12-01

The aim of our study was to analyze the variation of acoustic output, as expressed by the thermal (TI) and mechanical index (MI), during the learning curve for a fetal heart scan at 11-13 gestational weeks, with the introduction of a new ultrasound system. This was a prospective, observational study on 303 normal fetuses. The fetal heart was examined transabdominally using B-Mode and high definition (HD) color Doppler to obtain standard parameters: four-chamber, outflow tracts and three-vessel-trachea views. Data were analyzed in groups of 20 consecutive examinations and the percentage of successful examinations was calculated. TI and MI were retrieved from HD color Doppler examinations of the fetal heart and from pulsed-wave Doppler assessment of the tricuspid flow and ductus venosus. MI values from the color Doppler examination of the fetal heart showed a continuous decrease (0.81 to 0.75, p<0.001), along the learning phase. TI and MI indices from pulsed-wave Doppler evaluation of the tricuspid flow increased at the beginning of the learning phase and stabilized afterwards (0.34 to 0.36, p<0.05 and 0.37 to 0.4, p<0.001, respectively). TI from color Doppler exam of the heart and indices from ductus venosus assessment were very constant and did not change along the studied periods. The length of Doppler examination of the heart increased after about 80 cases by 25%, to a mean of 4 minutes (p<0.05). Safety indices from Doppler evaluation of the fetal heart and tricuspid flow vary during the learning curve for fetal heart assessment. Also, the occurrence of constant values suggests the potential for their supplementary active reduction. For a better adaptation to a new ultrasound technology, the sonographer should scan the fetal heart longer in the first trimester and follow displayed safety indices along the first 80 cases.

Medical subdomain classification of clinical notes using a machine learning-based natural language processing approach.

PubMed

Weng, Wei-Hung; Wagholikar, Kavishwar B; McCray, Alexa T; Szolovits, Peter; Chueh, Henry C

2017-12-01

The medical subdomain of a clinical note, such as cardiology or neurology, is useful content-derived metadata for developing machine learning downstream applications. To classify the medical subdomain of a note accurately, we have constructed a machine learning-based natural language processing (NLP) pipeline and developed medical subdomain classifiers based on the content of the note. We constructed the pipeline using the clinical NLP system, clinical Text Analysis and Knowledge Extraction System (cTAKES), the Unified Medical Language System (UMLS) Metathesaurus, Semantic Network, and learning algorithms to extract features from two datasets - clinical notes from Integrating Data for Analysis, Anonymization, and Sharing (iDASH) data repository (n = 431) and Massachusetts General Hospital (MGH) (n = 91,237), and built medical subdomain classifiers with different combinations of data representation methods and supervised learning algorithms. We evaluated the performance of classifiers and their portability across the two datasets. The convolutional recurrent neural network with neural word embeddings trained-medical subdomain classifier yielded the best performance measurement on iDASH and MGH datasets with area under receiver operating characteristic curve (AUC) of 0.975 and 0.991, and F1 scores of 0.845 and 0.870, respectively. Considering better clinical interpretability, linear support vector machine-trained medical subdomain classifier using hybrid bag-of-words and clinically relevant UMLS concepts as the feature representation, with term frequency-inverse document frequency (tf-idf)-weighting, outperformed other shallow learning classifiers on iDASH and MGH datasets with AUC of 0.957 and 0.964, and F1 scores of 0.932 and 0.934 respectively. We trained classifiers on one dataset, applied to the other dataset and yielded the threshold of F1 score of 0.7 in classifiers for half of the medical subdomains we studied. Our study shows that a supervised learning-based NLP approach is useful to develop medical subdomain classifiers. The deep learning algorithm with distributed word representation yields better performance yet shallow learning algorithms with the word and concept representation achieves comparable performance with better clinical interpretability. Portable classifiers may also be used across datasets from different institutions.

Gene Scanning of an Internalin B Gene Fragment Using High-Resolution Melting Curve Analysis as a Tool for Rapid Typing of Listeria monocytogenes

PubMed Central

Pietzka, Ariane T.; Stöger, Anna; Huhulescu, Steliana; Allerberger, Franz; Ruppitsch, Werner

2011-01-01

The ability to accurately track Listeria monocytogenes strains involved in outbreaks is essential for control and prevention of listeriosis. Because current typing techniques are time-consuming, cost-intensive, technically demanding, and difficult to standardize, we developed a rapid and cost-effective method for typing of L. monocytogenes. In all, 172 clinical L. monocytogenes isolates and 20 isolates from culture collections were typed by high-resolution melting (HRM) curve analysis of a specific locus of the internalin B gene (inlB). All obtained HRM curve profiles were verified by sequence analysis. The 192 tested L. monocytogenes isolates yielded 15 specific HRM curve profiles. Sequence analysis revealed that these 15 HRM curve profiles correspond to 18 distinct inlB sequence types. The HRM curve profiles obtained correlated with the five phylogenetic groups I.1, I.2, II.1, II.2, and III. Thus, HRM curve analysis constitutes an inexpensive assay and represents an improvement in typing relative to classical serotyping or multiplex PCR typing protocols. This method provides a rapid and powerful screening tool for simultaneous preliminary typing of up to 384 samples in approximately 2 hours. PMID:21227395

Supervised filters for EEG signal in naturally occurring epilepsy forecasting.

PubMed

Muñoz-Almaraz, Francisco Javier; Zamora-Martínez, Francisco; Botella-Rocamora, Paloma; Pardo, Juan

2017-01-01

Nearly 1% of the global population has Epilepsy. Forecasting epileptic seizures with an acceptable confidence level, could improve the disease treatment and thus the lifestyle of the people who suffer it. To do that the electroencephalogram (EEG) signal is usually studied through spectral power band filtering, but this paper proposes an alternative novel method of preprocessing the EEG signal based on supervised filters. Such filters have been employed in a machine learning algorithm, such as the K-Nearest Neighbor (KNN), to improve the prediction of seizures. The proposed solution extends with this novel approach an algorithm that was submitted to win the third prize of an international Data Science challenge promoted by Kaggle contest platform and the American Epilepsy Society, the Epilepsy Foundation, National Institutes of Health (NIH) and Mayo Clinic. A formal description of these preprocessing methods is presented and a detailed analysis in terms of Receiver Operating Characteristics (ROC) curve and Area Under ROC curve is performed. The obtained results show statistical significant improvements when compared with the spectral power band filtering (PBF) typical baseline. A trend between performance and the dataset size is observed, suggesting that the supervised filters bring better information, compared to the conventional PBF filters, as the dataset grows in terms of monitored variables (sensors) and time length. The paper demonstrates a better accuracy in forecasting when new filters are employed and its main contribution is in the field of machine learning algorithms to develop more accurate predictive systems.

Supervised filters for EEG signal in naturally occurring epilepsy forecasting

PubMed Central

2017-01-01

Nearly 1% of the global population has Epilepsy. Forecasting epileptic seizures with an acceptable confidence level, could improve the disease treatment and thus the lifestyle of the people who suffer it. To do that the electroencephalogram (EEG) signal is usually studied through spectral power band filtering, but this paper proposes an alternative novel method of preprocessing the EEG signal based on supervised filters. Such filters have been employed in a machine learning algorithm, such as the K-Nearest Neighbor (KNN), to improve the prediction of seizures. The proposed solution extends with this novel approach an algorithm that was submitted to win the third prize of an international Data Science challenge promoted by Kaggle contest platform and the American Epilepsy Society, the Epilepsy Foundation, National Institutes of Health (NIH) and Mayo Clinic. A formal description of these preprocessing methods is presented and a detailed analysis in terms of Receiver Operating Characteristics (ROC) curve and Area Under ROC curve is performed. The obtained results show statistical significant improvements when compared with the spectral power band filtering (PBF) typical baseline. A trend between performance and the dataset size is observed, suggesting that the supervised filters bring better information, compared to the conventional PBF filters, as the dataset grows in terms of monitored variables (sensors) and time length. The paper demonstrates a better accuracy in forecasting when new filters are employed and its main contribution is in the field of machine learning algorithms to develop more accurate predictive systems. PMID:28632737

Learning Curves: Making Quality Online Health Information Available at a Fitness Center

PubMed Central

Dobbins, Montie T.; Tarver, Talicia; Adams, Mararia; Jones, Dixie A.

2012-01-01

Meeting consumer health information needs can be a challenge. Research suggests that women seek health information from a variety of resources, including the Internet. In an effort to make women aware of reliable health information sources, the Louisiana State University Health Sciences Center – Shreveport Medical Library engaged in a partnership with a franchise location of Curves International, Inc. This article will discuss the project, its goals and its challenges. PMID:22545020

Monograph on propagation of sound waves in curved ducts

NASA Technical Reports Server (NTRS)

Rostafinski, Wojciech

1991-01-01

After reviewing and evaluating the existing material on sound propagation in curved ducts without flow, it seems strange that, except for Lord Rayleigh in 1878, no book on acoustics has treated the case of wave motion in bends. This monograph reviews the available analytical and experimental material, nearly 30 papers published on this subject so far, and concisely summarizes what has been learned about the motion of sound in hard-wall and acoustically lined cylindrical bends.

Impact of cap-assisted colonoscopy on the learning curve and quality in colonoscopy: a randomized controlled trial.

PubMed

Tang, Zhouwen; Zhang, Daniel S; Thrift, Aaron P; Patel, Kalpesh K

2018-03-01

Colonoscopy competency assessment in trainees traditionally has been informal. Comprehensive metrics such as the Assessment of Competency in Endoscopy (ACE) tool suggest that competency thresholds are higher than assumed. Cap-assisted colonoscopy (CAC) may improve competency, but data regarding novice trainees are lacking. We compared CAC versus standard colonoscopy (SC) performed by novice trainees in a randomized controlled trial. All colonoscopies performed by 3 gastroenterology fellows without prior experience were eligible for the study. Exclusion criteria included patient age <18 or >90 years, pregnancy, prior colon resection, diverticulitis, colon obstruction, severe hematochezia, referral for EMR, or a procedure done without patient sedation. Patients were randomized to either CAC or SC in a 1:1 fashion. The primary outcome was the independent cecal intubation rate (ICIR). Secondary outcomes were cecal intubation time, polyp detection rate, polyp miss rate, adenoma detection rate, ACE tool scores, and cumulative summation learning curves. A total of 203 colonoscopies were analyzed, 101 in CAC and 102 in SC. CAC resulted in a significantly higher cecal intubation rate, at 79.2% in CAC compared with 66.7% in SC (P = .04). Overall cecal intubation time was significantly shorter at 13.7 minutes for CAC versus 16.5 minutes for SC (P =.02). Cecal intubation time in the case of successful independent fellow intubation was not significantly different between CAC and SC (11.6 minutes vs 12.7 minutes; P = .29). Overall ACE tool motor and cognitive scores were higher with CAC. Learning curves for ICIR approached the competency threshold earlier with cap use but reached competency for only 1 fellow. The polyp detection rate, polyp miss rate, and adenoma detection rate were not significantly different between groups. CAC resulted in significant improvement in ICIR, overall ACE tool scores, and trend toward competency on learning curves when compared with SC in colonoscopy trainees without prior colonoscopy experience. (Clinical trial registration number: NCT02472730.). Copyright © 2018 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.

Defining the learning curve for multiparametric magnetic resonance imaging (MRI) of the prostate using MRI-transrectal ultrasonography (TRUS) fusion-guided transperineal prostate biopsies as a validation tool.

PubMed

Gaziev, Gabriele; Wadhwa, Karan; Barrett, Tristan; Koo, Brendan C; Gallagher, Ferdia A; Serrao, Eva; Frey, Julia; Seidenader, Jonas; Carmona, Lina; Warren, Anne; Gnanapragasam, Vincent; Doble, Andrew; Kastner, Christof

2016-01-01

To determine the accuracy of multiparametric magnetic resonance imaging (mpMRI) during the learning curve of radiologists using MRI targeted, transrectal ultrasonography (TRUS) guided transperineal fusion biopsy (MTTP) for validation. Prospective data on 340 men who underwent mpMRI (T2-weighted and diffusion-weighted MRI) followed by MTTP prostate biopsy, was collected according to Ginsburg Study Group and Standards for Reporting of Diagnostic Accuracy standards. MRI data were reported by two experienced radiologists and scored on a Likert scale. Biopsies were performed by consultant urologists not 'blinded' to the MRI result and men had both targeted and systematic sector biopsies, which were reviewed by a dedicated uropathologist. The cohorts were divided into groups representing five consecutive time intervals in the study. Sensitivity and specificity of positive MRI reports, prostate cancer detection by positive MRI, distribution of significant Gleason score and negative MRI with false negative for prostate cancer were calculated. Data were sequentially analysed and the learning curve was determined by comparing the first and last group. We detected a positive mpMRI in 64 patients from Group A (91%) and 52 patients from Group E (74%). The prostate cancer detection rate on mpMRI increased from 42% (27/64) in Group A to 81% (42/52) in Group E (P < 0.001). The prostate cancer detection rate by targeted biopsy increased from 27% (17/64) in Group A to 63% (33/52) in Group E (P < 0.001). The negative predictive value of MRI for significant cancer (>Gleason 3+3) was 88.9% in Group E compared with 66.6% in Group A. We demonstrate an improvement in detection of prostate cancer for MRI reporting over time, suggesting a learning curve for the technique. With an improved negative predictive value for significant cancer, decision for biopsy should be based on patient/surgeon factors and risk attributes alongside the MRI findings. © 2014 The Authors BJU International © 2014 BJU International Published by John Wiley & Sons Ltd.

Unsupervised classification of variable stars

NASA Astrophysics Data System (ADS)

Valenzuela, Lucas; Pichara, Karim

2018-03-01

During the past 10 years, a considerable amount of effort has been made to develop algorithms for automatic classification of variable stars. That has been primarily achieved by applying machine learning methods to photometric data sets where objects are represented as light curves. Classifiers require training sets to learn the underlying patterns that allow the separation among classes. Unfortunately, building training sets is an expensive process that demands a lot of human efforts. Every time data come from new surveys; the only available training instances are the ones that have a cross-match with previously labelled objects, consequently generating insufficient training sets compared with the large amounts of unlabelled sources. In this work, we present an algorithm that performs unsupervised classification of variable stars, relying only on the similarity among light curves. We tackle the unsupervised classification problem by proposing an untraditional approach. Instead of trying to match classes of stars with clusters found by a clustering algorithm, we propose a query-based method where astronomers can find groups of variable stars ranked by similarity. We also develop a fast similarity function specific for light curves, based on a novel data structure that allows scaling the search over the entire data set of unlabelled objects. Experiments show that our unsupervised model achieves high accuracy in the classification of different types of variable stars and that the proposed algorithm scales up to massive amounts of light curves.

da Vinci skills simulator for assessing learning curve and criterion-based training of robotic basic skills.

PubMed

Brinkman, Willem M; Luursema, Jan-Maarten; Kengen, Bas; Schout, Barbara M A; Witjes, J Alfred; Bekkers, Ruud L

2013-03-01

To answer 2 research questions: what are the learning curve patterns of novices on the da Vinci skills simulator parameters and what parameters are appropriate for criterion-based robotic training. A total of 17 novices completed 2 simulator sessions within 3 days. Each training session consisted of a warming-up exercise, followed by 5 repetitions of the "ring and rail II" task. Expert participants (n = 3) performed a warming-up exercise and 3 repetitions of the "ring and rail II" task on 1 day. We analyzed all 9 parameters of the simulator. Significant learning occurred on 5 parameters: overall score, time to complete, instrument collision, instruments out of view, and critical errors within 1-10 repetitions (P <.05). Economy of motion and excessive instrument force only showed improvement within the first 5 repetitions. No significant learning on the parameter drops and master workspace range was found. Using the expert overall performance score (n = 3) as a criterion (overall score 90%), 9 of 17 novice participants met the criterion within 10 repetitions. Most parameters showed that basic robotic skills are learned relatively quickly using the da Vinci skills simulator, but that 10 repetitions were not sufficient for most novices to reach an expert level. Some parameters seemed inappropriate for expert-based criterion training because either no learning occurred or the novice performance was equal to expert performance. Copyright © 2013 Elsevier Inc. All rights reserved.

Home and Preschool Learning Environments and Their Relations to the Development of Early Numeracy Skills

ERIC Educational Resources Information Center

Anders, Yvonne; Rossbach, Hans-Gunther; Weinert, Sabine; Ebert, Susanne; Kuger, Susanne; Lehrl, Simone; von Maurice, Jutta

2012-01-01

This study examined the influence of the quality of home and preschool learning environments on the development of early numeracy skills in Germany, drawing on a sample of 532 children in 97 preschools. Latent growth curve models were used to investigate early numeracy skills and their development from the first (average age: 3 years) to the third…

A Software Technology Transition Entropy Based Engineering Model

DTIC Science & Technology

2002-03-01

Systems Basics, p273). (Prigogine 1997 p81). It is not the place of this research to provide a mathematical formalism with theorems and lemmas. Rather...science). The ancient philosophers, 27 Pythagoras , Protagoras, Socrates, and Plato start the first discourse (the message) that has continued...unpacking of the technology "message" from Pythagoras . This process is characterized by accumulation learning, modeled by learning curves in

Use of Latent Growth Curve Modeling for Assessing the Effects of Summer and After-School Learning on Adolescent Students' Achievement Gap

ERIC Educational Resources Information Center

Lin, Chunn-Ying; Hsieh, Ya-Heng; Chen, Cheng-Hung

2015-01-01

Many Western researchers have found that the gaps in the learning progress between students from different socioeconomic statuses primarily occur due to the accumulated effects of long summer vacations, rather than during the school years. However, it remains to be seen whether these findings can be cross-culturally applied to children in Taiwan.…

Screening Electronic Health Record-Related Patient Safety Reports Using Machine Learning.

PubMed

Marella, William M; Sparnon, Erin; Finley, Edward

2017-03-01

The objective of this study was to develop a semiautomated approach to screening cases that describe hazards associated with the electronic health record (EHR) from a mandatory, population-based patient safety reporting system. Potentially relevant cases were identified through a query of the Pennsylvania Patient Safety Reporting System. A random sample of cases were manually screened for relevance and divided into training, testing, and validation data sets to develop a machine learning model. This model was used to automate screening of remaining potentially relevant cases. Of the 4 algorithms tested, a naive Bayes kernel performed best, with an area under the receiver operating characteristic curve of 0.927 ± 0.023, accuracy of 0.855 ± 0.033, and F score of 0.877 ± 0.027. The machine learning model and text mining approach described here are useful tools for identifying and analyzing adverse event and near-miss reports. Although reporting systems are beginning to incorporate structured fields on health information technology and the EHR, these methods can identify related events that reporters classify in other ways. These methods can facilitate analysis of legacy safety reports by retrieving health information technology-related and EHR-related events from databases without fields and controlled values focused on this subject and distinguishing them from reports in which the EHR is mentioned only in passing. Machine learning and text mining are useful additions to the patient safety toolkit and can be used to semiautomate screening and analysis of unstructured text in safety reports from frontline staff.

Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques

PubMed Central

Macyszyn, Luke; Akbari, Hamed; Pisapia, Jared M.; Da, Xiao; Attiah, Mark; Pigrish, Vadim; Bi, Yingtao; Pal, Sharmistha; Davuluri, Ramana V.; Roccograndi, Laura; Dahmane, Nadia; Martinez-Lage, Maria; Biros, George; Wolf, Ronald L.; Bilello, Michel; O'Rourke, Donald M.; Davatzikos, Christos

2016-01-01

Background MRI characteristics of brain gliomas have been used to predict clinical outcome and molecular tumor characteristics. However, previously reported imaging biomarkers have not been sufficiently accurate or reproducible to enter routine clinical practice and often rely on relatively simple MRI measures. The current study leverages advanced image analysis and machine learning algorithms to identify complex and reproducible imaging patterns predictive of overall survival and molecular subtype in glioblastoma (GB). Methods One hundred five patients with GB were first used to extract approximately 60 diverse features from preoperative multiparametric MRIs. These imaging features were used by a machine learning algorithm to derive imaging predictors of patient survival and molecular subtype. Cross-validation ensured generalizability of these predictors to new patients. Subsequently, the predictors were evaluated in a prospective cohort of 29 new patients. Results Survival curves yielded a hazard ratio of 10.64 for predicted long versus short survivors. The overall, 3-way (long/medium/short survival) accuracy in the prospective cohort approached 80%. Classification of patients into the 4 molecular subtypes of GB achieved 76% accuracy. Conclusions By employing machine learning techniques, we were able to demonstrate that imaging patterns are highly predictive of patient survival. Additionally, we found that GB subtypes have distinctive imaging phenotypes. These results reveal that when imaging markers related to infiltration, cell density, microvascularity, and blood–brain barrier compromise are integrated via advanced pattern analysis methods, they form very accurate predictive biomarkers. These predictive markers used solely preoperative images, hence they can significantly augment diagnosis and treatment of GB patients. PMID:26188015

Classification of Fowl Adenovirus Serotypes by Use of High-Resolution Melting-Curve Analysis of the Hexon Gene Region▿

PubMed Central

Steer, Penelope A.; Kirkpatrick, Naomi C.; O'Rourke, Denise; Noormohammadi, Amir H.

2009-01-01

Identification of fowl adenovirus (FAdV) serotypes is of importance in epidemiological studies of disease outbreaks and the adoption of vaccination strategies. In this study, real-time PCR and subsequent high-resolution melting (HRM)-curve analysis of three regions of the hexon gene were developed and assessed for their potential in differentiating 12 FAdV reference serotypes. The results were compared to previously described PCR and restriction enzyme analyses of the hexon gene. Both HRM-curve analysis of a 191-bp region of the hexon gene and restriction enzyme analysis failed to distinguish a number of serotypes used in this study. In addition, PCR of the region spanning nucleotides (nt) 144 to 1040 failed to amplify FAdV-5 in sufficient quantities for further analysis. However, HRM-curve analysis of the region spanning nt 301 to 890 proved a sensitive and specific method of differentiating all 12 serotypes. All melt curves were highly reproducible, and replicates of each serotype were correctly genotyped with a mean confidence value of more than 99% using normalized HRM curves. Sequencing analysis revealed that each profile was related to a unique sequence, with some sequences sharing greater than 94% identity. Melting-curve profiles were found to be related mainly to GC composition and distribution throughout the amplicons, regardless of sequence identity. The results presented in this study show that the closed-tube method of PCR and HRM-curve analysis provides an accurate, rapid, and robust genotyping technique for the identification of FAdV serotypes and can be used as a model for developing genotyping techniques for other pathogens. PMID:19036935

Thermoluminescence glow curve analysis and CGCD method for erbium doped CaZrO{sub 3} phosphor

DOE Office of Scientific and Technical Information (OSTI.GOV)

Tiwari, Ratnesh, E-mail: 31rati@gmail.com; Chopra, Seema

2016-05-06

The manuscript report the synthesis, thermoluminescence study at fixed concentration of Er{sup 3+} (1 mol%) doped CaZrO{sub 3} phosphor. The phosphors were prepared by modified solid state reaction method. The powder sample was characterized by thermoluminescence (TL) glow curve analysis. In TL glow curve the optimized concentration in 1mol% for UV irradiated sample. The kinetic parameters were calculated by computerized glow curve deconvolution (CGCD) techniaue. Trapping parameters gives the information of dosimetry loss in prepared phosphor and its usability in environmental monitoring and for personal monitoring. CGCD is the advance tool for analysis of complicated TL glow curves.

Learning curve with minimally invasive unicompartmental knee arthroplasty.

PubMed

Hamilton, William G; Ammeen, Deborah; Engh, C Anderson; Engh, Gerard A

2010-08-01

This study examined 445 consecutive minimally invasive unicompartmental knee arthroplasties (UKAs) from one institution to determine whether revision and reoperation rates would decrease as the number of cases performed increased, indicating the presence of a learning curve with this procedure. At a mean of 3.25 years, 26 knees required revision yielding an overall revision rate of 5.8%; survivorship at 2 years with revision as an end point was 96% +/- 1.7%. Both revisions and reoperations decreased over time but not significantly. For the first half of UKA cases performed vs the second half, revision rates fell from 5.0% to 2.5%, and reoperation rates fell from 8.1% to 5.4%. These data demonstrate that despite modifications made to improve surgical technique across time, a substantial complication rate with this procedure persists. Copyright (c) 2010 Elsevier Inc. All rights reserved.

Shallow Transits—Deep Learning. I. Feasibility Study of Deep Learning to Detect Periodic Transits of Exoplanets

NASA Astrophysics Data System (ADS)

Zucker, Shay; Giryes, Raja

2018-04-01

Transits of habitable planets around solar-like stars are expected to be shallow, and to have long periods, which means low information content. The current bottleneck in the detection of such transits is caused in large part by the presence of red (correlated) noise in the light curves obtained from the dedicated space telescopes. Based on the groundbreaking results deep learning achieves in many signal and image processing applications, we propose to use deep neural networks to solve this problem. We present a feasibility study, in which we applied a convolutional neural network on a simulated training set. The training set comprised light curves received from a hypothetical high-cadence space-based telescope. We simulated the red noise by using Gaussian Processes with a wide variety of hyper-parameters. We then tested the network on a completely different test set simulated in the same way. Our study proves that very difficult cases can indeed be detected. Furthermore, we show how detection trends can be studied and detection biases quantified. We have also checked the robustness of the neural-network performance against practical artifacts such as outliers and discontinuities, which are known to affect space-based high-cadence light curves. Future work will allow us to use the neural networks to characterize the transit model and identify individual transits. This new approach will certainly be an indispensable tool for the detection of habitable planets in the future planet-detection space missions such as PLATO.

Machine learning search for variable stars

NASA Astrophysics Data System (ADS)

Pashchenko, Ilya N.; Sokolovsky, Kirill V.; Gavras, Panagiotis

2018-04-01

Photometric variability detection is often considered as a hypothesis testing problem: an object is variable if the null hypothesis that its brightness is constant can be ruled out given the measurements and their uncertainties. The practical applicability of this approach is limited by uncorrected systematic errors. We propose a new variability detection technique sensitive to a wide range of variability types while being robust to outliers and underestimated measurement uncertainties. We consider variability detection as a classification problem that can be approached with machine learning. Logistic Regression (LR), Support Vector Machines (SVM), k Nearest Neighbours (kNN), Neural Nets (NN), Random Forests (RF), and Stochastic Gradient Boosting classifier (SGB) are applied to 18 features (variability indices) quantifying scatter and/or correlation between points in a light curve. We use a subset of Optical Gravitational Lensing Experiment phase two (OGLE-II) Large Magellanic Cloud (LMC) photometry (30 265 light curves) that was searched for variability using traditional methods (168 known variable objects) as the training set and then apply the NN to a new test set of 31 798 OGLE-II LMC light curves. Among 205 candidates selected in the test set, 178 are real variables, while 13 low-amplitude variables are new discoveries. The machine learning classifiers considered are found to be more efficient (select more variables and fewer false candidates) compared to traditional techniques using individual variability indices or their linear combination. The NN, SGB, SVM, and RF show a higher efficiency compared to LR and kNN.

Surgical experience and complications during endonasal sinus surgery.

PubMed

Keerl, R; Stankiewicz, J; Weber, R; Hosemann, W; Draf, W

1999-04-01

The introduction of optical aids for endonasal sinus surgery has not produced the expected drop in the rate of serious intraoperative complications. 1. Retrospectively, consecutive procedures of different surgeons were analyzed in regard to major complications (periorbital injury, orbital lesion, dural injury, endocranial lesion, damage to the internal carotid artery). The chronological distribution was transformed into a personal learning curve. 2. From our own experience and as surveyors, we analyzed the experiences of surgeons having encountered severe complications and compared them with the above-mentioned learning curve. In total, 1,500 operations carried out by five surgeons with 16 serious complications were assessed. For the learning curve, the following stages were defined. stage I: greatest risk of complication, with dural injury (1st to 30th operation); stage II, slighter risk of complication, with frequent periorbital injuries (31st to 180th operation); and stage III, least risk, corresponding to an experienced surgeon. Serious complications occur most frequently among experienced surgeons. The beginner enjoys the most effective type of assistance, in the form of personal guidance of an experienced surgeon who is constantly present during the first 30 operations, and who should then be readily available during the next 70 operative procedures. The use of multimedia software appears to be helpful, though its actual value still remains to be determined. The experienced surgeon in particular must be willing to exercise repeated self-criticism to keep his or her rate of complications to a minimum.

How to start a minimal access mitral valve program

PubMed Central

2013-01-01

The seven pillars of governance established by the National Health Service in the United Kingdom provide a useful framework for the process of introducing new procedures to a hospital. Drawing from local experience, the author present guidance for institutions considering establishing a minimal access mitral valve program. The seven pillars of governance apply to the practice of minimally invasive mitral valve surgery, based on the principle of patient-centred practice. The author delineate the benefits of minimally invasive mitral valve surgery in terms of: “clinical effectiveness”, including reduced length of hospital stay, “risk management effectiveness”, including conversion to sternotomy and aortic dissection, “patient experience” including improved cosmesis and quicker recovery, and the effectiveness of communication, resources and strategies in the implementation of minimally invasive mitral valve surgery. Finally, the author have identified seven learning curves experienced by surgeons involved in introducing a minimal access mitral valve program. The learning curves are defined as: techniques of mitral valve repair, Transoesophageal Echocardiography-guided cannulation, incisions, instruments, visualization, aortic occlusion and cardiopulmonary bypass strategies. From local experience, the author provide advice on how to reduce the learning curves, such as practising with the specialised instruments and visualization techniques during sternotomy cases. Underpinning the NHS pillars are the principles of systems awareness, teamwork, communication, ownership and leadership, all of which are paramount to performing any surgery but more so with minimal access surgery, as will be highlighted throughout this paper. PMID:24349981

Heidelberg retina tomograph measurements of the optic disc and parapapillary retina for detecting glaucoma analyzed by machine learning classifiers.

PubMed

Zangwill, Linda M; Chan, Kwokleung; Bowd, Christopher; Hao, Jicuang; Lee, Te-Won; Weinreb, Robert N; Sejnowski, Terrence J; Goldbaum, Michael H

2004-09-01

To determine whether topographical measurements of the parapapillary region analyzed by machine learning classifiers can detect early to moderate glaucoma better than similarly processed measurements obtained within the disc margin and to improve methods for optimization of machine learning classifier feature selection. One eye of each of 95 patients with early to moderate glaucomatous visual field damage and of each of 135 normal subjects older than 40 years participating in the longitudinal Diagnostic Innovations in Glaucoma Study (DIGS) were included. Heidelberg Retina Tomograph (HRT; Heidelberg Engineering, Dossenheim, Germany) mean height contour was measured in 36 equal sectors, both along the disc margin and in the parapapillary region (at a mean contour line radius of 1.7 mm). Each sector was evaluated individually and in combination with other sectors. Gaussian support vector machine (SVM) learning classifiers were used to interpret HRT sector measurements along the disc margin and in the parapapillary region, to differentiate between eyes with normal and glaucomatous visual fields and to compare the results with global and regional HRT parameter measurements. The area under the receiver operating characteristic (ROC) curve was used to measure diagnostic performance of the HRT parameters and to evaluate the cross-validation strategies and forward selection and backward elimination optimization techniques that were used to generate the reduced feature sets. The area under the ROC curve for mean height contour of the 36 sectors along the disc margin was larger than that for the mean height contour in the parapapillary region (0.97 and 0.85, respectively). Of the 36 individual sectors along the disc margin, those in the inferior region between 240 degrees and 300 degrees, had the largest area under the ROC curve (0.85-0.91). With SVM Gaussian techniques, the regional parameters showed the best ability to discriminate between normal eyes and eyes with glaucomatous visual field damage, followed by the global parameters, mean height contour measures along the disc margin, and mean height contour measures in the parapapillary region. The area under the ROC curve was 0.98, 0.94, 0.93, and 0.85, respectively. Cross-validation and optimization techniques demonstrated that good discrimination (99% of peak area under the ROC curve) can be obtained with a reduced number of HRT parameters. Mean height contour measurements along the disc margin discriminated between normal and glaucomatous eyes better than measurements obtained in the parapapillary region. Copyright Association for Research in Vision and Ophthalmology

Heidelberg Retina Tomograph Measurements of the Optic Disc and Parapapillary Retina for Detecting Glaucoma Analyzed by Machine Learning Classifiers

PubMed Central

Zangwill, Linda M.; Chan, Kwokleung; Bowd, Christopher; Hao, Jicuang; Lee, Te-Won; Weinreb, Robert N.; Sejnowski, Terrence J.; Goldbaum, Michael H.

2010-01-01

Purpose To determine whether topographical measurements of the parapapillary region analyzed by machine learning classifiers can detect early to moderate glaucoma better than similarly processed measurements obtained within the disc margin and to improve methods for optimization of machine learning classifier feature selection. Methods One eye of each of 95 patients with early to moderate glaucomatous visual field damage and of each of 135 normal subjects older than 40 years participating in the longitudinal Diagnostic Innovations in Glaucoma Study (DIGS) were included. Heidelberg Retina Tomograph (HRT; Heidelberg Engineering, Dossenheim, Germany) mean height contour was measured in 36 equal sectors, both along the disc margin and in the parapapillary region (at a mean contour line radius of 1.7 mm). Each sector was evaluated individually and in combination with other sectors. Gaussian support vector machine (SVM) learning classifiers were used to interpret HRT sector measurements along the disc margin and in the parapapillary region, to differentiate between eyes with normal and glaucomatous visual fields and to compare the results with global and regional HRT parameter measurements. The area under the receiver operating characteristic (ROC) curve was used to measure diagnostic performance of the HRT parameters and to evaluate the cross-validation strategies and forward selection and backward elimination optimization techniques that were used to generate the reduced feature sets. Results The area under the ROC curve for mean height contour of the 36 sectors along the disc margin was larger than that for the mean height contour in the parapapillary region (0.97 and 0.85, respectively). Of the 36 individual sectors along the disc margin, those in the inferior region between 240° and 300°, had the largest area under the ROC curve (0.85–0.91). With SVM Gaussian techniques, the regional parameters showed the best ability to discriminate between normal eyes and eyes with glaucomatous visual field damage, followed by the global parameters, mean height contour measures along the disc margin, and mean height contour measures in the parapapillary region. The area under the ROC curve was 0.98, 0.94, 0.93, and 0.85, respectively. Cross-validation and optimization techniques demonstrated that good discrimination (99% of peak area under the ROC curve) can be obtained with a reduced number of HRT parameters. Conclusions Mean height contour measurements along the disc margin discriminated between normal and glaucomatous eyes better than measurements obtained in the parapapillary region. PMID:15326133

Folded concave penalized learning in identifying multimodal MRI marker for Parkinson’s disease

PubMed Central

Liu, Hongcheng; Du, Guangwei; Zhang, Lijun; Lewis, Mechelle M.; Wang, Xue; Yao, Tao; Li, Runze; Huang, Xuemei

2016-01-01

Background Brain MRI holds promise to gauge different aspects of Parkinson’s disease (PD)-related pathological changes. Its analysis, however, is hindered by the high-dimensional nature of the data. New method This study introduces folded concave penalized (FCP) sparse logistic regression to identify biomarkers for PD from a large number of potential factors. The proposed statistical procedures target the challenges of high-dimensionality with limited data samples acquired. The maximization problem associated with the sparse logistic regression model is solved by local linear approximation. The proposed procedures then are applied to the empirical analysis of multimodal MRI data. Results From 45 features, the proposed approach identified 15 MRI markers and the UPSIT, which are known to be clinically relevant to PD. By combining the MRI and clinical markers, we can enhance substantially the specificity and sensitivity of the model, as indicated by the ROC curves. Comparison to existing methods We compare the folded concave penalized learning scheme with both the Lasso penalized scheme and the principle component analysis-based feature selection (PCA) in the Parkinson’s biomarker identification problem that takes into account both the clinical features and MRI markers. The folded concave penalty method demonstrates a substantially better clinical potential than both the Lasso and PCA in terms of specificity and sensitivity. Conclusions For the first time, we applied the FCP learning method to MRI biomarker discovery in PD. The proposed approach successfully identified MRI markers that are clinically relevant. Combining these biomarkers with clinical features can substantially enhance performance. PMID:27102045

Different verbal learning strategies in autism spectrum disorder: evidence from the Rey Auditory Verbal Learning Test.

PubMed

Bowler, Dermot M; Limoges, Elyse; Mottron, Laurent

2009-06-01

The Rey Auditory Verbal Learning Test, which requires the free recall of the same list of 15 unrelated words over 5 trials, was administered to 21 high-functioning adolescents and adults with autism spectrum disorder (ASD) and 21 matched typical individuals. The groups showed similar overall levels of free recall, rates of learning over trials and subjective organisation of their recall. However, the primacy portion of the serial position curve of the ASD participants showed slower growth over trials than that of the typical participants. The implications of this finding for our understanding of memory in ASD are discussed.

On the phenomenon of curved microcracks in /(S)/90n/s laminates - Their shapes, initiation angles and locations

NASA Technical Reports Server (NTRS)

Hu, Shoufeng; Bark, Jong S.; Nairn, John A.

1993-01-01

A variational analysis of the stress state in microcracked cross-ply laminates has been used to investigate the phenomenon of curved microcracking in /(S)/90n/s laminates. Previous investigators proposed that the initiation and orientation of curved microcracks are controlled by local maxima and stress trajectories of the principal stresses. We have implemented a principal stress model using a variational mechanics stress analysis and we were able to make predictions about curved microcracks. The predictions agree well with experimental observations and therefore support the assertion that the variational analysis gives an accurate stress state that is useful for modeling the microcracking properties of cross-ply laminates. An important prediction about curved microcracks is that they are a late stage of microcracking damage. They occur only when the crack density of straight microcracks exceeds the critical crack density for curved microcracking. The predicted critical crack density for curved microcracking agrees well with experimental observations.

Ten-year experience in managing a capitated ophthalmology carve-out by an academic eye center.

PubMed

Olson, R J

1997-01-01

A 10-year experience of managing a capitated opthalmology carve-out by an academic health unit is presented. Lessons learned regarding pricing, utilization, and managing this contract are discussed. Handling the cost of education and remaining competitive is presented as a not-insurmountable hurdle. Academic health units can compete in today's environment; however, the learning curve is steep and the problems many.

Alchemical and structural distribution based representation for universal quantum machine learning

NASA Astrophysics Data System (ADS)

Faber, Felix A.; Christensen, Anders S.; Huang, Bing; von Lilienfeld, O. Anatole

2018-06-01

We introduce a representation of any atom in any chemical environment for the automatized generation of universal kernel ridge regression-based quantum machine learning (QML) models of electronic properties, trained throughout chemical compound space. The representation is based on Gaussian distribution functions, scaled by power laws and explicitly accounting for structural as well as elemental degrees of freedom. The elemental components help us to lower the QML model's learning curve, and, through interpolation across the periodic table, even enable "alchemical extrapolation" to covalent bonding between elements not part of training. This point is demonstrated for the prediction of covalent binding in single, double, and triple bonds among main-group elements as well as for atomization energies in organic molecules. We present numerical evidence that resulting QML energy models, after training on a few thousand random training instances, reach chemical accuracy for out-of-sample compounds. Compound datasets studied include thousands of structurally and compositionally diverse organic molecules, non-covalently bonded protein side-chains, (H2O)40-clusters, and crystalline solids. Learning curves for QML models also indicate competitive predictive power for various other electronic ground state properties of organic molecules, calculated with hybrid density functional theory, including polarizability, heat-capacity, HOMO-LUMO eigenvalues and gap, zero point vibrational energy, dipole moment, and highest vibrational fundamental frequency.

Computer Aided Braille Trainer

PubMed Central

Sibert, Thomas W.

1984-01-01

The problems involved in teaching visually impaired persons to Braille are numerous. Training while the individual is still sighted and using a computer to assist is one way of shortening the learning curve. Such a solution is presented here.

Evaluation of pavement surface friction treatments.

DOT National Transportation Integrated Search

2011-12-01

The implementation of a pavement preservation program involves a learning curve with not only a determination to succeed, but : also the courage to fail. Also, successful implementation of pavement preservation program requires knowledge of the perfo...

Growth Patterns of Neuropsychological Functions in Indian Children

PubMed Central

Kar, Bhoomika R.; Rao, Shobini L.; Chandramouli, B. A.; Thennarasu, K.

2011-01-01

We investigated age-related differences in neuropsychological performance in 400 Indian school children (5–15 years of age). Functions of motor speed, attention, executive functions, visuospatial functions, comprehension, learning, and memory were examined. Growth curve analysis was performed. Different growth models fitted different cognitive functions. Neuropsychological task performance improved slowly between 5 and 7 years, moderately between 8 and 12 years and slowly between 13 and 15 years of age. The overall growth patterns of neuropsychological functions in Indian children have been discussed with the findings reported on American children. The present work describes non-linear, heterogeneous, and protracted age trends of neuropsychological functions in Indian children and adolescents. PMID:22053158

Motor Learning: An Analysis of 100 Trials of a Ski Slalom Game in Children with and without Developmental Coordination Disorder

PubMed Central

Smits-Engelsman, Bouwien C. M.; Jelsma, Lemke Dorothee; Ferguson, Gillian D.; Geuze, Reint H.

2015-01-01

Objective Although Developmental Coordination Disorder (DCD) is often characterized as a skill acquisition deficit disorder, few studies have addressed the process of motor learning. This study examined learning of a novel motor task; the Wii Fit ski slalom game. The main objectives were to determine: 1) whether learning occurs over 100 trial runs of the game, 2) if the learning curve is different between children with and without DCD, 3) if learning is different in an easier or harder version of the task, 4) if learning transfers to other balance tasks. Method 17 children with DCD (6–10 years) and a matched control group of 17 typically developing (TD) children engaged in 20 minutes of gaming, twice a week for five weeks. Each training session comprised of alternating trial runs, with five runs at an easy level and five runs at a difficult level. Wii scores, which combine speed and accuracy per run, were recorded. Standardized balance tasks were used to measure transfer. Results Significant differences in initial performance were found between groups on the Wii score and balance tasks. Both groups improved their Wii score over the five weeks. Improvement in the easy and in the hard task did not differ between groups. Retention in the time between training sessions was not different between TD and DCD groups either. The DCD group improved significantly on all balance tasks. Conclusions The findings in this study give a fairly coherent picture of the learning process over a medium time scale (5 weeks) in children novice to active computer games; they learn, retain and there is evidence of transfer to other balance tasks. The rate of motor learning is similar for those with and without DCD. Our results raise a number of questions about motor learning that need to be addressed in future research. PMID:26466324

Online adaptation and over-trial learning in macaque visuomotor control.

PubMed

Braun, Daniel A; Aertsen, Ad; Paz, Rony; Vaadia, Eilon; Rotter, Stefan; Mehring, Carsten

2011-01-01

When faced with unpredictable environments, the human motor system has been shown to develop optimized adaptation strategies that allow for online adaptation during the control process. Such online adaptation is to be contrasted to slower over-trial learning that corresponds to a trial-by-trial update of the movement plan. Here we investigate the interplay of both processes, i.e., online adaptation and over-trial learning, in a visuomotor experiment performed by macaques. We show that simple non-adaptive control schemes fail to perform in this task, but that a previously suggested adaptive optimal feedback control model can explain the observed behavior. We also show that over-trial learning as seen in learning and aftereffect curves can be explained by learning in a radial basis function network. Our results suggest that both the process of over-trial learning and the process of online adaptation are crucial to understand visuomotor learning.

Online Adaptation and Over-Trial Learning in Macaque Visuomotor Control

PubMed Central

Braun, Daniel A.; Aertsen, Ad; Paz, Rony; Vaadia, Eilon; Rotter, Stefan; Mehring, Carsten

2011-01-01

When faced with unpredictable environments, the human motor system has been shown to develop optimized adaptation strategies that allow for online adaptation during the control process. Such online adaptation is to be contrasted to slower over-trial learning that corresponds to a trial-by-trial update of the movement plan. Here we investigate the interplay of both processes, i.e., online adaptation and over-trial learning, in a visuomotor experiment performed by macaques. We show that simple non-adaptive control schemes fail to perform in this task, but that a previously suggested adaptive optimal feedback control model can explain the observed behavior. We also show that over-trial learning as seen in learning and aftereffect curves can be explained by learning in a radial basis function network. Our results suggest that both the process of over-trial learning and the process of online adaptation are crucial to understand visuomotor learning. PMID:21720526

pROC: an open-source package for R and S+ to analyze and compare ROC curves.

PubMed

Robin, Xavier; Turck, Natacha; Hainard, Alexandre; Tiberti, Natalia; Lisacek, Frédérique; Sanchez, Jean-Charles; Müller, Markus

2011-03-17

Receiver operating characteristic (ROC) curves are useful tools to evaluate classifiers in biomedical and bioinformatics applications. However, conclusions are often reached through inconsistent use or insufficient statistical analysis. To support researchers in their ROC curves analysis we developed pROC, a package for R and S+ that contains a set of tools displaying, analyzing, smoothing and comparing ROC curves in a user-friendly, object-oriented and flexible interface. With data previously imported into the R or S+ environment, the pROC package builds ROC curves and includes functions for computing confidence intervals, statistical tests for comparing total or partial area under the curve or the operating points of different classifiers, and methods for smoothing ROC curves. Intermediary and final results are visualised in user-friendly interfaces. A case study based on published clinical and biomarker data shows how to perform a typical ROC analysis with pROC. pROC is a package for R and S+ specifically dedicated to ROC analysis. It proposes multiple statistical tests to compare ROC curves, and in particular partial areas under the curve, allowing proper ROC interpretation. pROC is available in two versions: in the R programming language or with a graphical user interface in the S+ statistical software. It is accessible at http://expasy.org/tools/pROC/ under the GNU General Public License. It is also distributed through the CRAN and CSAN public repositories, facilitating its installation.

The use of machine learning for the identification of peripheral artery disease and future mortality risk.

PubMed

Ross, Elsie Gyang; Shah, Nigam H; Dalman, Ronald L; Nead, Kevin T; Cooke, John P; Leeper, Nicholas J

2016-11-01

A key aspect of the precision medicine effort is the development of informatics tools that can analyze and interpret "big data" sets in an automated and adaptive fashion while providing accurate and actionable clinical information. The aims of this study were to develop machine learning algorithms for the identification of disease and the prognostication of mortality risk and to determine whether such models perform better than classical statistical analyses. Focusing on peripheral artery disease (PAD), patient data were derived from a prospective, observational study of 1755 patients who presented for elective coronary angiography. We employed multiple supervised machine learning algorithms and used diverse clinical, demographic, imaging, and genomic information in a hypothesis-free manner to build models that could identify patients with PAD and predict future mortality. Comparison was made to standard stepwise linear regression models. Our machine-learned models outperformed stepwise logistic regression models both for the identification of patients with PAD (area under the curve, 0.87 vs 0.76, respectively; P = .03) and for the prediction of future mortality (area under the curve, 0.76 vs 0.65, respectively; P = .10). Both machine-learned models were markedly better calibrated than the stepwise logistic regression models, thus providing more accurate disease and mortality risk estimates. Machine learning approaches can produce more accurate disease classification and prediction models. These tools may prove clinically useful for the automated identification of patients with highly morbid diseases for which aggressive risk factor management can improve outcomes. Copyright © 2016 Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.

Sensitivity analysis for linear structural equation models, longitudinal mediation with latent growth models and blended learning in biostatistics education

NASA Astrophysics Data System (ADS)

Sullivan, Adam John

In chapter 1, we consider the biases that may arise when an unmeasured confounder is omitted from a structural equation model (SEM) and sensitivity analysis techniques to correct for such biases. We give an analysis of which effects in an SEM are and are not biased by an unmeasured confounder. It is shown that a single unmeasured confounder will bias not just one but numerous effects in an SEM. We present sensitivity analysis techniques to correct for biases in total, direct, and indirect effects when using SEM analyses, and illustrate these techniques with a study of aging and cognitive function. In chapter 2, we consider longitudinal mediation with latent growth curves. We define the direct and indirect effects using counterfactuals and consider the assumptions needed for identifiability of those effects. We develop models with a binary treatment/exposure followed by a model where treatment/exposure changes with time allowing for treatment/exposure-mediator interaction. We thus formalize mediation analysis with latent growth curve models using counterfactuals, makes clear the assumptions and extends these methods to allow for exposure mediator interactions. We present and illustrate the techniques with a study on Multiple Sclerosis(MS) and depression. In chapter 3, we report on a pilot study in blended learning that took place during the Fall 2013 and Summer 2014 semesters here at Harvard. We blended the traditional BIO 200: Principles of Biostatistics and created ID 200: Principles of Biostatistics and epidemiology. We used materials from the edX course PH207x: Health in Numbers: Quantitative Methods in Clinical & Public Health Research and used. These materials were used as a video textbook in which students would watch a given number of these videos prior to class. Using surveys as well as exam data we informally assess these blended classes from the student's perspective as well as a comparison of these students with students in another course, BIO 201: Introduction to Statistical Methods in Fall 2013 as well as students from BIO 200 in Fall semesters of 1992 and 1993. We then suggest improvements upon our original course designs and follow up with an informal look at how these implemented changes affected the second offering of the newly blended ID 200 in Summer 2014.

29 CFR 525.12 - Terms and conditions of special minimum wage certificates.

Code of Federal Regulations, 2011 CFR

2011-07-01

... provide them with an opportunity to overcome the initial learning curve. The persons observed shall be... of machines are not available, a second work measurement should be conducted. (i) Each worker with a...

Evaluation of pavement surface friction treatments : [technical summary].

DOT National Transportation Integrated Search

2011-01-01

The implementation of a pavement preservation program involves a learning curve with not only a determination to succeed, but also the courage to fail. Successful implementation of pavement preservation program requires knowledge of the performance o...