Study of Tissue Phantoms, Tissues, and Contrast Agent with the Biophotoacoustic Radar and Comparison to Ultrasound Imaging for Deep Subsurface Imaging

NASA Astrophysics Data System (ADS)

Alwi, R.; Telenkov, S.; Mandelis, A.; Gu, F.

2012-11-01

In this study, the imaging capability of our wide-spectrum frequency-domain photoacoustic (FD-PA) imaging alias "photoacoustic radar" methodology for imaging of soft tissues is explored. A practical application of the mathematical correlation processing method with relatively long (1 ms) frequency-modulated optical excitation is demonstrated for reconstruction of the spatial location of the PA sources. Image comparison with ultrasound (US) modality was investigated to see the complementarity between the two techniques. The obtained results with a phased array probe on tissue phantoms and their comparison to US images demonstrated that the FD-PA technique has strong potential for deep subsurface imaging with excellent contrast and high signal-to-noise ratio. FD-PA images of blood vessels in a human wrist and an in vivo subcutaneous tumor in a rat model are presented. As in other imaging modalities, the employment of contrast agents is desirable to improve the capability of medical diagnostics. Therefore, this study also evaluated and characterized the use of Food and Drug Administration (FDA)-approved superparamagnetic iron oxide nanoparticles (SPION) as PA contrast agents.

A deep learning approach to estimate chemically-treated collagenous tissue nonlinear anisotropic stress-strain responses from microscopy images.

PubMed

Liang, Liang; Liu, Minliang; Sun, Wei

2017-11-01

Biological collagenous tissues comprised of networks of collagen fibers are suitable for a broad spectrum of medical applications owing to their attractive mechanical properties. In this study, we developed a noninvasive approach to estimate collagenous tissue elastic properties directly from microscopy images using Machine Learning (ML) techniques. Glutaraldehyde-treated bovine pericardium (GLBP) tissue, widely used in the fabrication of bioprosthetic heart valves and vascular patches, was chosen to develop a representative application. A Deep Learning model was designed and trained to process second harmonic generation (SHG) images of collagen networks in GLBP tissue samples, and directly predict the tissue elastic mechanical properties. The trained model is capable of identifying the overall tissue stiffness with a classification accuracy of 84%, and predicting the nonlinear anisotropic stress-strain curves with average regression errors of 0.021 and 0.031. Thus, this study demonstrates the feasibility and great potential of using the Deep Learning approach for fast and noninvasive assessment of collagenous tissue elastic properties from microstructural images. In this study, we developed, to our best knowledge, the first Deep Learning-based approach to estimate the elastic properties of collagenous tissues directly from noninvasive second harmonic generation images. The success of this study holds promise for the use of Machine Learning techniques to noninvasively and efficiently estimate the mechanical properties of many structure-based biological materials, and it also enables many potential applications such as serving as a quality control tool to select tissue for the manufacturing of medical devices (e.g. bioprosthetic heart valves). Copyright © 2017 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

High-Resolution Ultrasound-Switchable Fluorescence Imaging in Centimeter-Deep Tissue Phantoms with High Signal-To-Noise Ratio and High Sensitivity via Novel Contrast Agents.

PubMed

Cheng, Bingbing; Bandi, Venugopal; Wei, Ming-Yuan; Pei, Yanbo; D'Souza, Francis; Nguyen, Kytai T; Hong, Yi; Yuan, Baohong

2016-01-01

For many years, investigators have sought after high-resolution fluorescence imaging in centimeter-deep tissue because many interesting in vivo phenomena-such as the presence of immune system cells, tumor angiogenesis, and metastasis-may be located deep in tissue. Previously, we developed a new imaging technique to achieve high spatial resolution in sub-centimeter deep tissue phantoms named continuous-wave ultrasound-switchable fluorescence (CW-USF). The principle is to use a focused ultrasound wave to externally and locally switch on and off the fluorophore emission from a small volume (close to ultrasound focal volume). By making improvements in three aspects of this technique: excellent near-infrared USF contrast agents, a sensitive frequency-domain USF imaging system, and an effective signal processing algorithm, for the first time this study has achieved high spatial resolution (~ 900 μm) in 3-centimeter-deep tissue phantoms with high signal-to-noise ratio (SNR) and high sensitivity (3.4 picomoles of fluorophore in a volume of 68 nanoliters can be detected). We have achieved these results in both tissue-mimic phantoms and porcine muscle tissues. We have also demonstrated multi-color USF to image and distinguish two fluorophores with different wavelengths, which might be very useful for simultaneously imaging of multiple targets and observing their interactions in the future. This work has opened the door for future studies of high-resolution centimeter-deep tissue fluorescence imaging.

Deep Learning MR Imaging-based Attenuation Correction for PET/MR Imaging.

PubMed

Liu, Fang; Jang, Hyungseok; Kijowski, Richard; Bradshaw, Tyler; McMillan, Alan B

2018-02-01

Purpose To develop and evaluate the feasibility of deep learning approaches for magnetic resonance (MR) imaging-based attenuation correction (AC) (termed deep MRAC) in brain positron emission tomography (PET)/MR imaging. Materials and Methods A PET/MR imaging AC pipeline was built by using a deep learning approach to generate pseudo computed tomographic (CT) scans from MR images. A deep convolutional auto-encoder network was trained to identify air, bone, and soft tissue in volumetric head MR images coregistered to CT data for training. A set of 30 retrospective three-dimensional T1-weighted head images was used to train the model, which was then evaluated in 10 patients by comparing the generated pseudo CT scan to an acquired CT scan. A prospective study was carried out for utilizing simultaneous PET/MR imaging for five subjects by using the proposed approach. Analysis of covariance and paired-sample t tests were used for statistical analysis to compare PET reconstruction error with deep MRAC and two existing MR imaging-based AC approaches with CT-based AC. Results Deep MRAC provides an accurate pseudo CT scan with a mean Dice coefficient of 0.971 ± 0.005 for air, 0.936 ± 0.011 for soft tissue, and 0.803 ± 0.021 for bone. Furthermore, deep MRAC provides good PET results, with average errors of less than 1% in most brain regions. Significantly lower PET reconstruction errors were realized with deep MRAC (-0.7% ± 1.1) compared with Dixon-based soft-tissue and air segmentation (-5.8% ± 3.1) and anatomic CT-based template registration (-4.8% ± 2.2). Conclusion The authors developed an automated approach that allows generation of discrete-valued pseudo CT scans (soft tissue, bone, and air) from a single high-spatial-resolution diagnostic-quality three-dimensional MR image and evaluated it in brain PET/MR imaging. This deep learning approach for MR imaging-based AC provided reduced PET reconstruction error relative to a CT-based standard within the brain compared

Deep-tissue two-photon imaging in brain and peripheral nerve with a compact high-pulse energy ytterbium fiber laser

NASA Astrophysics Data System (ADS)

Fontaine, Arjun K.; Kirchner, Matthew S.; Caldwell, John H.; Weir, Richard F.; Gibson, Emily A.

2018-02-01

Two-photon microscopy is a powerful tool of current scientific research, allowing optical visualization of structures below the surface of tissues. This is of particular value in neuroscience, where optically accessing regions within the brain is critical for the continued advancement in understanding of neural circuits. However, two-photon imaging at significant depths have typically used Ti:Sapphire based amplifiers that are prohibitively expensive and bulky. In this study, we demonstrate deep tissue two-photon imaging using a compact, inexpensive, turnkey operated Ytterbium fiber laser (Y-Fi, KM Labs). The laser is based on all-normal dispersion (ANDi) that provides short pulse durations and high pulse energies. Depth measurements obtained in ex vivo mouse cortex exceed those obtainable with standard two-photon microscopes using Ti:Sapphire lasers. In addition to demonstrating the capability of deep-tissue imaging in the brain, we investigated imaging depth in highly-scattering white matter with measurements in sciatic nerve showing limited optical penetration of heavily myelinated nerve tissue relative to grey matter.

High-Resolution Ultrasound-Switchable Fluorescence Imaging in Centimeter-Deep Tissue Phantoms with High Signal-To-Noise Ratio and High Sensitivity via Novel Contrast Agents

PubMed Central

Cheng, Bingbing; Bandi, Venugopal; Wei, Ming-Yuan; Pei, Yanbo; D’Souza, Francis; Nguyen, Kytai T.; Hong, Yi; Yuan, Baohong

2016-01-01

For many years, investigators have sought after high-resolution fluorescence imaging in centimeter-deep tissue because many interesting in vivo phenomena—such as the presence of immune system cells, tumor angiogenesis, and metastasis—may be located deep in tissue. Previously, we developed a new imaging technique to achieve high spatial resolution in sub-centimeter deep tissue phantoms named continuous-wave ultrasound-switchable fluorescence (CW-USF). The principle is to use a focused ultrasound wave to externally and locally switch on and off the fluorophore emission from a small volume (close to ultrasound focal volume). By making improvements in three aspects of this technique: excellent near-infrared USF contrast agents, a sensitive frequency-domain USF imaging system, and an effective signal processing algorithm, for the first time this study has achieved high spatial resolution (~ 900 μm) in 3-centimeter-deep tissue phantoms with high signal-to-noise ratio (SNR) and high sensitivity (3.4 picomoles of fluorophore in a volume of 68 nanoliters can be detected). We have achieved these results in both tissue-mimic phantoms and porcine muscle tissues. We have also demonstrated multi-color USF to image and distinguish two fluorophores with different wavelengths, which might be very useful for simultaneously imaging of multiple targets and observing their interactions in the future. This work has opened the door for future studies of high-resolution centimeter-deep tissue fluorescence imaging. PMID:27829050

Automated assessment of breast tissue density in non-contrast 3D CT images without image segmentation based on a deep CNN

NASA Astrophysics Data System (ADS)

Zhou, Xiangrong; Kano, Takuya; Koyasu, Hiromi; Li, Shuo; Zhou, Xinxin; Hara, Takeshi; Matsuo, Masayuki; Fujita, Hiroshi

2017-03-01

This paper describes a novel approach for the automatic assessment of breast density in non-contrast three-dimensional computed tomography (3D CT) images. The proposed approach trains and uses a deep convolutional neural network (CNN) from scratch to classify breast tissue density directly from CT images without segmenting the anatomical structures, which creates a bottleneck in conventional approaches. Our scheme determines breast density in a 3D breast region by decomposing the 3D region into several radial 2D-sections from the nipple, and measuring the distribution of breast tissue densities on each 2D section from different orientations. The whole scheme is designed as a compact network without the need for post-processing and provides high robustness and computational efficiency in clinical settings. We applied this scheme to a dataset of 463 non-contrast CT scans obtained from 30- to 45-year-old-women in Japan. The density of breast tissue in each CT scan was assigned to one of four categories (glandular tissue within the breast <25%, 25%-50%, 50%-75%, and >75%) by a radiologist as ground truth. We used 405 CT scans for training a deep CNN and the remaining 58 CT scans for testing the performance. The experimental results demonstrated that the findings of the proposed approach and those of the radiologist were the same in 72% of the CT scans among the training samples and 76% among the testing samples. These results demonstrate the potential use of deep CNN for assessing breast tissue density in non-contrast 3D CT images.

Deep learning based tissue analysis predicts outcome in colorectal cancer.

PubMed

Bychkov, Dmitrii; Linder, Nina; Turkki, Riku; Nordling, Stig; Kovanen, Panu E; Verrill, Clare; Walliander, Margarita; Lundin, Mikael; Haglund, Caj; Lundin, Johan

2018-02-21

Image-based machine learning and deep learning in particular has recently shown expert-level accuracy in medical image classification. In this study, we combine convolutional and recurrent architectures to train a deep network to predict colorectal cancer outcome based on images of tumour tissue samples. The novelty of our approach is that we directly predict patient outcome, without any intermediate tissue classification. We evaluate a set of digitized haematoxylin-eosin-stained tumour tissue microarray (TMA) samples from 420 colorectal cancer patients with clinicopathological and outcome data available. The results show that deep learning-based outcome prediction with only small tissue areas as input outperforms (hazard ratio 2.3; CI 95% 1.79-3.03; AUC 0.69) visual histological assessment performed by human experts on both TMA spot (HR 1.67; CI 95% 1.28-2.19; AUC 0.58) and whole-slide level (HR 1.65; CI 95% 1.30-2.15; AUC 0.57) in the stratification into low- and high-risk patients. Our results suggest that state-of-the-art deep learning techniques can extract more prognostic information from the tissue morphology of colorectal cancer than an experienced human observer.

Transmission in near-infrared optical windows for deep brain imaging.

PubMed

Shi, Lingyan; Sordillo, Laura A; Rodríguez-Contreras, Adrián; Alfano, Robert

2016-01-01

Near-infrared (NIR) radiation has been employed using one- and two-photon excitation of fluorescence imaging at wavelengths 650-950 nm (optical window I) for deep brain imaging; however, longer wavelengths in NIR have been overlooked due to a lack of suitable NIR-low band gap semiconductor imaging detectors and/or femtosecond laser sources. This research introduces three new optical windows in NIR and demonstrates their potential for deep brain tissue imaging. The transmittances are measured in rat brain tissue in the second (II, 1,100-1,350 nm), third (III, 1,600-1,870 nm), and fourth (IV, centered at 2,200 nm) NIR optical tissue windows. The relationship between transmission and tissue thickness is measured and compared with the theory. Due to a reduction in scattering and minimal absorption, window III is shown to be the best for deep brain imaging, and windows II and IV show similar but better potential for deep imaging than window I. © 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Versatile Polymer Nanoparticles as Two-Photon-Triggered Photosensitizers for Simultaneous Cellular, Deep-Tissue Imaging, and Photodynamic Therapy.

PubMed

Guo, Liang; Ge, Jiechao; Liu, Qian; Jia, Qingyan; Zhang, Hongyan; Liu, Weimin; Niu, Guangle; Liu, Sha; Gong, Jianru; Hackbarth, Steffen; Wang, Pengfei

2017-06-01

Clinical applications of current photodynamic therapy (PDT) photosensitizers (PSs) are often limited by their absorption in the UV-vis range that possesses limited tissue penetration ability, leading to ineffective therapeutic response for deep-seated tumors. Alternatively, two-photon excited PS (TPE-PS) using NIR light triggered is one the most promising candidates for PDT improvement. Herein, multimodal polymer nanoparticles (PNPs) from polythiophene derivative as two-photon fluorescence imaging as well as two-photon-excited PDT agent are developed. The prepared PNPs exhibit excellent water dispersibility, high photostability and pH stability, strong fluorescence brightness, and low dark toxicity. More importantly, the PNPs also possess other outstanding features including: (1) the high 1 O 2 quantum yield; (2) the strong two-photon-induced fluorescence and efficient 1 O 2 generation; (3) the specific accumulation in lysosomes of HeLa cells; and (4) the imaging detection depth up to 2100 µm in the mock tissue under two-photon. The multifunctional PNPs are promising candidates as TPE-PDT agent for simultaneous cellular, deep-tissue imaging, and highly efficient in vivo PDT of cancer. © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Breaking the acoustic diffraction limit via nonlinear effect and thermal confinement for potential deep-tissue high-resolution imaging

PubMed Central

Yuan, Baohong; Pei, Yanbo; Kandukuri, Jayanth

2013-01-01

Our recently developed ultrasound-switchable fluorescence (USF) imaging technique showed that it was feasible to conduct high-resolution fluorescence imaging in a centimeter-deep turbid medium. Because the spatial resolution of this technique highly depends on the ultrasound-induced temperature focal size (UTFS), minimization of UTFS becomes important for further improving the spatial resolution USF technique. In this study, we found that UTFS can be significantly reduced below the diffraction-limited acoustic intensity focal size via nonlinear acoustic effects and thermal confinement by appropriately controlling ultrasound power and exposure time, which can be potentially used for deep-tissue high-resolution imaging. PMID:23479498

Deep learning for tissue microarray image-based outcome prediction in patients with colorectal cancer

NASA Astrophysics Data System (ADS)

Bychkov, Dmitrii; Turkki, Riku; Haglund, Caj; Linder, Nina; Lundin, Johan

2016-03-01

Recent advances in computer vision enable increasingly accurate automated pattern classification. In the current study we evaluate whether a convolutional neural network (CNN) can be trained to predict disease outcome in patients with colorectal cancer based on images of tumor tissue microarray samples. We compare the prognostic accuracy of CNN features extracted from the whole, unsegmented tissue microarray spot image, with that of CNN features extracted from the epithelial and non-epithelial compartments, respectively. The prognostic accuracy of visually assessed histologic grade is used as a reference. The image data set consists of digitized hematoxylin-eosin (H and E) stained tissue microarray samples obtained from 180 patients with colorectal cancer. The patient samples represent a variety of histological grades, have data available on a series of clinicopathological variables including long-term outcome and ground truth annotations performed by experts. The CNN features extracted from images of the epithelial tissue compartment significantly predicted outcome (hazard ratio (HR) 2.08; CI95% 1.04-4.16; area under the curve (AUC) 0.66) in a test set of 60 patients, as compared to the CNN features extracted from unsegmented images (HR 1.67; CI95% 0.84-3.31, AUC 0.57) and visually assessed histologic grade (HR 1.96; CI95% 0.99-3.88, AUC 0.61). As a conclusion, a deep-learning classifier can be trained to predict outcome of colorectal cancer based on images of H and E stained tissue microarray samples and the CNN features extracted from the epithelial compartment only resulted in a prognostic discrimination comparable to that of visually determined histologic grade.

Deep Learning in Medical Image Analysis.

PubMed

Shen, Dinggang; Wu, Guorong; Suk, Heung-Il

2017-06-21

This review covers computer-assisted analysis of images in the field of medical imaging. Recent advances in machine learning, especially with regard to deep learning, are helping to identify, classify, and quantify patterns in medical images. At the core of these advances is the ability to exploit hierarchical feature representations learned solely from data, instead of features designed by hand according to domain-specific knowledge. Deep learning is rapidly becoming the state of the art, leading to enhanced performance in various medical applications. We introduce the fundamentals of deep learning methods and review their successes in image registration, detection of anatomical and cellular structures, tissue segmentation, computer-aided disease diagnosis and prognosis, and so on. We conclude by discussing research issues and suggesting future directions for further improvement.

Deep two-photon microscopic imaging through brain tissue using the second singlet state from fluorescent agent chlorophyll α in spinach leaf

NASA Astrophysics Data System (ADS)

Shi, Lingyan; Rodríguez-Contreras, Adrián; Budansky, Yury; Pu, Yang; An Nguyen, Thien; Alfano, Robert R.

2014-06-01

Two-photon (2P) excitation of the second singlet (S) state was studied to achieve deep optical microscopic imaging in brain tissue when both the excitation (800 nm) and emission (685 nm) wavelengths lie in the "tissue optical window" (650 to 950 nm). S2 state technique was used to investigate chlorophyll α (Chl α) fluorescence inside a spinach leaf under a thick layer of freshly sliced rat brain tissue in combination with 2P microscopic imaging. Strong emission at the peak wavelength of 685 nm under the 2P S state of Chl α enabled the imaging depth up to 450 μm through rat brain tissue.

Deep two-photon microscopic imaging through brain tissue using the second singlet state from fluorescent agent chlorophyll α in spinach leaf.

PubMed

Shi, Lingyan; Rodríguez-Contreras, Adrián; Budansky, Yury; Pu, Yang; Nguyen, Thien An; Alfano, Robert R

2014-06-01

Two-photon (2P) excitation of the second singlet (S₂) state was studied to achieve deep optical microscopic imaging in brain tissue when both the excitation (800 nm) and emission (685 nm) wavelengths lie in the "tissue optical window" (650 to 950 nm). S₂ state technique was used to investigate chlorophyll α (Chl α) fluorescence inside a spinach leaf under a thick layer of freshly sliced rat brain tissue in combination with 2P microscopic imaging. Strong emission at the peak wavelength of 685 nm under the 2P S₂ state of Chl α enabled the imaging depth up to 450 μm through rat brain tissue.

Deep Learning in Medical Image Analysis

PubMed Central

Shen, Dinggang; Wu, Guorong; Suk, Heung-Il

2016-01-01

The computer-assisted analysis for better interpreting images have been longstanding issues in the medical imaging field. On the image-understanding front, recent advances in machine learning, especially, in the way of deep learning, have made a big leap to help identify, classify, and quantify patterns in medical images. Specifically, exploiting hierarchical feature representations learned solely from data, instead of handcrafted features mostly designed based on domain-specific knowledge, lies at the core of the advances. In that way, deep learning is rapidly proving to be the state-of-the-art foundation, achieving enhanced performances in various medical applications. In this article, we introduce the fundamentals of deep learning methods; review their successes to image registration, anatomical/cell structures detection, tissue segmentation, computer-aided disease diagnosis or prognosis, and so on. We conclude by raising research issues and suggesting future directions for further improvements. PMID:28301734

Realistic tissue visualization using photoacoustic image

NASA Astrophysics Data System (ADS)

Cho, Seonghee; Managuli, Ravi; Jeon, Seungwan; Kim, Jeesu; Kim, Chulhong

2018-02-01

Visualization methods are very important in biomedical imaging. As a technology that understands life, biomedical imaging has the unique advantage of providing the most intuitive information in the image. This advantage of biomedical imaging can be greatly improved by choosing a special visualization method. This is more complicated in volumetric data. Volume data has the advantage of containing 3D spatial information. Unfortunately, the data itself cannot directly represent the potential value. Because images are always displayed in 2D space, visualization is the key and creates the real value of volume data. However, image processing of 3D data requires complicated algorithms for visualization and high computational burden. Therefore, specialized algorithms and computing optimization are important issues in volume data. Photoacoustic-imaging is a unique imaging modality that can visualize the optical properties of deep tissue. Because the color of the organism is mainly determined by its light absorbing component, photoacoustic data can provide color information of tissue, which is closer to real tissue color. In this research, we developed realistic tissue visualization using acoustic-resolution photoacoustic volume data. To achieve realistic visualization, we designed specialized color transfer function, which depends on the depth of the tissue from the skin. We used direct ray casting method and processed color during computing shader parameter. In the rendering results, we succeeded in obtaining similar texture results from photoacoustic data. The surface reflected rays were visualized in white, and the reflected color from the deep tissue was visualized red like skin tissue. We also implemented the CUDA algorithm in an OpenGL environment for real-time interactive imaging.

Spatial Statistics for Segmenting Histological Structures in H&E Stained Tissue Images.

PubMed

Nguyen, Luong; Tosun, Akif Burak; Fine, Jeffrey L; Lee, Adrian V; Taylor, D Lansing; Chennubhotla, S Chakra

2017-07-01

Segmenting a broad class of histological structures in transmitted light and/or fluorescence-based images is a prerequisite for determining the pathological basis of cancer, elucidating spatial interactions between histological structures in tumor microenvironments (e.g., tumor infiltrating lymphocytes), facilitating precision medicine studies with deep molecular profiling, and providing an exploratory tool for pathologists. This paper focuses on segmenting histological structures in hematoxylin- and eosin-stained images of breast tissues, e.g., invasive carcinoma, carcinoma in situ, atypical and normal ducts, adipose tissue, and lymphocytes. We propose two graph-theoretic segmentation methods based on local spatial color and nuclei neighborhood statistics. For benchmarking, we curated a data set of 232 high-power field breast tissue images together with expertly annotated ground truth. To accurately model the preference for histological structures (ducts, vessels, tumor nets, adipose, etc.) over the remaining connective tissue and non-tissue areas in ground truth annotations, we propose a new region-based score for evaluating segmentation algorithms. We demonstrate the improvement of our proposed methods over the state-of-the-art algorithms in both region- and boundary-based performance measures.

Self-recognition in corals facilitates deep-sea habitat engineering

USGS Publications Warehouse

Hennige, Sebastian J; Morrison, Cheryl L.; Form, Armin U.; Buscher, Janina; Kamenos, Nicholas A.; Roberts, J. Murray

2014-01-01

The ability of coral reefs to engineer complex three-dimensional habitats is central to their success and the rich biodiversity they support. In tropical reefs, encrusting coralline algae bind together substrates and dead coral framework to make continuous reef structures, but beyond the photic zone, the cold-water coral Lophelia pertusa also forms large biogenic reefs, facilitated by skeletal fusion. Skeletal fusion in tropical corals can occur in closely related or juvenile individuals as a result of non-aggressive skeletal overgrowth or allogeneic tissue fusion, but contact reactions in many species result in mortality if there is no ‘self-recognition’ on a broad species level. This study reveals areas of ‘flawless’ skeletal fusion in Lophelia pertusa, potentially facilitated by allogeneic tissue fusion, are identified as having small aragonitic crystals or low levels of crystal organisation, and strong molecular bonding. Regardless of the mechanism, the recognition of ‘self’ between adjacent L. pertusa colonies leads to no observable mortality, facilitates ecosystem engineering and reduces aggression-related energetic expenditure in an environment where energy conservation is crucial. The potential for self-recognition at a species level, and subsequent skeletal fusion in framework-forming cold-water corals is an important first step in understanding their significance as ecological engineers in deep-seas worldwide.

Predictive validity of granulation tissue color measured by digital image analysis for deep pressure ulcer healing: a multicenter prospective cohort study.

PubMed

Iizaka, Shinji; Kaitani, Toshiko; Sugama, Junko; Nakagami, Gojiro; Naito, Ayumi; Koyanagi, Hiroe; Konya, Chizuko; Sanada, Hiromi

2013-01-01

This multicenter prospective cohort study examined the predictive validity of granulation tissue color evaluated by digital image analysis for deep pressure ulcer healing. Ninety-one patients with deep pressure ulcers were followed for 3 weeks. From a wound photograph taken at baseline, an image representing the granulation red index (GRI) was processed in which a redder color represented higher values. We calculated the average GRI over granulation tissue and the proportion of pixels exceeding the threshold intensity of 80 for the granulation tissue surface (%GRI80) and wound surface (%wound red index 80). In the receiver operating characteristics curve analysis, most GRI parameters had adequate discriminative values for both improvement of the DESIGN-R total score and wound closure. Ulcers were categorized by the obtained cutoff points of the average GRI (≤80, >80), %GRI80 (≤55, >55-80, >80%), and %wound red index 80 (≤25, >25-50, >50%). In the linear mixed model, higher classes for all GRI parameters showed significantly greater relative improvement in overall wound severity during the 3 weeks after adjustment for patient characteristics and wound locations. Assessment of granulation tissue color by digital image analysis will be useful as an objective monitoring tool for granulation tissue quality or surrogate outcomes of pressure ulcer healing. © 2012 by the Wound Healing Society.

A simple model for deep tissue attenuation correction and large organ analysis of Cerenkov luminescence imaging

NASA Astrophysics Data System (ADS)

Habte, Frezghi; Natarajan, Arutselvan; Paik, David S.; Gambhir, Sanjiv S.

2014-03-01

Cerenkov luminescence imaging (CLI) is an emerging cost effective modality that uses conventional small animal optical imaging systems and clinically available radionuclide probes for light emission. CLI has shown good correlation with PET for organs of high uptake such as kidney, spleen, thymus and subcutaneous tumors in mouse models. However, CLI has limitations for deep tissue quantitative imaging since the blue-weighted spectral characteristics of Cerenkov radiation attenuates highly by mammalian tissue. Large organs such as the liver have also shown higher signal due to the contribution of emission of light from a greater thickness of tissue. In this study, we developed a simple model that estimates the effective tissue attenuation coefficient in order to correct the CLI signal intensity with a priori estimated depth and thickness of specific organs. We used several thin slices of ham to build a phantom with realistic attenuation. We placed radionuclide sources inside the phantom at different tissue depths and imaged it using an IVIS Spectrum (Perkin-Elmer, Waltham, MA, USA) and Inveon microPET (Preclinical Solutions Siemens, Knoxville, TN). We also performed CLI and PET of mouse models and applied the proposed attenuation model to correct CLI measurements. Using calibration factors obtained from phantom study that converts the corrected CLI measurements to %ID/g, we obtained an average difference of less that 10% for spleen and less than 35% for liver compared to conventional PET measurements. Hence, the proposed model has a capability of correcting the CLI signal to provide comparable measurements with PET data.

Adult stem cell lineage tracing and deep tissue imaging

PubMed Central

Fink, Juergen; Andersson-Rolf, Amanda; Koo, Bon-Kyoung

2015-01-01

Lineage tracing is a widely used method for understanding cellular dynamics in multicellular organisms during processes such as development, adult tissue maintenance, injury repair and tumorigenesis. Advances in tracing or tracking methods, from light microscopy-based live cell tracking to fluorescent label-tracing with two-photon microscopy, together with emerging tissue clearing strategies and intravital imaging approaches have enabled scientists to decipher adult stem and progenitor cell properties in various tissues and in a wide variety of biological processes. Although technical advances have enabled time-controlled genetic labeling and simultaneous live imaging, a number of obstacles still need to be overcome. In this review, we aim to provide an in-depth description of the traditional use of lineage tracing as well as current strategies and upcoming new methods of labeling and imaging. [BMB Reports 2015; 48(12): 655-667] PMID:26634741

An optical clearing technique for plant tissues allowing deep imaging and compatible with fluorescence microscopy.

PubMed

Warner, Cherish A; Biedrzycki, Meredith L; Jacobs, Samuel S; Wisser, Randall J; Caplan, Jeffrey L; Sherrier, D Janine

2014-12-01

We report on a nondestructive clearing technique that enhances transmission of light through specimens from diverse plant species, opening unique opportunities for microscope-enabled plant research. After clearing, plant organs and thick tissue sections are amenable to deep imaging. The clearing method is compatible with immunocytochemistry techniques and can be used in concert with common fluorescent probes, including widely adopted protein tags such as GFP, which has fluorescence that is preserved during the clearing process. © 2014 American Society of Plant Biologists. All Rights Reserved.

Adaptive focus for deep tissue using diffuse backscatter

NASA Astrophysics Data System (ADS)

Kress, Jeremy; Pourrezaei, Kambiz

2014-02-01

A system integrating high density diffuse optical imaging with adaptive optics using MEMS for deep tissue interaction is presented. In this system, a laser source is scanned over a high density fiber bundle using Digital Micromirror Device (DMD) and channeled to a tissue phantom. Backscatter is then collected from the tissue phantom by a high density fiber array of different fiber type and channeled to CMOS sensor for image acquisition. Intensity focus is directly verified using a second CMOS sensor which measures intensity transmitted though the tissue phantom. A set of training patterns are displayed on the DMD and backscatter is numerically fit to the transmission intensity. After the training patterns are displayed, adaptive focus is performed using only the backscatter and fitting functions. Additionally, tissue reconstruction and prediction of interference focusing by photoacoustic and optical tomographic methods is discussed. Finally, potential NIR applications such as in-vivo adaptive neural photostimulation and cancer targeting are discussed.

3D printed optical phantoms and deep tissue imaging for in vivo applications including oral surgery

NASA Astrophysics Data System (ADS)

Bentz, Brian Z.; Costas, Alfonso; Gaind, Vaibhav; Garcia, Jose M.; Webb, Kevin J.

2017-03-01

Progress in developing optical imaging for biomedical applications requires customizable and often complex objects known as "phantoms" for testing, evaluation, and calibration. This work demonstrates that 3D printing is an ideal method for fabricating such objects, allowing intricate inhomogeneities to be placed at exact locations in complex or anatomically realistic geometries, a process that is difficult or impossible using molds. We show printed mouse phantoms we have fabricated for developing deep tissue fluorescence imaging methods, and measurements of both their optical and mechanical properties. Additionally, we present a printed phantom of the human mouth that we use to develop an artery localization method to assist in oral surgery.

A metadata-aware application for remote scoring and exchange of tissue microarray images

PubMed Central

2013-01-01

Background The use of tissue microarrays (TMA) and advances in digital scanning microscopy has enabled the collection of thousands of tissue images. There is a need for software tools to annotate, query and share this data amongst researchers in different physical locations. Results We have developed an open source web-based application for remote scoring of TMA images, which exploits the value of Microsoft Silverlight Deep Zoom to provide a intuitive interface for zooming and panning around digital images. We use and extend existing XML-based standards to ensure that the data collected can be archived and that our system is interoperable with other standards-compliant systems. Conclusion The application has been used for multi-centre scoring of TMA slides composed of tissues from several Phase III breast cancer trials and ten different studies participating in the International Breast Cancer Association Consortium (BCAC). The system has enabled researchers to simultaneously score large collections of TMA and export the standardised data to integrate with pathological and clinical outcome data, thereby facilitating biomarker discovery. PMID:23635078

Deep tissue massage: What are we talking about?

PubMed

Koren, Yogev; Kalichman, Leonid

2018-04-01

Massage is a common treatment in complementary and integrative medicine. Deep tissue massage, a form of therapeutic massage, has become more and more popular in recent years. Hence, the use of massage generally and deep tissue massage specifically, should be evaluated as any other modality of therapy to establish its efficacy and safety. To determine the definitions used for deep tissue massage in the scientific literature and to review the current scientific evidence for its efficacy and safety. Narrative review. There is no commonly accepted definition of deep tissue massage in the literature. The definition most frequently used is the intention of the therapist. We suggest separating the definitions of deep massage and deep tissue massage as follows: deep massage should be used to describe the intention of the therapist to treat deep tissue by using any form of massage and deep tissue massage should be used to describe a specific and independent method of massage therapy, utilizing the specific set of principles and techniques as defined by Riggs: "The understanding of the layers of the body, and the ability to work with tissue in these layers to relax, lengthen, and release holding patterns in the most effective and energy efficient way possible within the client's parameters of comfort". Heterogeneity of techniques and protocols used in published studies have made it difficult to draw any clear conclusions. Favorable outcomes may result from deep tissue massage in pain populations and patients with decreased range of motion. In addition, several rare serious adverse events were found related to deep tissue massage, probably as a result of the forceful application of massage therapy. Future research of deep tissue massage should be based on a common definition, classification system and the use of common comparators as controls. Copyright © 2017 Elsevier Ltd. All rights reserved.

Technical Note: Deep learning based MRAC using rapid ultra-short echo time imaging.

PubMed

Jang, Hyungseok; Liu, Fang; Zhao, Gengyan; Bradshaw, Tyler; McMillan, Alan B

2018-05-15

In this study, we explore the feasibility of a novel framework for MR-based attenuation correction for PET/MR imaging based on deep learning via convolutional neural networks, which enables fully automated and robust estimation of a pseudo CT image based on ultrashort echo time (UTE), fat, and water images obtained by a rapid MR acquisition. MR images for MRAC are acquired using dual echo ramped hybrid encoding (dRHE), where both UTE and out-of-phase echo images are obtained within a short single acquisition (35 sec). Tissue labeling of air, soft tissue, and bone in the UTE image is accomplished via a deep learning network that was pre-trained with T1-weighted MR images. UTE images are used as input to the network, which was trained using labels derived from co-registered CT images. The tissue labels estimated by deep learning are refined by a conditional random field based correction. The soft tissue labels are further separated into fat and water components using the two-point Dixon method. The estimated bone, air, fat, and water images are then assigned appropriate Hounsfield units, resulting in a pseudo CT image for PET attenuation correction. To evaluate the proposed MRAC method, PET/MR imaging of the head was performed on 8 human subjects, where Dice similarity coefficients of the estimated tissue labels and relative PET errors were evaluated through comparison to a registered CT image. Dice coefficients for air (within the head), soft tissue, and bone labels were 0.76±0.03, 0.96±0.006, and 0.88±0.01. In PET quantification, the proposed MRAC method produced relative PET errors less than 1% within most brain regions. The proposed MRAC method utilizing deep learning with transfer learning and an efficient dRHE acquisition enables reliable PET quantification with accurate and rapid pseudo CT generation. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.

Deep convolutional neural network and 3D deformable approach for tissue segmentation in musculoskeletal magnetic resonance imaging.

PubMed

Liu, Fang; Zhou, Zhaoye; Jang, Hyungseok; Samsonov, Alexey; Zhao, Gengyan; Kijowski, Richard

2018-04-01

To describe and evaluate a new fully automated musculoskeletal tissue segmentation method using deep convolutional neural network (CNN) and three-dimensional (3D) simplex deformable modeling to improve the accuracy and efficiency of cartilage and bone segmentation within the knee joint. A fully automated segmentation pipeline was built by combining a semantic segmentation CNN and 3D simplex deformable modeling. A CNN technique called SegNet was applied as the core of the segmentation method to perform high resolution pixel-wise multi-class tissue classification. The 3D simplex deformable modeling refined the output from SegNet to preserve the overall shape and maintain a desirable smooth surface for musculoskeletal structure. The fully automated segmentation method was tested using a publicly available knee image data set to compare with currently used state-of-the-art segmentation methods. The fully automated method was also evaluated on two different data sets, which include morphological and quantitative MR images with different tissue contrasts. The proposed fully automated segmentation method provided good segmentation performance with segmentation accuracy superior to most of state-of-the-art methods in the publicly available knee image data set. The method also demonstrated versatile segmentation performance on both morphological and quantitative musculoskeletal MR images with different tissue contrasts and spatial resolutions. The study demonstrates that the combined CNN and 3D deformable modeling approach is useful for performing rapid and accurate cartilage and bone segmentation within the knee joint. The CNN has promising potential applications in musculoskeletal imaging. Magn Reson Med 79:2379-2391, 2018. © 2017 International Society for Magnetic Resonance in Medicine. © 2017 International Society for Magnetic Resonance in Medicine.

Automatical and accurate segmentation of cerebral tissues in fMRI dataset with combination of image processing and deep learning

NASA Astrophysics Data System (ADS)

Kong, Zhenglun; Luo, Junyi; Xu, Shengpu; Li, Ting

2018-02-01

Image segmentation plays an important role in medical science. One application is multimodality imaging, especially the fusion of structural imaging with functional imaging, which includes CT, MRI and new types of imaging technology such as optical imaging to obtain functional images. The fusion process require precisely extracted structural information, in order to register the image to it. Here we used image enhancement, morphometry methods to extract the accurate contours of different tissues such as skull, cerebrospinal fluid (CSF), grey matter (GM) and white matter (WM) on 5 fMRI head image datasets. Then we utilized convolutional neural network to realize automatic segmentation of images in deep learning way. Such approach greatly reduced the processing time compared to manual and semi-automatic segmentation and is of great importance in improving speed and accuracy as more and more samples being learned. The contours of the borders of different tissues on all images were accurately extracted and 3D visualized. This can be used in low-level light therapy and optical simulation software such as MCVM. We obtained a precise three-dimensional distribution of brain, which offered doctors and researchers quantitative volume data and detailed morphological characterization for personal precise medicine of Cerebral atrophy/expansion. We hope this technique can bring convenience to visualization medical and personalized medicine.

Hyperspectral imaging in SWIR: from stain-free microscopy to deep tissue imaging (Conference Presentation)

NASA Astrophysics Data System (ADS)

Berezin, Mikhail Y.

2016-03-01

Recent advances in relatively unexplored short wave infrared (SWIR) range from 800-1600 nm detectors make wide-field imaging in this spectral range attractive to biology. The distinct advantages of SWIR region over the visible and near infrared (NIR) in tissue analysis are two-fold: (i) high abundance endogenous chromophores (i.e. water and lipids) enable tissue component differentiation based on wavelength-dependent absorption properties and (ii) the weak scattering of tissue permits better resolution of imaging in thick specimens. When combined with high spectral resolution, SWIR imaging produces a spectroscopic image, where every pixel corresponds to the entire high-resolution spectrum. This hyperspectral (HS) approach provides rich information about the relative abundance of individual chromophores and their interactions that contribute to the intensity and location of the optical signal. The presentation discusses the challenges in the SWIR-HS instrument design and data analysis and demonstrates some of the promising applications of this technology in life science and medicine.

Hyperspectral Image Enhancement and Mixture Deep-Learning Classification of Corneal Epithelium Injuries

PubMed Central

Md Noor, Siti Salwa; Michael, Kaleena; Marshall, Stephen; Ren, Jinchang

2017-01-01

In our preliminary study, the reflectance signatures obtained from hyperspectral imaging (HSI) of normal and abnormal corneal epithelium tissues of porcine show similar morphology with subtle differences. Here we present image enhancement algorithms that can be used to improve the interpretability of data into clinically relevant information to facilitate diagnostics. A total of 25 corneal epithelium images without the application of eye staining were used. Three image feature extraction approaches were applied for image classification: (i) image feature classification from histogram using a support vector machine with a Gaussian radial basis function (SVM-GRBF); (ii) physical image feature classification using deep-learning Convolutional Neural Networks (CNNs) only; and (iii) the combined classification of CNNs and SVM-Linear. The performance results indicate that our chosen image features from the histogram and length-scale parameter were able to classify with up to 100% accuracy; particularly, at CNNs and CNNs-SVM, by employing 80% of the data sample for training and 20% for testing. Thus, in the assessment of corneal epithelium injuries, HSI has high potential as a method that could surpass current technologies regarding speed, objectivity, and reliability. PMID:29144388

Hyperspectral Image Enhancement and Mixture Deep-Learning Classification of Corneal Epithelium Injuries.

PubMed

Noor, Siti Salwa Md; Michael, Kaleena; Marshall, Stephen; Ren, Jinchang

2017-11-16

In our preliminary study, the reflectance signatures obtained from hyperspectral imaging (HSI) of normal and abnormal corneal epithelium tissues of porcine show similar morphology with subtle differences. Here we present image enhancement algorithms that can be used to improve the interpretability of data into clinically relevant information to facilitate diagnostics. A total of 25 corneal epithelium images without the application of eye staining were used. Three image feature extraction approaches were applied for image classification: (i) image feature classification from histogram using a support vector machine with a Gaussian radial basis function (SVM-GRBF); (ii) physical image feature classification using deep-learning Convolutional Neural Networks (CNNs) only; and (iii) the combined classification of CNNs and SVM-Linear. The performance results indicate that our chosen image features from the histogram and length-scale parameter were able to classify with up to 100% accuracy; particularly, at CNNs and CNNs-SVM, by employing 80% of the data sample for training and 20% for testing. Thus, in the assessment of corneal epithelium injuries, HSI has high potential as a method that could surpass current technologies regarding speed, objectivity, and reliability.

Clutter elimination for deep clinical optoacoustic imaging using localised vibration tagging (LOVIT)☆

PubMed Central

Jaeger, Michael; Bamber, Jeffrey C.; Frenz, Martin

2013-01-01

This paper investigates a novel method which allows clutter elimination in deep optoacoustic imaging. Clutter significantly limits imaging depth in clinical optoacoustic imaging, when irradiation optics and ultrasound detector are integrated in a handheld probe for flexible imaging of the human body. Strong optoacoustic transients generated at the irradiation site obscure weak signals from deep inside the tissue, either directly by propagating towards the probe, or via acoustic scattering. In this study we demonstrate that signals of interest can be distinguished from clutter by tagging them at the place of origin with localised tissue vibration induced by the acoustic radiation force in a focused ultrasonic beam. We show phantom results where this technique allowed almost full clutter elimination and thus strongly improved contrast for deep imaging. Localised vibration tagging by means of acoustic radiation force is especially promising for integration into ultrasound systems that already have implemented radiation force elastography. PMID:25302147

Three-Photon Luminescence of Gold Nanorods and Its Applications for High Contrast Tissue and Deep In Vivo Brain Imaging

PubMed Central

Wang, Shaowei; Xi, Wang; Cai, Fuhong; Zhao, Xinyuan; Xu, Zhengping; Qian, Jun; He, Sailing

2015-01-01

Gold nanoparticles can be used as contrast agents for bio-imaging applications. Here we studied multi-photon luminescence (MPL) of gold nanorods (GNRs), under the excitation of femtosecond (fs) lasers. GNRs functionalized with polyethylene glycol (PEG) molecules have high chemical and optical stability, and can be used as multi-photon luminescent nanoprobes for deep in vivo imaging of live animals. We have found that the depth of in vivo imaging is dependent upon the transmission and focal capability of the excitation light interacting with the GNRs. Our study focused on the comparison of MPL from GNRs with two different aspect ratios, as well as their ex vivo and in vivo imaging effects under 760 nm and 1000 nm excitation, respectively. Both of these wavelengths were located at an optically transparent window of biological tissue (700-1000 nm). PEGylated GNRs, which were intravenously injected into mice via the tail vein and accumulated in major organs and tumor tissue, showed high image contrast due to distinct three-photon luminescence (3PL) signals upon irradiation of a 1000 nm fs laser. Concerning in vivo mouse brain imaging, the 3PL imaging depth of GNRs under 1000 nm fs excitation could reach 600 μm, which was approximately 170 μm deeper than the two-photon luminescence (2PL) imaging depth of GNRs with a fs excitation of 760 nm. PMID:25553113

Three-photon tissue imaging using moxifloxacin.

PubMed

Lee, Seunghun; Lee, Jun Ho; Wang, Taejun; Jang, Won Hyuk; Yoon, Yeoreum; Kim, Bumju; Jun, Yong Woong; Kim, Myoung Joon; Kim, Ki Hean

2018-06-20

Moxifloxacin is an antibiotic used in clinics and has recently been used as a clinically compatible cell-labeling agent for two-photon (2P) imaging. Although 2P imaging with moxifloxacin labeling visualized cells inside tissues using enhanced fluorescence, the imaging depth was quite limited because of the relatively short excitation wavelength (<800 nm) used. In this study, the feasibility of three-photon (3P) excitation of moxifloxacin using a longer excitation wavelength and moxifloxacin-based 3P imaging were tested to increase the imaging depth. Moxifloxacin fluorescence via 3P excitation was detected at a >1000 nm excitation wavelength. After obtaining the excitation and emission spectra of moxifloxacin, moxifloxacin-based 3P imaging was applied to ex vivo mouse bladder and ex vivo mouse small intestine tissues and compared with moxifloxacin-based 2P imaging by switching the excitation wavelength of a Ti:sapphire oscillator between near 1030 and 780 nm. Both moxifloxacin-based 2P and 3P imaging visualized cellular structures in the tissues via moxifloxacin labeling, but the image contrast was better with 3P imaging than with 2P imaging at the same imaging depths. The imaging speed and imaging depth of moxifloxacin-based 3P imaging using a Ti:sapphire oscillator were limited by insufficient excitation power. Therefore, we constructed a new system for moxifloxacin-based 3P imaging using a high-energy Yb fiber laser at 1030 nm and used it for in vivo deep tissue imaging of a mouse small intestine. Moxifloxacin-based 3P imaging could be useful for clinical applications with enhanced imaging depth.

A Non-Invasive Deep Tissue PH Monitor.

DTIC Science & Technology

1995-08-11

disturbances in acid-base regulation may have serious effects on metabolic activity, circulation, and the central nervous system. Currently, acid-base...to tissue ischemia than is arterial pH. Consequently, a non-invasive deep tissue pH monitor has enormous value as a mechanism for rapid and effective ...achieved, and improve our understanding of what physical effects are important to successful non-invasive deep tissue pH monitoring. This last statement

Deep Convolutional Neural Networks for Multi-Modality Isointense Infant Brain Image Segmentation

PubMed Central

Zhang, Wenlu; Li, Rongjian; Deng, Houtao; Wang, Li; Lin, Weili; Ji, Shuiwang; Shen, Dinggang

2015-01-01

The segmentation of infant brain tissue images into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) plays an important role in studying early brain development in health and disease. In the isointense stage (approximately 6–8 months of age), WM and GM exhibit similar levels of intensity in both T1 and T2 MR images, making the tissue segmentation very challenging. Only a small number of existing methods have been designed for tissue segmentation in this isointense stage; however, they only used a single T1 or T2 images, or the combination of T1 and T2 images. In this paper, we propose to use deep convolutional neural networks (CNNs) for segmenting isointense stage brain tissues using multi-modality MR images. CNNs are a type of deep models in which trainable filters and local neighborhood pooling operations are applied alternatingly on the raw input images, resulting in a hierarchy of increasingly complex features. Specifically, we used multimodality information from T1, T2, and fractional anisotropy (FA) images as inputs and then generated the segmentation maps as outputs. The multiple intermediate layers applied convolution, pooling, normalization, and other operations to capture the highly nonlinear mappings between inputs and outputs. We compared the performance of our approach with that of the commonly used segmentation methods on a set of manually segmented isointense stage brain images. Results showed that our proposed model significantly outperformed prior methods on infant brain tissue segmentation. In addition, our results indicated that integration of multi-modality images led to significant performance improvement. PMID:25562829

Bioluminescence resonance energy transfer (BRET) imaging of protein–protein interactions within deep tissues of living subjects

PubMed Central

Dragulescu-Andrasi, Anca; Chan, Carmel T.; Massoud, Tarik F.; Gambhir, Sanjiv S.

2011-01-01

Identifying protein–protein interactions (PPIs) is essential for understanding various disease mechanisms and developing new therapeutic approaches. Current methods for assaying cellular intermolecular interactions are mainly used for cells in culture and have limited use for the noninvasive assessment of small animal disease models. Here, we describe red light-emitting reporter systems based on bioluminescence resonance energy transfer (BRET) that allow for assaying PPIs both in cell culture and deep tissues of small animals. These BRET systems consist of the recently developed Renilla reniformis luciferase (RLuc) variants RLuc8 and RLuc8.6, used as BRET donors, combined with two red fluorescent proteins, TagRFP and TurboFP635, as BRET acceptors. In addition to the native coelenterazine luciferase substrate, we used the synthetic derivative coelenterazine-v, which further red-shifts the emission maxima of Renilla luciferases by 35 nm. We show the use of these BRET systems for ratiometric imaging of both cells in culture and deep-tissue small animal tumor models and validate their applicability for studying PPIs in mice in the context of rapamycin-induced FK506 binding protein 12 (FKBP12)-FKBP12 rapamycin binding domain (FRB) association. These red light-emitting BRET systems have great potential for investigating PPIs in the context of drug screening and target validation applications. PMID:21730157

Deep learning based classification for head and neck cancer detection with hyperspectral imaging in an animal model

NASA Astrophysics Data System (ADS)

Ma, Ling; Lu, Guolan; Wang, Dongsheng; Wang, Xu; Chen, Zhuo Georgia; Muller, Susan; Chen, Amy; Fei, Baowei

2017-03-01

Hyperspectral imaging (HSI) is an emerging imaging modality that can provide a noninvasive tool for cancer detection and image-guided surgery. HSI acquires high-resolution images at hundreds of spectral bands, providing big data to differentiating different types of tissue. We proposed a deep learning based method for the detection of head and neck cancer with hyperspectral images. Since the deep learning algorithm can learn the feature hierarchically, the learned features are more discriminative and concise than the handcrafted features. In this study, we adopt convolutional neural networks (CNN) to learn the deep feature of pixels for classifying each pixel into tumor or normal tissue. We evaluated our proposed classification method on the dataset containing hyperspectral images from 12 tumor-bearing mice. Experimental results show that our method achieved an average accuracy of 91.36%. The preliminary study demonstrated that our deep learning method can be applied to hyperspectral images for detecting head and neck tumors in animal models.

Cellular imaging of deep organ using two-photon Bessel light-sheet nonlinear structured illumination microscopy

PubMed Central

Zhao, Ming; Zhang, Han; Li, Yu; Ashok, Amit; Liang, Rongguang; Zhou, Weibin; Peng, Leilei

2014-01-01

In vivo fluorescent cellular imaging of deep internal organs is highly challenging, because the excitation needs to penetrate through strong scattering tissue and the emission signal is degraded significantly by photon diffusion induced by tissue-scattering. We report that by combining two-photon Bessel light-sheet microscopy with nonlinear structured illumination microscopy (SIM), live samples up to 600 microns wide can be imaged by light-sheet microscopy with 500 microns penetration depth, and diffused background in deep tissue light-sheet imaging can be reduced to obtain clear images at cellular resolution in depth beyond 200 microns. We demonstrate in vivo two-color imaging of pronephric glomeruli and vasculature of zebrafish kidney, whose cellular structures located at the center of the fish body are revealed in high clarity by two-color two-photon Bessel light-sheet SIM. PMID:24876996

A bright cyan-excitable orange fluorescent protein facilitates dual-emission microscopy and enhances bioluminescence imaging in vivo

PubMed Central

Chu, Jun; Oh, Young-Hee; Sens, Alex; Ataie, Niloufar; Dana, Hod; Macklin, John J.; Laviv, Tal; Welf, Erik S.; Dean, Kevin M.; Zhang, Feijie; Kim, Benjamin B.; Tang, Clement Tran; Hu, Michelle; Baird, Michelle A.; Davidson, Michael W.; Kay, Mark A.; Fiolka, Reto; Yasuda, Ryohei; Kim, Douglas S.; Ng, Ho-Leung; Lin, Michael Z.

2016-01-01

Orange-red fluorescent proteins (FPs) are widely used in biomedical research for multiplexed epifluorescence microscopy with GFP-based probes, but their different excitation requirements make multiplexing with new advanced microscopy methods difficult. Separately, orange-red FPs are useful for deep-tissue imaging in mammals due to the relative tissue transmissibility of orange-red light, but their dependence on illumination limits their sensitivity as reporters in deep tissues. Here we describe CyOFP1, a bright engineered orange-red FP that is excitable by cyan light. We show that CyOFP1 enables single-excitation multiplexed imaging with GFP-based probes in single-photon and two-photon microscopy, including time-lapse imaging in light-sheet systems. CyOFP1 also serves as an efficient acceptor for resonance energy transfer from the highly catalytic blue-emitting luciferase NanoLuc. An optimized fusion of CyOFP1 and NanoLuc, called Antares, functions as a highly sensitive bioluminescent reporter in vivo, producing substantially brighter signals from deep tissues than firefly luciferase and other bioluminescent proteins. PMID:27240196

Spectroscopic Imaging of Deep Tissue through Photoacoustic Detection of Molecular Vibration

PubMed Central

Wang, Pu; Rajian, Justin R.; Cheng, Ji-Xin

2013-01-01

The quantized vibration of chemical bonds provides a way of imaging target molecules in a complex tissue environment. Photoacoustic detection of harmonic vibrational transitions provides an approach to visualize tissue content beyond the ballistic photon regime. This method involves pulsed laser excitation of overtone transitions in target molecules inside a tissue. Fast relaxation of the vibrational energy into heat results in a local temperature rise on the order of mK and a subsequent generation of acoustic waves detectable with an ultrasonic transducer. In this perspective, we review recent advances that demonstrate the advantages of vibration-based photoacoustic imaging and illustrate its potential in diagnosing cardiovascular plaques. An outlook into future development of vibrational photoacoustic endoscopy and tomography is provided. PMID:24073304

Bond-selective imaging of deep tissue through the optical window between 1600 and 1850 nm.

PubMed

Wang, Pu; Wang, Han-Wei; Sturek, Michael; Cheng, Ji-Xin

2012-01-01

We report the employment of an optical window between 1600 nm and 1850 nm for bond-selective deep tissue imaging through harmonic vibrational excitation and acoustic detection of resultant pressure waves. In this window where a local minimum of water absorption resides, we found a 5 times enhancement of photoacoustic signal by first overtone excitation of the methylene group CH(2) at 1730 nm, compared to the second overtone excitation at 1210 nm. The enhancement allows 3D mapping of intramuscular fat with improved contrast and of lipid deposition inside an atherosclerotic artery wall in the presence of blood. Moreover, lipid and protein are differentiated based on the first overtone absorption profiles of CH(2) and methyl group CH(3) in this window. Copyright © 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Photoacoustic diagnosis of burns in rats: two-dimensional photo-acoustic imaging of burned tissue

NASA Astrophysics Data System (ADS)

Yamazaki, Mutsuo; Sato, Shunichi; Saito, Daizo; Okada, Yoshiaki; Kurita, Akira; Kikuchi, Makoto; Ashida, Hiroshi; Obara, Minoru

2003-06-01

We previously reported that for rat burn models, deep dermal burns and deep burns can be well differentiated by measuring the propagation time of the photoacoustic signals originated from the blood in the healthy skin tissue under the damaged tissue layer. However, the diagnosis was based on point measurement in the wound, and therefore site-dependent information on the injuries was not obtained; such information is very important for diagnosis of extended burns. In the present study, we scanned a photoacoustic detector on the wound and constructed two-dimensional (2-D) images of the blood-originated photoacoustic signals for superficial dermal burns (SDB), deep dermal burns (DDB), deep burns (DB), and healthy skins (control) in rats. For each burn model, site-dependent variation of the signal was observed; the variation probably reflects the distribution of blood vessels in the skin tissue. In spite of the variation, clear differentiation was obtained between SDB, DDB, and DB from the 2D images. The images were constructed as a function of post burn time. Temporal signal variation will be also presented.

Three-Dimensional Imaging of the Intracellular Fate of Plasmid DNA and Transgene Expression: ZsGreen1 and Tissue Clearing Method CUBIC Are an Optimal Combination for Multicolor Deep Imaging in Murine Tissues.

PubMed

Fumoto, Shintaro; Nishimura, Koyo; Nishida, Koyo; Kawakami, Shigeru

2016-01-01

Evaluation methods for determining the distribution of transgene expression in the body and the in vivo fate of viral and non-viral vectors are necessary for successful development of in vivo gene delivery systems. Here, we evaluated the spatial distribution of transgene expression using tissue clearing methods. After hydrodynamic injection of plasmid DNA into mice, whole tissues were subjected to tissue clearing. Tissue clearing followed by confocal laser scanning microscopy enabled evaluation of the three-dimensional distribution of transgene expression without preparation of tissue sections. Among the tested clearing methods (ClearT2, SeeDB, and CUBIC), CUBIC was the most suitable method for determining the spatial distribution of transgene expression in not only the liver but also other tissues such as the kidney and lung. In terms of the type of fluorescent protein, the observable depth for green fluorescent protein ZsGreen1 was slightly greater than that for red fluorescent protein tdTomato. We observed a depth of ~1.5 mm for the liver and 500 μm for other tissues without preparation of tissue sections. Furthermore, we succeeded in multicolor deep imaging of the intracellular fate of plasmid DNA in the murine liver. Thus, tissue clearing would be a powerful approach for determining the spatial distribution of plasmid DNA and transgene expression in various murine tissues.

Photoacoustic physio-chemical analysis and its implementation in deep tissue with a catheter setup

NASA Astrophysics Data System (ADS)

Xu, Guan; Meng, Zhou-xian; Lin, Jian-die D.; Cheng, Qian; Wang, Xueding

2015-03-01

Photoacoustic (PA) measurements encode the information associated with both physical microstructures and chemical contents in biological tissues. A two-dimensional physio-chemical spectrogram (PCS) can be formulated by combining the power spectra of PA signals acquired at a series of optical wavelengths. The analysis of PCS, or namely PA physio-chemical analysis (PAPCA), enables the quantification of the relative concentrations and the spatial distributions of a variety of chemical components in the tissue. This study validated the feasibility of PAPCA in characterizing liver conditions during the progression of non-alcoholic fatty liver disease. A catheter based setup facilitating measurement in deep tissues was also tested.

Fiber-based tunable repetition rate source for deep tissue two-photon fluorescence microscopy.

PubMed

Charan, Kriti; Li, Bo; Wang, Mengran; Lin, Charles P; Xu, Chris

2018-05-01

Deep tissue multiphoton imaging requires high peak power to enhance signal and low average power to prevent thermal damage. Both goals can be advantageously achieved through laser repetition rate tuning instead of simply adjusting the average power. We show that the ideal repetition rate for deep two-photon imaging in the mouse brain is between 1 and 10 MHz, and we present a fiber-based source with an arbitrarily tunable repetition rate within this range. The performance of the new source is compared to a mode-locked Ti:Sapphire (Ti:S) laser for in vivo imaging of mouse brain vasculature. At 2.5 MHz, the fiber source requires 5.1 times less average power to obtain the same signal as a standard Ti:S laser operating at 80 MHz.

Fiber-based tunable repetition rate source for deep tissue two-photon fluorescence microscopy

PubMed Central

Charan, Kriti; Li, Bo; Wang, Mengran; Lin, Charles P.; Xu, Chris

2018-01-01

Deep tissue multiphoton imaging requires high peak power to enhance signal and low average power to prevent thermal damage. Both goals can be advantageously achieved through laser repetition rate tuning instead of simply adjusting the average power. We show that the ideal repetition rate for deep two-photon imaging in the mouse brain is between 1 and 10 MHz, and we present a fiber-based source with an arbitrarily tunable repetition rate within this range. The performance of the new source is compared to a mode-locked Ti:Sapphire (Ti:S) laser for in vivo imaging of mouse brain vasculature. At 2.5 MHz, the fiber source requires 5.1 times less average power to obtain the same signal as a standard Ti:S laser operating at 80 MHz. PMID:29760989

Imaging deep skeletal muscle structure using a high-sensitivity ultrathin side-viewing optical coherence tomography needle probe

PubMed Central

Yang, Xiaojie; Lorenser, Dirk; McLaughlin, Robert A.; Kirk, Rodney W.; Edmond, Matthew; Simpson, M. Cather; Grounds, Miranda D.; Sampson, David D.

2013-01-01

We have developed an extremely miniaturized optical coherence tomography (OCT) needle probe (outer diameter 310 µm) with high sensitivity (108 dB) to enable minimally invasive imaging of cellular structure deep within skeletal muscle. Three-dimensional volumetric images were acquired from ex vivo mouse tissue, examining both healthy and pathological dystrophic muscle. Individual myofibers were visualized as striations in the images. Degradation of cellular structure in necrotic regions was seen as a loss of these striations. Tendon and connective tissue were also visualized. The observed structures were validated against co-registered hematoxylin and eosin (H&E) histology sections. These images of internal cellular structure of skeletal muscle acquired with an OCT needle probe demonstrate the potential of this technique to visualize structure at the microscopic level deep in biological tissue in situ. PMID:24466482

Characterisation of signal enhancements achieved when utilizing a photon diode in deep Raman spectroscopy of tissue

PubMed Central

Vardaki, Martha Z.; Matousek, Pavel; Stone, Nicholas

2016-01-01

We characterise the performance of a beam enhancing element (‘photon diode’) for use in deep Raman spectroscopy (DRS) of biological tissues. The optical component enhances the number of laser photons coupled into a tissue sample by returning escaping photons back into it at the illumination zone. The method is compatible with transmission Raman spectroscopy, a deep Raman spectroscopy concept, and its implementation leads to considerable enhancement of detected Raman photon rates. In the past, the enhancement concept was demonstrated with a variety of samples (pharmaceutical tablets, tissue, etc) but it was not systematically characterized with biological tissues. In this study, we investigate the enhancing properties of the photon diode in the transmission Raman geometry as a function of: a) the depth and b) the optical properties of tissue samples. Liquid tissue phantoms were employed to facilitate systematic variation of optical properties. These were chosen to mimic optical properties of human tissues, including breast and prostate. The obtained results evidence that a photon diode can enhance Raman signals of tissues by a maximum of × 2.4, although it can also decrease the signals created towards the back of samples that exhibit high scattering or absorption properties. PMID:27375932

Fiber-optic fluorescence imaging

PubMed Central

Flusberg, Benjamin A; Cocker, Eric D; Piyawattanametha, Wibool; Jung, Juergen C; Cheung, Eunice L M; Schnitzer, Mark J

2010-01-01

Optical fibers guide light between separate locations and enable new types of fluorescence imaging. Fiber-optic fluorescence imaging systems include portable handheld microscopes, flexible endoscopes well suited for imaging within hollow tissue cavities and microendoscopes that allow minimally invasive high-resolution imaging deep within tissue. A challenge in the creation of such devices is the design and integration of miniaturized optical and mechanical components. Until recently, fiber-based fluorescence imaging was mainly limited to epifluorescence and scanning confocal modalities. Two new classes of photonic crystal fiber facilitate ultrashort pulse delivery for fiber-optic two-photon fluorescence imaging. An upcoming generation of fluorescence imaging devices will be based on microfabricated device components. PMID:16299479

Relevance of deep learning to facilitate the diagnosis of HER2 status in breast cancer

NASA Astrophysics Data System (ADS)

Vandenberghe, Michel E.; Scott, Marietta L. J.; Scorer, Paul W.; Söderberg, Magnus; Balcerzak, Denis; Barker, Craig

2017-04-01

Tissue biomarker scoring by pathologists is central to defining the appropriate therapy for patients with cancer. Yet, inter-pathologist variability in the interpretation of ambiguous cases can affect diagnostic accuracy. Modern artificial intelligence methods such as deep learning have the potential to supplement pathologist expertise to ensure constant diagnostic accuracy. We developed a computational approach based on deep learning that automatically scores HER2, a biomarker that defines patient eligibility for anti-HER2 targeted therapies in breast cancer. In a cohort of 71 breast tumour resection samples, automated scoring showed a concordance of 83% with a pathologist. The twelve discordant cases were then independently reviewed, leading to a modification of diagnosis from initial pathologist assessment for eight cases. Diagnostic discordance was found to be largely caused by perceptual differences in assessing HER2 expression due to high HER2 staining heterogeneity. This study provides evidence that deep learning aided diagnosis can facilitate clinical decision making in breast cancer by identifying cases at high risk of misdiagnosis.

In vivo tomographic imaging of deep seated cancer using fluorescence lifetime contrast

PubMed Central

Rice, William L.; Shcherbakova, Daria M; Verkusha, Vladislav V.; Kumar, Anand T.N.

2015-01-01

Preclinical cancer research would benefit from non-invasive imaging methods that allow tracking and visualization of early stage metastasis in vivo. While fluorescent proteins revolutionized intravital microscopy, two major challenges which still remain are tissue autofluorescence and hemoglobin absorption, which act to limit intravital optical techniques to large or subcutaneous tumors. Here we employ time-domain technology for the effective separation of tissue autofluorescence from extrinsic fluorophores, based on their distinct fluorescence lifetimes. Additionally, we employ cancer cells labelled with near infra-red fluorescent proteins (iRFP) to allow deep-tissue imaging. Our results demonstrate that time-domain imaging allows the detection of metastasis in deep-seated organs of living mice with a more than 20-fold increase in sensitivity compared to conventional continuous wave techniques. Furthermore, the distinct fluorescence lifetimes of each iRFP enables lifetime multiplexing of three different tumors, each expressing unique iRFP labels in the same animal. Fluorescence tomographic reconstructions reveal 3D distributions of iRFP720-expressing cancer cells in lungs and brain of live mice, allowing ready longitudinal monitoring of cancer cell fate with greater sensitivity than otherwise currently possible. PMID:25670171

Overview of deep learning in medical imaging.

PubMed

Suzuki, Kenji

2017-09-01

The use of machine learning (ML) has been increasing rapidly in the medical imaging field, including computer-aided diagnosis (CAD), radiomics, and medical image analysis. Recently, an ML area called deep learning emerged in the computer vision field and became very popular in many fields. It started from an event in late 2012, when a deep-learning approach based on a convolutional neural network (CNN) won an overwhelming victory in the best-known worldwide computer vision competition, ImageNet Classification. Since then, researchers in virtually all fields, including medical imaging, have started actively participating in the explosively growing field of deep learning. In this paper, the area of deep learning in medical imaging is overviewed, including (1) what was changed in machine learning before and after the introduction of deep learning, (2) what is the source of the power of deep learning, (3) two major deep-learning models: a massive-training artificial neural network (MTANN) and a convolutional neural network (CNN), (4) similarities and differences between the two models, and (5) their applications to medical imaging. This review shows that ML with feature input (or feature-based ML) was dominant before the introduction of deep learning, and that the major and essential difference between ML before and after deep learning is the learning of image data directly without object segmentation or feature extraction; thus, it is the source of the power of deep learning, although the depth of the model is an important attribute. The class of ML with image input (or image-based ML) including deep learning has a long history, but recently gained popularity due to the use of the new terminology, deep learning. There are two major models in this class of ML in medical imaging, MTANN and CNN, which have similarities as well as several differences. In our experience, MTANNs were substantially more efficient in their development, had a higher performance, and required a

Ultrafast ultrasound localization microscopy for deep super-resolution vascular imaging

NASA Astrophysics Data System (ADS)

Errico, Claudia; Pierre, Juliette; Pezet, Sophie; Desailly, Yann; Lenkei, Zsolt; Couture, Olivier; Tanter, Mickael

2015-11-01

Non-invasive imaging deep into organs at microscopic scales remains an open quest in biomedical imaging. Although optical microscopy is still limited to surface imaging owing to optical wave diffusion and fast decorrelation in tissue, revolutionary approaches such as fluorescence photo-activated localization microscopy led to a striking increase in resolution by more than an order of magnitude in the last decade. In contrast with optics, ultrasonic waves propagate deep into organs without losing their coherence and are much less affected by in vivo decorrelation processes. However, their resolution is impeded by the fundamental limits of diffraction, which impose a long-standing trade-off between resolution and penetration. This limits clinical and preclinical ultrasound imaging to a sub-millimetre scale. Here we demonstrate in vivo that ultrasound imaging at ultrafast frame rates (more than 500 frames per second) provides an analogue to optical localization microscopy by capturing the transient signal decorrelation of contrast agents—inert gas microbubbles. Ultrafast ultrasound localization microscopy allowed both non-invasive sub-wavelength structural imaging and haemodynamic quantification of rodent cerebral microvessels (less than ten micrometres in diameter) more than ten millimetres below the tissue surface, leading to transcranial whole-brain imaging within short acquisition times (tens of seconds). After intravenous injection, single echoes from individual microbubbles were detected through ultrafast imaging. Their localization, not limited by diffraction, was accumulated over 75,000 images, yielding 1,000,000 events per coronal plane and statistically independent pixels of ten micrometres in size. Precise temporal tracking of microbubble positions allowed us to extract accurately in-plane velocities of the blood flow with a large dynamic range (from one millimetre per second to several centimetres per second). These results pave the way for deep non

Ultrafast ultrasound localization microscopy for deep super-resolution vascular imaging.

PubMed

Errico, Claudia; Pierre, Juliette; Pezet, Sophie; Desailly, Yann; Lenkei, Zsolt; Couture, Olivier; Tanter, Mickael

2015-11-26

Non-invasive imaging deep into organs at microscopic scales remains an open quest in biomedical imaging. Although optical microscopy is still limited to surface imaging owing to optical wave diffusion and fast decorrelation in tissue, revolutionary approaches such as fluorescence photo-activated localization microscopy led to a striking increase in resolution by more than an order of magnitude in the last decade. In contrast with optics, ultrasonic waves propagate deep into organs without losing their coherence and are much less affected by in vivo decorrelation processes. However, their resolution is impeded by the fundamental limits of diffraction, which impose a long-standing trade-off between resolution and penetration. This limits clinical and preclinical ultrasound imaging to a sub-millimetre scale. Here we demonstrate in vivo that ultrasound imaging at ultrafast frame rates (more than 500 frames per second) provides an analogue to optical localization microscopy by capturing the transient signal decorrelation of contrast agents--inert gas microbubbles. Ultrafast ultrasound localization microscopy allowed both non-invasive sub-wavelength structural imaging and haemodynamic quantification of rodent cerebral microvessels (less than ten micrometres in diameter) more than ten millimetres below the tissue surface, leading to transcranial whole-brain imaging within short acquisition times (tens of seconds). After intravenous injection, single echoes from individual microbubbles were detected through ultrafast imaging. Their localization, not limited by diffraction, was accumulated over 75,000 images, yielding 1,000,000 events per coronal plane and statistically independent pixels of ten micrometres in size. Precise temporal tracking of microbubble positions allowed us to extract accurately in-plane velocities of the blood flow with a large dynamic range (from one millimetre per second to several centimetres per second). These results pave the way for deep non

Dual targeting luminescent gold nanoclusters for tumor imaging and deep tissue therapy.

PubMed

Chen, Dan; Li, Bowen; Cai, Songhua; Wang, Peng; Peng, Shuwen; Sheng, Yuanzhi; He, Yuanyuan; Gu, Yueqing; Chen, Haiyan

2016-09-01

Dual targeting towards both extracellular and intracellular receptors specific to tumor is a significant approach for cancer diagnosis and therapy. In the present study, a novel nano-platform (AuNC-cRGD-Apt) with dual targeting function was initially established by conjugating gold nanocluster (AuNC) with cyclic RGD (cRGD) that is specific to αvβ3integrins over-expressed on the surface of tumor tissues and aptamer AS1411 (Apt) that is of high affinity to nucleolin over-expressed in the cytoplasm and nucleus of tumor cells. Then, AuNC-cRGD-Apt was further functionalized with near infrared (NIR) fluorescence dye (MPA), giving a NIR fluorescent dual-targeting probe AuNC-MPA-cRGD-Apt. AuNC-MPA-cRGD-Apt displays low cytotoxicity and favorable tumor-targeting capability at both in vitro and in vivo level, suggesting its clinical potential for tumor imaging. Additionally, Doxorubicin (DOX), a widely used clinical chemotherapeutic drug that kill cancer cells by intercalating DNA in cellular nucleus, was immobilized onto AuNC-cRGD-Apt forming a pro-drug, AuNC-DOX-cRGD-Apt. The enhanced tumor affinity, deep tumor penetration and improved anti-tumor activity of this pro-drug were demonstrated in different tumor cell lines, tumor spheroid and tumor-bearing mouse models. Results in this study suggest not only the prospect of non-toxic AuNC modified with two targeting ligands for tumor targeted imaging, but also confirm the promising future of dual targeting AuNC as a core for the design of prodrug in the field of cancer therapy. Copyright © 2016 Elsevier Ltd. All rights reserved.

Prediction of Tissue Outcome and Assessment of Treatment Effect in Acute Ischemic Stroke Using Deep Learning.

PubMed

Nielsen, Anne; Hansen, Mikkel Bo; Tietze, Anna; Mouridsen, Kim

2018-06-01

Treatment options for patients with acute ischemic stroke depend on the volume of salvageable tissue. This volume assessment is currently based on fixed thresholds and single imagine modalities, limiting accuracy. We wish to develop and validate a predictive model capable of automatically identifying and combining acute imaging features to accurately predict final lesion volume. Using acute magnetic resonance imaging, we developed and trained a deep convolutional neural network (CNN deep ) to predict final imaging outcome. A total of 222 patients were included, of which 187 were treated with rtPA (recombinant tissue-type plasminogen activator). The performance of CNN deep was compared with a shallow CNN based on the perfusion-weighted imaging biomarker Tmax (CNN Tmax ), a shallow CNN based on a combination of 9 different biomarkers (CNN shallow ), a generalized linear model, and thresholding of the diffusion-weighted imaging biomarker apparent diffusion coefficient (ADC) at 600×10 -6 mm 2 /s (ADC thres ). To assess whether CNN deep is capable of differentiating outcomes of ±intravenous rtPA, patients not receiving intravenous rtPA were included to train CNN deep, -rtpa to access a treatment effect. The networks' performances were evaluated using visual inspection, area under the receiver operating characteristic curve (AUC), and contrast. CNN deep yields significantly better performance in predicting final outcome (AUC=0.88±0.12) than generalized linear model (AUC=0.78±0.12; P =0.005), CNN Tmax (AUC=0.72±0.14; P <0.003), and ADC thres (AUC=0.66±0.13; P <0.0001) and a substantially better performance than CNN shallow (AUC=0.85±0.11; P =0.063). Measured by contrast, CNN deep improves the predictions significantly, showing superiority to all other methods ( P ≤0.003). CNN deep also seems to be able to differentiate outcomes based on treatment strategy with the volume of final infarct being significantly different ( P =0.048). The considerable prediction

Seeing through Musculoskeletal Tissues: Improving In Situ Imaging of Bone and the Lacunar Canalicular System through Optical Clearing

PubMed Central

Berke, Ian M.; Miola, Joseph P.; David, Michael A.; Smith, Melanie K.; Price, Christopher

2016-01-01

In situ, cells of the musculoskeletal system reside within complex and often interconnected 3-D environments. Key to better understanding how 3-D tissue and cellular environments regulate musculoskeletal physiology, homeostasis, and health is the use of robust methodologies for directly visualizing cell-cell and cell-matrix architecture in situ. However, the use of standard optical imaging techniques is often of limited utility in deep imaging of intact musculoskeletal tissues due to the highly scattering nature of biological tissues. Drawing inspiration from recent developments in the deep-tissue imaging field, we describe the application of immersion based optical clearing techniques, which utilize the principle of refractive index (RI) matching between the clearing/mounting media and tissue under observation, to improve the deep, in situ imaging of musculoskeletal tissues. To date, few optical clearing techniques have been applied specifically to musculoskeletal tissues, and a systematic comparison of the clearing ability of optical clearing agents in musculoskeletal tissues has yet to be fully demonstrated. In this study we tested the ability of eight different aqueous and non-aqueous clearing agents, with RIs ranging from 1.45 to 1.56, to optically clear murine knee joints and cortical bone. We demonstrated and quantified the ability of these optical clearing agents to clear musculoskeletal tissues and improve both macro- and micro-scale imaging of musculoskeletal tissue across several imaging modalities (stereomicroscopy, spectroscopy, and one-, and two-photon confocal microscopy) and investigational techniques (dynamic bone labeling and en bloc tissue staining). Based upon these findings we believe that optical clearing, in combination with advanced imaging techniques, has the potential to complement classical musculoskeletal analysis techniques; opening the door for improved in situ investigation and quantification of musculoskeletal tissues. PMID:26930293

Seeing through Musculoskeletal Tissues: Improving In Situ Imaging of Bone and the Lacunar Canalicular System through Optical Clearing.

PubMed

Berke, Ian M; Miola, Joseph P; David, Michael A; Smith, Melanie K; Price, Christopher

2016-01-01

In situ, cells of the musculoskeletal system reside within complex and often interconnected 3-D environments. Key to better understanding how 3-D tissue and cellular environments regulate musculoskeletal physiology, homeostasis, and health is the use of robust methodologies for directly visualizing cell-cell and cell-matrix architecture in situ. However, the use of standard optical imaging techniques is often of limited utility in deep imaging of intact musculoskeletal tissues due to the highly scattering nature of biological tissues. Drawing inspiration from recent developments in the deep-tissue imaging field, we describe the application of immersion based optical clearing techniques, which utilize the principle of refractive index (RI) matching between the clearing/mounting media and tissue under observation, to improve the deep, in situ imaging of musculoskeletal tissues. To date, few optical clearing techniques have been applied specifically to musculoskeletal tissues, and a systematic comparison of the clearing ability of optical clearing agents in musculoskeletal tissues has yet to be fully demonstrated. In this study we tested the ability of eight different aqueous and non-aqueous clearing agents, with RIs ranging from 1.45 to 1.56, to optically clear murine knee joints and cortical bone. We demonstrated and quantified the ability of these optical clearing agents to clear musculoskeletal tissues and improve both macro- and micro-scale imaging of musculoskeletal tissue across several imaging modalities (stereomicroscopy, spectroscopy, and one-, and two-photon confocal microscopy) and investigational techniques (dynamic bone labeling and en bloc tissue staining). Based upon these findings we believe that optical clearing, in combination with advanced imaging techniques, has the potential to complement classical musculoskeletal analysis techniques; opening the door for improved in situ investigation and quantification of musculoskeletal tissues.

Blind CT image quality assessment via deep learning strategy: initial study

NASA Astrophysics Data System (ADS)

Li, Sui; He, Ji; Wang, Yongbo; Liao, Yuting; Zeng, Dong; Bian, Zhaoying; Ma, Jianhua

2018-03-01

Computed Tomography (CT) is one of the most important medical imaging modality. CT images can be used to assist in the detection and diagnosis of lesions and to facilitate follow-up treatment. However, CT images are vulnerable to noise. Actually, there are two major source intrinsically causing the CT data noise, i.e., the X-ray photo statistics and the electronic noise background. Therefore, it is necessary to doing image quality assessment (IQA) in CT imaging before diagnosis and treatment. Most of existing CT images IQA methods are based on human observer study. However, these methods are impractical in clinical for their complex and time-consuming. In this paper, we presented a blind CT image quality assessment via deep learning strategy. A database of 1500 CT images is constructed, containing 300 high-quality images and 1200 corresponding noisy images. Specifically, the high-quality images were used to simulate the corresponding noisy images at four different doses. Then, the images are scored by the experienced radiologists by the following attributes: image noise, artifacts, edge and structure, overall image quality, and tumor size and boundary estimation with five-point scale. We trained a network for learning the non-liner map from CT images to subjective evaluation scores. Then, we load the pre-trained model to yield predicted score from the test image. To demonstrate the performance of the deep learning network in IQA, correlation coefficients: Pearson Linear Correlation Coefficient (PLCC) and Spearman Rank Order Correlation Coefficient (SROCC) are utilized. And the experimental result demonstrate that the presented deep learning based IQA strategy can be used in the CT image quality assessment.

High Definition Confocal Imaging Modalities for the Characterization of Tissue-Engineered Substitutes.

PubMed

Mayrand, Dominique; Fradette, Julie

2018-01-01

Optimal imaging methods are necessary in order to perform a detailed characterization of thick tissue samples from either native or engineered tissues. Tissue-engineered substitutes are featuring increasing complexity including multiple cell types and capillary-like networks. Therefore, technical approaches allowing the visualization of the inner structural organization and cellular composition of tissues are needed. This chapter describes an optical clearing technique which facilitates the detailed characterization of whole-mount samples from skin and adipose tissues (ex vivo tissues and in vitro tissue-engineered substitutes) when combined with spectral confocal microscopy and quantitative analysis on image renderings.

Segmenting Brain Tissues from Chinese Visible Human Dataset by Deep-Learned Features with Stacked Autoencoder

PubMed Central

Zhao, Guangjun; Wang, Xuchu; Niu, Yanmin; Tan, Liwen; Zhang, Shao-Xiang

2016-01-01

Cryosection brain images in Chinese Visible Human (CVH) dataset contain rich anatomical structure information of tissues because of its high resolution (e.g., 0.167 mm per pixel). Fast and accurate segmentation of these images into white matter, gray matter, and cerebrospinal fluid plays a critical role in analyzing and measuring the anatomical structures of human brain. However, most existing automated segmentation methods are designed for computed tomography or magnetic resonance imaging data, and they may not be applicable for cryosection images due to the imaging difference. In this paper, we propose a supervised learning-based CVH brain tissues segmentation method that uses stacked autoencoder (SAE) to automatically learn the deep feature representations. Specifically, our model includes two successive parts where two three-layer SAEs take image patches as input to learn the complex anatomical feature representation, and then these features are sent to Softmax classifier for inferring the labels. Experimental results validated the effectiveness of our method and showed that it outperformed four other classical brain tissue detection strategies. Furthermore, we reconstructed three-dimensional surfaces of these tissues, which show their potential in exploring the high-resolution anatomical structures of human brain. PMID:27057543

Segmenting Brain Tissues from Chinese Visible Human Dataset by Deep-Learned Features with Stacked Autoencoder.

PubMed

Zhao, Guangjun; Wang, Xuchu; Niu, Yanmin; Tan, Liwen; Zhang, Shao-Xiang

2016-01-01

Cryosection brain images in Chinese Visible Human (CVH) dataset contain rich anatomical structure information of tissues because of its high resolution (e.g., 0.167 mm per pixel). Fast and accurate segmentation of these images into white matter, gray matter, and cerebrospinal fluid plays a critical role in analyzing and measuring the anatomical structures of human brain. However, most existing automated segmentation methods are designed for computed tomography or magnetic resonance imaging data, and they may not be applicable for cryosection images due to the imaging difference. In this paper, we propose a supervised learning-based CVH brain tissues segmentation method that uses stacked autoencoder (SAE) to automatically learn the deep feature representations. Specifically, our model includes two successive parts where two three-layer SAEs take image patches as input to learn the complex anatomical feature representation, and then these features are sent to Softmax classifier for inferring the labels. Experimental results validated the effectiveness of our method and showed that it outperformed four other classical brain tissue detection strategies. Furthermore, we reconstructed three-dimensional surfaces of these tissues, which show their potential in exploring the high-resolution anatomical structures of human brain.

Deep learning and three-compartment breast imaging in breast cancer diagnosis

NASA Astrophysics Data System (ADS)

Drukker, Karen; Huynh, Benjamin Q.; Giger, Maryellen L.; Malkov, Serghei; Avila, Jesus I.; Fan, Bo; Joe, Bonnie; Kerlikowske, Karla; Drukteinis, Jennifer S.; Kazemi, Leila; Pereira, Malesa M.; Shepherd, John

2017-03-01

We investigated whether deep learning has potential to aid in the diagnosis of breast cancer when applied to mammograms and biologic tissue composition images derived from three-compartment (3CB) imaging. The dataset contained diagnostic mammograms and 3CB images (water, lipid, and protein content) of biopsy-sampled BIRADS 4 and 5 lesions in 195 patients. In 58 patients, the lesion manifested as a mass (13 malignant vs. 45 benign), in 87 as microcalcifications (19 vs. 68), and in 56 as (focal) asymmetry or architectural distortion (11 vs. 45). Six patients had both a mass and calcifications. For each mammogram and corresponding 3CB images, a 128x128 region of interest containing the lesion was selected by an expert radiologist and used directly as input to a deep learning method pretrained on a very large independent set of non-medical images. We used a nested leave-one-out-by-case (patient) model selection and classification protocol. The area under the ROC curve (AUC) for the task of distinguishing between benign and malignant lesions was used as performance metric. For the cases with mammographic masses, the AUC increased from 0.83 (mammograms alone) to 0.89 (mammograms+3CB, p=.162). For the microcalcification and asymmetry/architectural distortion cases the AUC increased from 0.84 to 0.91 (p=.116) and from 0.61 to 0.87 (p=.006), respectively. Our results indicate great potential for the application of deep learning methods in the diagnosis of breast cancer and additional knowledge of the biologic tissue composition appeared to improve performance, especially for lesions mammographically manifesting as asymmetries or architectural distortions.

Sex differences and hormonal modulation of deep tissue pain

PubMed Central

Traub, Richard J.; Ji, Yaping

2013-01-01

Women disproportionately suffer from many deep tissue pain conditions. Experimental studies show that women have lower pain thresholds, higher pain ratings and less tolerance to a range of painful stimuli. Most clinical and epidemiological reports suggest female gonadal hormones modulate pain for some, but not all, conditions. Similarly, animal studies support greater nociceptive sensitivity in females in many deep tissue pain models. Gonadal hormones modulate responses in primary afferents, dorsal horn neurons and supraspinal sites, but the direction of modulation is variable. This review will examine sex differences in deep tissue pain in humans and animals focusing on the role of gonadal hormones (mainly estradiol) as an underlying component of the modulation of pain sensitivity. PMID:23872333

Combining deep learning and coherent anti-Stokes Raman scattering imaging for automated differential diagnosis of lung cancer

NASA Astrophysics Data System (ADS)

Weng, Sheng; Xu, Xiaoyun; Li, Jiasong; Wong, Stephen T. C.

2017-10-01

Lung cancer is the most prevalent type of cancer and the leading cause of cancer-related deaths worldwide. Coherent anti-Stokes Raman scattering (CARS) is capable of providing cellular-level images and resolving pathologically related features on human lung tissues. However, conventional means of analyzing CARS images requires extensive image processing, feature engineering, and human intervention. This study demonstrates the feasibility of applying a deep learning algorithm to automatically differentiate normal and cancerous lung tissue images acquired by CARS. We leverage the features learned by pretrained deep neural networks and retrain the model using CARS images as the input. We achieve 89.2% accuracy in classifying normal, small-cell carcinoma, adenocarcinoma, and squamous cell carcinoma lung images. This computational method is a step toward on-the-spot diagnosis of lung cancer and can be further strengthened by the efforts aimed at miniaturizing the CARS technique for fiber-based microendoscopic imaging.

Deep Imaging Survey

NASA Image and Video Library

2003-07-25

This is the first Deep Imaging Survey image taken by NASA Galaxy Evolution Explorer. On June 22 and 23, 2003, the spacecraft obtained this near ultraviolet image of the Groth region by adding multiple orbits for a total exposure time of 14,000 seconds. Tens of thousands of objects can be identified in this picture. http://photojournal.jpl.nasa.gov/catalog/PIA04627

An activity-based near-infrared glucuronide trapping probe for imaging β-glucuronidase expression in deep tissues.

PubMed

Cheng, Ta-Chun; Roffler, Steve R; Tzou, Shey-Cherng; Chuang, Kuo-Hsiang; Su, Yu-Cheng; Chuang, Chih-Hung; Kao, Chien-Han; Chen, Chien-Shu; Harn, I-Hong; Liu, Kuan-Yi; Cheng, Tian-Lu; Leu, Yu-Ling

2012-02-15

β-glucuronidase is an attractive reporter and prodrug-converting enzyme. The development of near-IR (NIR) probes for imaging of β-glucuronidase activity would be ideal to allow estimation of reporter expression and for personalized glucuronide prodrug cancer therapy in preclinical studies. However, NIR glucuronide probes are not yet available. In this work, we developed two fluorescent probes for detection of β-glucuronidase activity, one for the NIR range (containing IR-820 dye) and the other for the visible range [containing fluorescein isothiocyanate (FITC)], by utilizing a difluoromethylphenol-glucuronide moiety (TrapG) to trap the fluorochromes in the vicinity of the active enzyme. β-glucuronidase-mediated hydrolysis of the glucuronyl bond of TrapG generates a highly reactive alkylating group that facilitates the attachment of the fluorochrome to nucleophilic moieties located near β-glucuronidase-expressing sites. FITC-TrapG was selectively trapped on purified β-glucuronidase or β-glucuronidase-expressing CT26 cells (CT26/mβG) but not on bovine serum albumin or non-β-glucuronidase-expressing CT26 cells used as controls. β-glucuronidase-activated FITC-TrapG did not interfere with β-glucuronidase activity and could label bystander proteins near β-glucuronidase. Both FITC-TrapG and NIR-TrapG specifically imaged subcutaneous CT26/mβG tumors, but only NIR-TrapG could image CT26/mβG tumors transplanted deep in the liver. Thus NIR-TrapG may provide a valuable tool for visualizing β-glucuronidase activity in vivo.

Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification.

PubMed

Sladojevic, Srdjan; Arsenovic, Marko; Anderla, Andras; Culibrk, Dubravko; Stefanovic, Darko

2016-01-01

The latest generation of convolutional neural networks (CNNs) has achieved impressive results in the field of image classification. This paper is concerned with a new approach to the development of plant disease recognition model, based on leaf image classification, by the use of deep convolutional networks. Novel way of training and the methodology used facilitate a quick and easy system implementation in practice. The developed model is able to recognize 13 different types of plant diseases out of healthy leaves, with the ability to distinguish plant leaves from their surroundings. According to our knowledge, this method for plant disease recognition has been proposed for the first time. All essential steps required for implementing this disease recognition model are fully described throughout the paper, starting from gathering images in order to create a database, assessed by agricultural experts. Caffe, a deep learning framework developed by Berkley Vision and Learning Centre, was used to perform the deep CNN training. The experimental results on the developed model achieved precision between 91% and 98%, for separate class tests, on average 96.3%.

Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification

PubMed Central

Sladojevic, Srdjan; Arsenovic, Marko; Culibrk, Dubravko; Stefanovic, Darko

2016-01-01

The latest generation of convolutional neural networks (CNNs) has achieved impressive results in the field of image classification. This paper is concerned with a new approach to the development of plant disease recognition model, based on leaf image classification, by the use of deep convolutional networks. Novel way of training and the methodology used facilitate a quick and easy system implementation in practice. The developed model is able to recognize 13 different types of plant diseases out of healthy leaves, with the ability to distinguish plant leaves from their surroundings. According to our knowledge, this method for plant disease recognition has been proposed for the first time. All essential steps required for implementing this disease recognition model are fully described throughout the paper, starting from gathering images in order to create a database, assessed by agricultural experts. Caffe, a deep learning framework developed by Berkley Vision and Learning Centre, was used to perform the deep CNN training. The experimental results on the developed model achieved precision between 91% and 98%, for separate class tests, on average 96.3%. PMID:27418923

Deep and Structured Robust Information Theoretic Learning for Image Analysis.

PubMed

Deng, Yue; Bao, Feng; Deng, Xuesong; Wang, Ruiping; Kong, Youyong; Dai, Qionghai

2016-07-07

This paper presents a robust information theoretic (RIT) model to reduce the uncertainties, i.e. missing and noisy labels, in general discriminative data representation tasks. The fundamental pursuit of our model is to simultaneously learn a transformation function and a discriminative classifier that maximize the mutual information of data and their labels in the latent space. In this general paradigm, we respectively discuss three types of the RIT implementations with linear subspace embedding, deep transformation and structured sparse learning. In practice, the RIT and deep RIT are exploited to solve the image categorization task whose performances will be verified on various benchmark datasets. The structured sparse RIT is further applied to a medical image analysis task for brain MRI segmentation that allows group-level feature selections on the brain tissues.

Relevance of deep learning to facilitate the diagnosis of HER2 status in breast cancer

PubMed Central

Vandenberghe, Michel E.; Scott, Marietta L. J.; Scorer, Paul W.; Söderberg, Magnus; Balcerzak, Denis; Barker, Craig

2017-01-01

Tissue biomarker scoring by pathologists is central to defining the appropriate therapy for patients with cancer. Yet, inter-pathologist variability in the interpretation of ambiguous cases can affect diagnostic accuracy. Modern artificial intelligence methods such as deep learning have the potential to supplement pathologist expertise to ensure constant diagnostic accuracy. We developed a computational approach based on deep learning that automatically scores HER2, a biomarker that defines patient eligibility for anti-HER2 targeted therapies in breast cancer. In a cohort of 71 breast tumour resection samples, automated scoring showed a concordance of 83% with a pathologist. The twelve discordant cases were then independently reviewed, leading to a modification of diagnosis from initial pathologist assessment for eight cases. Diagnostic discordance was found to be largely caused by perceptual differences in assessing HER2 expression due to high HER2 staining heterogeneity. This study provides evidence that deep learning aided diagnosis can facilitate clinical decision making in breast cancer by identifying cases at high risk of misdiagnosis. PMID:28378829

RGD-conjugated two-photon absorbing near-IR emitting fluorescent probes for tumor vascular imaging (Conference Presentation)

NASA Astrophysics Data System (ADS)

Belfield, Kevin D.; Yue, Xiling; Morales, Alma R.; Githaiga, Grace W.; Woodward, Adam W.; Tang, Simon; Sawada, Junko; Komatsu, Masanobu; Liu, Xuan

2016-03-01

Observation of the activation and inhibition of angiogenesis processes is important in the progression of cancer. Application of targeting peptides, such as a small peptide that contains adjacent L-arginine (R), glycine (G) and L-aspartic acid (D) residues can afford high selectivity and deep penetration in vessel imaging. To facilitate deep tissue vasculature imaging, probes that can be excited via two-photon absorption (2PA) in the near-infrared (NIR) and subsequently emit in the NIR are essential. In this study, the enhancement of tissue image quality with RGD conjugates was investigated with new NIR-emitting pyranyl fluorophore derivatives in two-photon fluorescence microscopy. Linear and nonlinear photophysical properties of the new probes were comprehensively characterized; significantly the probes exhibited good 2PA over a broad spectral range from 700-1100 nm. Cell and tissue images were then acquired and examined, revealing deep penetration and high contrast with the new pyranyl RGD-conjugates up to 350 μm in tumor tissue.

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

Combining deep learning and coherent anti-Stokes Raman scattering imaging for automated differential diagnosis of lung cancer.

PubMed

Weng, Sheng; Xu, Xiaoyun; Li, Jiasong; Wong, Stephen T C

2017-10-01

Lung cancer is the most prevalent type of cancer and the leading cause of cancer-related deaths worldwide. Coherent anti-Stokes Raman scattering (CARS) is capable of providing cellular-level images and resolving pathologically related features on human lung tissues. However, conventional means of analyzing CARS images requires extensive image processing, feature engineering, and human intervention. This study demonstrates the feasibility of applying a deep learning algorithm to automatically differentiate normal and cancerous lung tissue images acquired by CARS. We leverage the features learned by pretrained deep neural networks and retrain the model using CARS images as the input. We achieve 89.2% accuracy in classifying normal, small-cell carcinoma, adenocarcinoma, and squamous cell carcinoma lung images. This computational method is a step toward on-the-spot diagnosis of lung cancer and can be further strengthened by the efforts aimed at miniaturizing the CARS technique for fiber-based microendoscopic imaging. (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE).

Assessment of deep tissue hyperalgesia in the groin - a method comparison of electrical vs. pressure stimulation.

PubMed

Aasvang, E K; Werner, M U; Kehlet, H

2014-09-01

Deep pain complaints are more frequent than cutaneous in post-surgical patients, and a prevalent finding in quantitative sensory testing studies. However, the preferred assessment method - pressure algometry - is indirect and tissue unspecific, hindering advances in treatment and preventive strategies. Thus, there is a need for development of methods with direct stimulation of suspected hyperalgesic tissues to identify the peripheral origin of nociceptive input. We compared the reliability of an ultrasound-guided needle stimulation protocol of electrical detection and pain thresholds to pressure algometry, by performing identical test-retest sequences 10 days apart, in deep tissues in the groin region. Electrical stimulation was performed by five up-and-down staircase series of single impulses of 0.04 ms duration, starting from 0 mA in increments of 0.2 mA until a threshold was reached and descending until sensation was lost. Method reliability was assessed by Bland-Altman plots, descriptive statistics, coefficients of variance and intraclass correlation coefficients. The electrical stimulation method was comparable to pressure algometry regarding 10 days test-retest repeatability, but with superior same-day reliability for electrical stimulation (P < 0.05). Between-subject variance rather than within-subject variance was the main source for test variation. There were no systematic differences in electrical thresholds across tissues and locations (P > 0.05). The presented tissue-specific direct deep tissue electrical stimulation technique has equal or superior reliability compared with the indirect tissue-unspecific stimulation by pressure algometry. This method may facilitate advances in mechanism based preventive and treatment strategies in acute and chronic post-surgical pain states. © 2014 The Acta Anaesthesiologica Scandinavica Foundation. Published by John Wiley & Sons Ltd.

Recent progress in tissue optical clearing for spectroscopic application

NASA Astrophysics Data System (ADS)

Sdobnov, A. Yu.; Darvin, M. E.; Genina, E. A.; Bashkatov, A. N.; Lademann, J.; Tuchin, V. V.

2018-05-01

This paper aims to review recent progress in optical clearing of the skin and over naturally turbid biological tissues and blood using this technique in vivo and in vitro with multiphoton microscopy, confocal Raman microscopy, confocal microscopy, NIR spectroscopy, optical coherence tomography, and laser speckle contrast imaging. Basic principles of the technique, its safety, advantages and limitations are discussed. The application of optical clearing agent on a tissue allows for controlling the optical properties of tissue. Optical clearing-induced reduction of tissue scattering significantly facilitates the observation of deep-located tissue regions, at the same time improving the resolution and image contrast for a variety of optical imaging methods suitable for clinical applications, such as diagnostics and laser treatment of skin diseases, mucosal tumor imaging, laser disruption of pathological abnormalities, etc. Structural images of different skin layers obtained ex vivo for porcine ear skin samples at application of Omnipaque™ and glycerol solutions during 60 min. Red color corresponds to TPEAF signal channel. Green color corresponds to SHG signal channel.

Groth Deep Image

NASA Image and Video Library

2003-07-25

This ultraviolet color blowup of the Groth Deep Image was taken by NASA Galaxy Evolution Explorer on June 22 and June 23, 2003. Many hundreds of galaxies are detected in this portion of the image. NASA astronomers believe the faint red galaxies are 6 billion light years away. http://photojournal.jpl.nasa.gov/catalog/PIA04625

In vivo mapping of current density distribution in brain tissues during deep brain stimulation (DBS)

NASA Astrophysics Data System (ADS)

Sajib, Saurav Z. K.; Oh, Tong In; Kim, Hyung Joong; Kwon, Oh In; Woo, Eung Je

2017-01-01

New methods for in vivo mapping of brain responses during deep brain stimulation (DBS) are indispensable to secure clinical applications. Assessment of current density distribution, induced by internally injected currents, may provide an alternative method for understanding the therapeutic effects of electrical stimulation. The current flow and pathway are affected by internal conductivity, and can be imaged using magnetic resonance-based conductivity imaging methods. Magnetic resonance electrical impedance tomography (MREIT) is an imaging method that can enable highly resolved mapping of electromagnetic tissue properties such as current density and conductivity of living tissues. In the current study, we experimentally imaged current density distribution of in vivo canine brains by applying MREIT to electrical stimulation. The current density maps of three canine brains were calculated from the measured magnetic flux density data. The absolute current density values of brain tissues, including gray matter, white matter, and cerebrospinal fluid were compared to assess the active regions during DBS. The resulting current density in different tissue types may provide useful information about current pathways and volume activation for adjusting surgical planning and understanding the therapeutic effects of DBS.

Automatic segmentation of the prostate on CT images using deep learning and multi-atlas fusion

NASA Astrophysics Data System (ADS)

Ma, Ling; Guo, Rongrong; Zhang, Guoyi; Tade, Funmilayo; Schuster, David M.; Nieh, Peter; Master, Viraj; Fei, Baowei

2017-02-01

Automatic segmentation of the prostate on CT images has many applications in prostate cancer diagnosis and therapy. However, prostate CT image segmentation is challenging because of the low contrast of soft tissue on CT images. In this paper, we propose an automatic segmentation method by combining a deep learning method and multi-atlas refinement. First, instead of segmenting the whole image, we extract the region of interesting (ROI) to delete irrelevant regions. Then, we use the convolutional neural networks (CNN) to learn the deep features for distinguishing the prostate pixels from the non-prostate pixels in order to obtain the preliminary segmentation results. CNN can automatically learn the deep features adapting to the data, which are different from some handcrafted features. Finally, we select some similar atlases to refine the initial segmentation results. The proposed method has been evaluated on a dataset of 92 prostate CT images. Experimental results show that our method achieved a Dice similarity coefficient of 86.80% as compared to the manual segmentation. The deep learning based method can provide a useful tool for automatic segmentation of the prostate on CT images and thus can have a variety of clinical applications.

A Deep Learning Approach to Digitally Stain Optical Coherence Tomography Images of the Optic Nerve Head.

PubMed

Devalla, Sripad Krishna; Chin, Khai Sing; Mari, Jean-Martial; Tun, Tin A; Strouthidis, Nicholas G; Aung, Tin; Thiéry, Alexandre H; Girard, Michaël J A

2018-01-01

To develop a deep learning approach to digitally stain optical coherence tomography (OCT) images of the optic nerve head (ONH). A horizontal B-scan was acquired through the center of the ONH using OCT (Spectralis) for one eye of each of 100 subjects (40 healthy and 60 glaucoma). All images were enhanced using adaptive compensation. A custom deep learning network was then designed and trained with the compensated images to digitally stain (i.e., highlight) six tissue layers of the ONH. The accuracy of our algorithm was assessed (against manual segmentations) using the dice coefficient, sensitivity, specificity, intersection over union (IU), and accuracy. We studied the effect of compensation, number of training images, and performance comparison between glaucoma and healthy subjects. For images it had not yet assessed, our algorithm was able to digitally stain the retinal nerve fiber layer + prelamina, the RPE, all other retinal layers, the choroid, and the peripapillary sclera and lamina cribrosa. For all tissues, the dice coefficient, sensitivity, specificity, IU, and accuracy (mean) were 0.84 ± 0.03, 0.92 ± 0.03, 0.99 ± 0.00, 0.89 ± 0.03, and 0.94 ± 0.02, respectively. Our algorithm performed significantly better when compensated images were used for training (P < 0.001). Besides offering a good reliability, digital staining also performed well on OCT images of both glaucoma and healthy individuals. Our deep learning algorithm can simultaneously stain the neural and connective tissues of the ONH, offering a framework to automatically measure multiple key structural parameters of the ONH that may be critical to improve glaucoma management.

Biodynamic imaging for phenotypic profiling of three-dimensional tissue culture

NASA Astrophysics Data System (ADS)

Sun, Hao; Merrill, Daniel; An, Ran; Turek, John; Matei, Daniela; Nolte, David D.

2017-01-01

Three-dimensional (3-D) tissue culture represents a more biologically relevant environment for testing new drugs compared to conventional two-dimensional cancer cell culture models. Biodynamic imaging is a high-content 3-D optical imaging technology based on low-coherence interferometry and digital holography that uses dynamic speckle as high-content image contrast to probe deep inside 3-D tissue. Speckle contrast is shown to be a scaling function of the acquisition time relative to the persistence time of intracellular transport and hence provides a measure of cellular activity. Cellular responses of 3-D multicellular spheroids to paclitaxel are compared among three different growth techniques: rotating bioreactor (BR), hanging-drop (HD), and nonadherent (U-bottom, UB) plate spheroids, compared with ex vivo living tissues. HD spheroids have the most homogeneous tissue, whereas BR spheroids display large sample-to-sample variability as well as spatial heterogeneity. The responses of BR-grown tumor spheroids to paclitaxel are more similar to those of ex vivo biopsies than the responses of spheroids grown using HD or plate methods. The rate of mitosis inhibition by application of taxol is measured through tissue dynamics spectroscopic imaging, demonstrating the ability to monitor antimitotic chemotherapy. These results illustrate the potential use of low-coherence digital holography for 3-D pharmaceutical screening applications.

Optical toolkits for in vivo deep tissue laser scanning microscopy: a primer

NASA Astrophysics Data System (ADS)

Lee, Woei Ming; McMenamin, Thomas; Li, Yongxiao

2018-06-01

Life at the microscale is animated and multifaceted. The impact of dynamic in vivo microscopy in small animals has opened up opportunities to peer into a multitude of biological processes at the cellular scale in their native microenvironments. Laser scanning microscopy (LSM) coupled with targeted fluorescent proteins has become an indispensable tool to enable dynamic imaging in vivo at high temporal and spatial resolutions. In the last few decades, the technique has been translated from imaging cells in thin samples to mapping cells in the thick biological tissue of living organisms. Here, we sought to provide a concise overview of the design considerations of a LSM that enables cellular and subcellular imaging in deep tissue. Individual components under review include: long working distance microscope objectives, laser scanning technologies, adaptive optics devices, beam shaping technologies and photon detectors, with an emphasis on more recent advances. The review will conclude with the latest innovations in automated optical microscopy, which would impact tracking and quantification of heterogeneous populations of cells in vivo.

Deep, noninvasive imaging and surgical guidance of submillimeter tumors using targeted M13-stabilized single-walled carbon nanotubes

PubMed Central

Ghosh, Debadyuti; Bagley, Alexander F.; Na, Young Jeong; Birrer, Michael J.; Bhatia, Sangeeta N.; Belcher, Angela M.

2014-01-01

Highly sensitive detection of small, deep tumors for early diagnosis and surgical interventions remains a challenge for conventional imaging modalities. Second-window near-infrared light (NIR2, 950–1,400 nm) is promising for in vivo fluorescence imaging due to deep tissue penetration and low tissue autofluorescence. With their intrinsic fluorescence in the NIR2 regime and lack of photobleaching, single-walled carbon nanotubes (SWNTs) are potentially attractive contrast agents to detect tumors. Here, targeted M13 virus-stabilized SWNTs are used to visualize deep, disseminated tumors in vivo. This targeted nanoprobe, which uses M13 to stably display both tumor-targeting peptides and an SWNT imaging probe, demonstrates excellent tumor-to-background uptake and exhibits higher signal-to-noise performance compared with visible and near-infrared (NIR1) dyes for delineating tumor nodules. Detection and excision of tumors by a gynecological surgeon improved with SWNT image guidance and led to the identification of submillimeter tumors. Collectively, these findings demonstrate the promise of targeted SWNT nanoprobes for noninvasive disease monitoring and guided surgery. PMID:25214538

Deep, noninvasive imaging and surgical guidance of submillimeter tumors using targeted M13-stabilized single-walled carbon nanotubes.

PubMed

Ghosh, Debadyuti; Bagley, Alexander F; Na, Young Jeong; Birrer, Michael J; Bhatia, Sangeeta N; Belcher, Angela M

2014-09-23

Highly sensitive detection of small, deep tumors for early diagnosis and surgical interventions remains a challenge for conventional imaging modalities. Second-window near-infrared light (NIR2, 950-1,400 nm) is promising for in vivo fluorescence imaging due to deep tissue penetration and low tissue autofluorescence. With their intrinsic fluorescence in the NIR2 regime and lack of photobleaching, single-walled carbon nanotubes (SWNTs) are potentially attractive contrast agents to detect tumors. Here, targeted M13 virus-stabilized SWNTs are used to visualize deep, disseminated tumors in vivo. This targeted nanoprobe, which uses M13 to stably display both tumor-targeting peptides and an SWNT imaging probe, demonstrates excellent tumor-to-background uptake and exhibits higher signal-to-noise performance compared with visible and near-infrared (NIR1) dyes for delineating tumor nodules. Detection and excision of tumors by a gynecological surgeon improved with SWNT image guidance and led to the identification of submillimeter tumors. Collectively, these findings demonstrate the promise of targeted SWNT nanoprobes for noninvasive disease monitoring and guided surgery.

Time-lapse imaging of disease progression in deep brain areas using fluorescence microendoscopy

PubMed Central

Barretto, Robert P. J.; Ko, Tony H.; Jung, Juergen C.; Wang, Tammy J.; Capps, George; Waters, Allison C.; Ziv, Yaniv; Attardo, Alessio; Recht, Lawrence; Schnitzer, Mark J.

2013-01-01

The combination of intravital microscopy and animal models of disease has propelled studies of disease mechanisms and treatments. However, many disorders afflict tissues inaccessible to light microscopy in live subjects. Here we introduce cellular-level time-lapse imaging deep within the live mammalian brain by one- and two-photon fluorescence microendoscopy over multiple weeks. Bilateral imaging sites allowed longitudinal comparisons within individual subjects, including of normal and diseased tissues. Using this approach we tracked CA1 hippocampal pyramidal neuron dendrites in adult mice, revealing these dendrites' extreme stability (>8,000 day mean lifetime) and rare examples of their structural alterations. To illustrate disease studies, we tracked deep lying gliomas by observing tumor growth, visualizing three-dimensional vasculature structure, and determining microcirculatory speeds. Average erythrocyte speeds in gliomas declined markedly as the disease advanced, notwithstanding significant increases in capillary diameters. Time-lapse microendoscopy will be applicable to studies of numerous disorders, including neurovascular, neurological, cancerous, and trauma-induced conditions. PMID:21240263

Computational ghost imaging using deep learning

NASA Astrophysics Data System (ADS)

Shimobaba, Tomoyoshi; Endo, Yutaka; Nishitsuji, Takashi; Takahashi, Takayuki; Nagahama, Yuki; Hasegawa, Satoki; Sano, Marie; Hirayama, Ryuji; Kakue, Takashi; Shiraki, Atsushi; Ito, Tomoyoshi

2018-04-01

Computational ghost imaging (CGI) is a single-pixel imaging technique that exploits the correlation between known random patterns and the measured intensity of light transmitted (or reflected) by an object. Although CGI can obtain two- or three-dimensional images with a single or a few bucket detectors, the quality of the reconstructed images is reduced by noise due to the reconstruction of images from random patterns. In this study, we improve the quality of CGI images using deep learning. A deep neural network is used to automatically learn the features of noise-contaminated CGI images. After training, the network is able to predict low-noise images from new noise-contaminated CGI images.

Identifying mitosis deep in tissue using dynamic light scattering fluctuation spectroscopy

NASA Astrophysics Data System (ADS)

An, Ran; Jeong, Kwan; Turek, John; Nolte, David

2012-03-01

In the cell cycle, mitosis is the most dramatic phase, especially in Telophase and Cytokinesis. For single cells and cell monolayer, there are precise microscopic studies of mitosis, while for 3-D tissue such as tumor spheroids the light signal is obscured by the high background of diffusely scattered light. Therefore, the mitosis phase cannot be detected deep inside 3-D tissue using conventional microscopic techniques. In this work, we detect mitosis in living tissue using Tissue Dynamic Imaging (TDI). We trace depth-gated dynamic speckles from a tumor spheroid (up to 1mm in diameter) using coherence-gated digital holography imaging. Frequency-versus-time spectrograms depend on specific types of perturbation such as cell shape change, membrane undulation and cell organelles movements. By using these spectral responses as functional finger prints, we can identify mitosis events from different voxels at a specified depth inside tumor spheroids. By performing B-scans of the tumor spheroid, we generate 3-D mitosis maps (or movies) for the entire tumor spheroids. We show that for healthy tumor spheroids, the mitosis events only happen within the proliferating shell. We also compare results when anti-cancer drugs are applied to arrest, release and synchronize mitosis. This shows the application of TDI for drug screening. The technique can identify and monitor complex motilities inside 3-D tissue with a strong potential for drug diagnosis and developmental biology studies.

A simultaneous multimodal imaging system for tissue functional parameters

NASA Astrophysics Data System (ADS)

Ren, Wenqi; Zhang, Zhiwu; Wu, Qiang; Zhang, Shiwu; Xu, Ronald

2014-02-01

Simultaneous and quantitative assessment of skin functional characteristics in different modalities will facilitate diagnosis and therapy in many clinical applications such as wound healing. However, many existing clinical practices and multimodal imaging systems are subjective, qualitative, sequential for multimodal data collection, and need co-registration between different modalities. To overcome these limitations, we developed a multimodal imaging system for quantitative, non-invasive, and simultaneous imaging of cutaneous tissue oxygenation and blood perfusion parameters. The imaging system integrated multispectral and laser speckle imaging technologies into one experimental setup. A Labview interface was developed for equipment control, synchronization, and image acquisition. Advanced algorithms based on a wide gap second derivative reflectometry and laser speckle contrast analysis (LASCA) were developed for accurate reconstruction of tissue oxygenation and blood perfusion respectively. Quantitative calibration experiments and a new style of skinsimulating phantom were designed to verify the accuracy and reliability of the imaging system. The experimental results were compared with a Moor tissue oxygenation and perfusion monitor. For In vivo testing, a post-occlusion reactive hyperemia (PORH) procedure in human subject and an ongoing wound healing monitoring experiment using dorsal skinfold chamber models were conducted to validate the usability of our system for dynamic detection of oxygenation and perfusion parameters. In this study, we have not only setup an advanced multimodal imaging system for cutaneous tissue oxygenation and perfusion parameters but also elucidated its potential for wound healing assessment in clinical practice.

Deep Learning Nuclei Detection in Digitized Histology Images by Superpixels.

PubMed

Sornapudi, Sudhir; Stanley, Ronald Joe; Stoecker, William V; Almubarak, Haidar; Long, Rodney; Antani, Sameer; Thoma, George; Zuna, Rosemary; Frazier, Shelliane R

2018-01-01

Advances in image analysis and computational techniques have facilitated automatic detection of critical features in histopathology images. Detection of nuclei is critical for squamous epithelium cervical intraepithelial neoplasia (CIN) classification into normal, CIN1, CIN2, and CIN3 grades. In this study, a deep learning (DL)-based nuclei segmentation approach is investigated based on gathering localized information through the generation of superpixels using a simple linear iterative clustering algorithm and training with a convolutional neural network. The proposed approach was evaluated on a dataset of 133 digitized histology images and achieved an overall nuclei detection (object-based) accuracy of 95.97%, with demonstrated improvement over imaging-based and clustering-based benchmark techniques. The proposed DL-based nuclei segmentation Method with superpixel analysis has shown improved segmentation results in comparison to state-of-the-art methods.

Deep Learning Nuclei Detection in Digitized Histology Images by Superpixels

PubMed Central

Sornapudi, Sudhir; Stanley, Ronald Joe; Stoecker, William V.; Almubarak, Haidar; Long, Rodney; Antani, Sameer; Thoma, George; Zuna, Rosemary; Frazier, Shelliane R.

2018-01-01

Background: Advances in image analysis and computational techniques have facilitated automatic detection of critical features in histopathology images. Detection of nuclei is critical for squamous epithelium cervical intraepithelial neoplasia (CIN) classification into normal, CIN1, CIN2, and CIN3 grades. Methods: In this study, a deep learning (DL)-based nuclei segmentation approach is investigated based on gathering localized information through the generation of superpixels using a simple linear iterative clustering algorithm and training with a convolutional neural network. Results: The proposed approach was evaluated on a dataset of 133 digitized histology images and achieved an overall nuclei detection (object-based) accuracy of 95.97%, with demonstrated improvement over imaging-based and clustering-based benchmark techniques. Conclusions: The proposed DL-based nuclei segmentation Method with superpixel analysis has shown improved segmentation results in comparison to state-of-the-art methods. PMID:29619277

The effect of adjunctive noncontact low frequency ultrasound on deep tissue pressure injury.

PubMed

Honaker, Jeremy S; Forston, Michael R; Davis, Emily A; Weisner, Michelle M; Morgan, Jennifer A; Sacca, Emily

2016-11-01

The optimal treatment for deep tissue pressure injuries has not been determined. Deep tissue pressure injuries represent a more ominous early stage pressure injury that may evolve into full thickness ulceration despite implementing the standard of care for pressure injury. A longitudinal prospective historical case control study design was used to determine the effectiveness of noncontact low frequency ultrasound plus standard of care (treatment group) in comparison to standard of care (control group) in reducing deep tissue pressure injury severity, total surface area, and final pressure injury stage. The Honaker Suspected Deep Tissue Injury Severity Scale (range 3-18[more severe]) was used to determine deep tissue pressure injury severity at enrollment (Time 1) and discharge (Time 2). A total of 60 subjects (Treatment = 30; Control= 30) were enrolled in the study. In comparison to the control group mean deep tissue pressure injury total surface area change at Time 2 (0.3 cm 2 ), the treatment group had a greater decrease (8.8 cm 2 ) that was significant (t = 2.41, p = 0.014, r 2  = 0.10). In regards to the Honaker Suspected Deep Tissue Injury Severity Scale scores, the treatment group had a significantly lower score (7.6) in comparison to the control group (11.9) at time 2, with a mean difference of 4.6 (t = 6.146, p = 0.0001, r 2  = 0.39). When considering the final pressure ulcer stage at Time 2, the control group were mostly composed of unstageable pressure ulcer (57%) and deep tissue pressure injury severity (27%). In contrast, the treatment group final pressure ulcer stages were less severe and were mostly composed of stage 2 pressure injury (50%) and deep tissue pressure injury severity (23%) were the most common at time 2. The results of this study have shown that deep tissue pressure injury severity treated with noncontact low frequency ultrasound within 5 days of onset and in conjunction with standard of care may improve

Hello World Deep Learning in Medical Imaging.

PubMed

Lakhani, Paras; Gray, Daniel L; Pett, Carl R; Nagy, Paul; Shih, George

2018-05-03

There is recent popularity in applying machine learning to medical imaging, notably deep learning, which has achieved state-of-the-art performance in image analysis and processing. The rapid adoption of deep learning may be attributed to the availability of machine learning frameworks and libraries to simplify their use. In this tutorial, we provide a high-level overview of how to build a deep neural network for medical image classification, and provide code that can help those new to the field begin their informatics projects.

Biodynamic imaging for phenotypic profiling of three-dimensional tissue culture

PubMed Central

Sun, Hao; Merrill, Daniel; An, Ran; Turek, John; Matei, Daniela; Nolte, David D.

2017-01-01

Abstract. Three-dimensional (3-D) tissue culture represents a more biologically relevant environment for testing new drugs compared to conventional two-dimensional cancer cell culture models. Biodynamic imaging is a high-content 3-D optical imaging technology based on low-coherence interferometry and digital holography that uses dynamic speckle as high-content image contrast to probe deep inside 3-D tissue. Speckle contrast is shown to be a scaling function of the acquisition time relative to the persistence time of intracellular transport and hence provides a measure of cellular activity. Cellular responses of 3-D multicellular spheroids to paclitaxel are compared among three different growth techniques: rotating bioreactor (BR), hanging-drop (HD), and nonadherent (U-bottom, UB) plate spheroids, compared with ex vivo living tissues. HD spheroids have the most homogeneous tissue, whereas BR spheroids display large sample-to-sample variability as well as spatial heterogeneity. The responses of BR-grown tumor spheroids to paclitaxel are more similar to those of ex vivo biopsies than the responses of spheroids grown using HD or plate methods. The rate of mitosis inhibition by application of taxol is measured through tissue dynamics spectroscopic imaging, demonstrating the ability to monitor antimitotic chemotherapy. These results illustrate the potential use of low-coherence digital holography for 3-D pharmaceutical screening applications. PMID:28301634

Autofluorescence-based diagnostic UV imaging of tissues and cells

NASA Astrophysics Data System (ADS)

Renkoski, Timothy E.

Cancer is the second leading cause of death in the United States, and its early diagnosis is critical to improving treatment options and patient outcomes. In autofluorescence (AF) imaging, light of controlled wavelengths is projected onto tissue, absorbed by specific molecules, and re-emitted at longer wavelengths. Images of re-emitted light are used together with spectral information to infer tissue functional information and diagnosis. This dissertation describes AF imaging studies of three different organs using data collected from fresh human surgical specimens. In the ovary study, illumination was at 365 nm, and images were captured at 8 emission wavelengths. Measurements from a multispectral imaging system and fiber optic probe were used to map tissue diagnosis at every image pixel. For the colon and pancreas studies, instrumentation was developed extending AF imaging capability to sub-300 nm excitation. Images excited in the deep UV revealed tryptophan and protein content which are believed to change with disease state. Several excitation wavelength bands from 280 nm to 440 nm were investigated. Microscopic AF images collected in the pancreas study included both cultured and primary cells. Several findings are reported. A method of transforming fiber optic probe spectra for direct comparison with imager spectra was devised. Normalization of AF data by green reflectance data was found useful in correcting hemoglobin absorption. Ratio images, both AF and reflectance, were formulated to highlight growths in the colon. Novel tryptophan AF images were found less useful for colon diagnostics than the new ratio techniques. Microscopic tryptophan AF images produce useful visualization of cellular protein content, but their diagnostic value requires further study.

In vivo three-photon imaging of deep cerebellum

NASA Astrophysics Data System (ADS)

Wang, Mengran; Wang, Tianyu; Wu, Chunyan; Li, Bo; Ouzounov, Dimitre G.; Sinefeld, David; Guru, Akash; Nam, Hyung-Song; Capecchi, Mario R.; Warden, Melissa R.; Xu, Chris

2018-02-01

We demonstrate three-photon microscopy (3PM) of mouse cerebellum at 1 mm depth by imaging both blood vessels and neurons. We compared 3PM and 2PM in the mouse cerebellum for imaging green (using excitation sources at 1300 nm and 920 nm, respectively) and red fluorescence (using excitation sources at 1680 nm and 1064 nm, respectively). 3PM enabled deeper imaging than 2PM because the use of longer excitation wavelength reduces the scattering in biological tissue and the higher order nonlinear excitation provides better 3D localization. To illustrate these two advantages quantitatively, we measured the signal decay as well as the signal-to-background ratio (SBR) as a function of depth. We performed 2-photon imaging from the brain surface all the way down to the area where the SBR reaches 1, while at the same depth, 3PM still has SBR above 30. The segmented decay curve shows that the mouse cerebellum has different effective attenuation lengths at different depths, indicating heterogeneous tissue property for this brain region. We compared the third harmonic generation (THG) signal, which is used to visualize myelinated fibers, with the decay curve. We found that the regions with shorter effective attenuation lengths correspond to the regions with more fibers. Our results indicate that the widespread, non-uniformly distributed myelinated fibers adds heterogeneity to mouse cerebellum, which poses additional challenges in deep imaging of this brain region.

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.

Deep Tissue Photoacoustic Imaging Using a Miniaturized 2-D Capacitive Micromachined Ultrasonic Transducer Array

PubMed Central

Kothapalli, Sri-Rajasekhar; Ma, Te-Jen; Vaithilingam, Srikant; Oralkan, Ömer

2014-01-01

In this paper, we demonstrate 3-D photoacoustic imaging (PAI) of light absorbing objects embedded as deep as 5 cm inside strong optically scattering phantoms using a miniaturized (4 mm × 4 mm × 500 µm), 2-D capacitive micromachined ultrasonic transducer (CMUT) array of 16 × 16 elements with a center frequency of 5.5 MHz. Two-dimensional tomographic images and 3-D volumetric images of the objects placed at different depths are presented. In addition, we studied the sensitivity of CMUT-based PAI to the concentration of indocyanine green dye at 5 cm depth inside the phantom. Under optimized experimental conditions, the objects at 5 cm depth can be imaged with SNR of about 35 dB and a spatial resolution of approximately 500 µm. Results demonstrate that CMUTs with integrated front-end amplifier circuits are an attractive choice for achieving relatively high depth sensitivity for PAI. PMID:22249594

Wishart Deep Stacking Network for Fast POLSAR Image Classification.

PubMed

Jiao, Licheng; Liu, Fang

2016-05-11

Inspired by the popular deep learning architecture - Deep Stacking Network (DSN), a specific deep model for polarimetric synthetic aperture radar (POLSAR) image classification is proposed in this paper, which is named as Wishart Deep Stacking Network (W-DSN). First of all, a fast implementation of Wishart distance is achieved by a special linear transformation, which speeds up the classification of POLSAR image and makes it possible to use this polarimetric information in the following Neural Network (NN). Then a single-hidden-layer neural network based on the fast Wishart distance is defined for POLSAR image classification, which is named as Wishart Network (WN) and improves the classification accuracy. Finally, a multi-layer neural network is formed by stacking WNs, which is in fact the proposed deep learning architecture W-DSN for POLSAR image classification and improves the classification accuracy further. In addition, the structure of WN can be expanded in a straightforward way by adding hidden units if necessary, as well as the structure of the W-DSN. As a preliminary exploration on formulating specific deep learning architecture for POLSAR image classification, the proposed methods may establish a simple but clever connection between POLSAR image interpretation and deep learning. The experiment results tested on real POLSAR image show that the fast implementation of Wishart distance is very efficient (a POLSAR image with 768000 pixels can be classified in 0.53s), and both the single-hidden-layer architecture WN and the deep learning architecture W-DSN for POLSAR image classification perform well and work efficiently.

Reconstruction of initial pressure from limited view photoacoustic images using deep learning

NASA Astrophysics Data System (ADS)

Waibel, Dominik; Gröhl, Janek; Isensee, Fabian; Kirchner, Thomas; Maier-Hein, Klaus; Maier-Hein, Lena

2018-02-01

Quantification of tissue properties with photoacoustic (PA) imaging typically requires a highly accurate representation of the initial pressure distribution in tissue. Almost all PA scanners reconstruct the PA image only from a partial scan of the emitted sound waves. Especially handheld devices, which have become increasingly popular due to their versatility and ease of use, only provide limited view data because of their geometry. Owing to such limitations in hardware as well as to the acoustic attenuation in tissue, state-of-the-art reconstruction methods deliver only approximations of the initial pressure distribution. To overcome the limited view problem, we present a machine learning-based approach to the reconstruction of initial pressure from limited view PA data. Our method involves a fully convolutional deep neural network based on a U-Net-like architecture with pixel-wise regression loss on the acquired PA images. It is trained and validated on in silico data generated with Monte Carlo simulations. In an initial study we found an increase in accuracy over the state-of-the-art when reconstructing simulated linear-array scans of blood vessels.

Single myelin fiber imaging in living rodents without labeling by deep optical coherence microscopy

NASA Astrophysics Data System (ADS)

Ben Arous, Juliette; Binding, Jonas; Léger, Jean-François; Casado, Mariano; Topilko, Piotr; Gigan, Sylvain; Claude Boccara, A.; Bourdieu, Laurent

2011-11-01

Myelin sheath disruption is responsible for multiple neuropathies in the central and peripheral nervous system. Myelin imaging has thus become an important diagnosis tool. However, in vivo imaging has been limited to either low-resolution techniques unable to resolve individual fibers or to low-penetration imaging of single fibers, which cannot provide quantitative information about large volumes of tissue, as required for diagnostic purposes. Here, we perform myelin imaging without labeling and at micron-scale resolution with >300-μm penetration depth on living rodents. This was achieved with a prototype [termed deep optical coherence microscopy (deep-OCM)] of a high-numerical aperture infrared full-field optical coherence microscope, which includes aberration correction for the compensation of refractive index mismatch and high-frame-rate interferometric measurements. We were able to measure the density of individual myelinated fibers in the rat cortex over a large volume of gray matter. In the peripheral nervous system, deep-OCM allows, after minor surgery, in situ imaging of single myelinated fibers over a large fraction of the sciatic nerve. This allows quantitative comparison of normal and Krox20 mutant mice, in which myelination in the peripheral nervous system is impaired. This opens promising perspectives for myelin chronic imaging in demyelinating diseases and for minimally invasive medical diagnosis.

Single myelin fiber imaging in living rodents without labeling by deep optical coherence microscopy.

PubMed

Ben Arous, Juliette; Binding, Jonas; Léger, Jean-François; Casado, Mariano; Topilko, Piotr; Gigan, Sylvain; Boccara, A Claude; Bourdieu, Laurent

2011-11-01

Myelin sheath disruption is responsible for multiple neuropathies in the central and peripheral nervous system. Myelin imaging has thus become an important diagnosis tool. However, in vivo imaging has been limited to either low-resolution techniques unable to resolve individual fibers or to low-penetration imaging of single fibers, which cannot provide quantitative information about large volumes of tissue, as required for diagnostic purposes. Here, we perform myelin imaging without labeling and at micron-scale resolution with >300-μm penetration depth on living rodents. This was achieved with a prototype [termed deep optical coherence microscopy (deep-OCM)] of a high-numerical aperture infrared full-field optical coherence microscope, which includes aberration correction for the compensation of refractive index mismatch and high-frame-rate interferometric measurements. We were able to measure the density of individual myelinated fibers in the rat cortex over a large volume of gray matter. In the peripheral nervous system, deep-OCM allows, after minor surgery, in situ imaging of single myelinated fibers over a large fraction of the sciatic nerve. This allows quantitative comparison of normal and Krox20 mutant mice, in which myelination in the peripheral nervous system is impaired. This opens promising perspectives for myelin chronic imaging in demyelinating diseases and for minimally invasive medical diagnosis.

Thermalnet: a Deep Convolutional Network for Synthetic Thermal Image Generation

NASA Astrophysics Data System (ADS)

Kniaz, V. V.; Gorbatsevich, V. S.; Mizginov, V. A.

2017-05-01

Deep convolutional neural networks have dramatically changed the landscape of the modern computer vision. Nowadays methods based on deep neural networks show the best performance among image recognition and object detection algorithms. While polishing of network architectures received a lot of scholar attention, from the practical point of view the preparation of a large image dataset for a successful training of a neural network became one of major challenges. This challenge is particularly profound for image recognition in wavelengths lying outside the visible spectrum. For example no infrared or radar image datasets large enough for successful training of a deep neural network are available to date in public domain. Recent advances of deep neural networks prove that they are also capable to do arbitrary image transformations such as super-resolution image generation, grayscale image colorisation and imitation of style of a given artist. Thus a natural question arise: how could be deep neural networks used for augmentation of existing large image datasets? This paper is focused on the development of the Thermalnet deep convolutional neural network for augmentation of existing large visible image datasets with synthetic thermal images. The Thermalnet network architecture is inspired by colorisation deep neural networks.

Automated Grading of Gliomas using Deep Learning in Digital Pathology Images: A modular approach with ensemble of convolutional neural networks.

PubMed

Ertosun, Mehmet Günhan; Rubin, Daniel L

2015-01-01

Brain glioma is the most common primary malignant brain tumors in adults with different pathologic subtypes: Lower Grade Glioma (LGG) Grade II, Lower Grade Glioma (LGG) Grade III, and Glioblastoma Multiforme (GBM) Grade IV. The survival and treatment options are highly dependent of this glioma grade. We propose a deep learning-based, modular classification pipeline for automated grading of gliomas using digital pathology images. Whole tissue digitized images of pathology slides obtained from The Cancer Genome Atlas (TCGA) were used to train our deep learning modules. Our modular pipeline provides diagnostic quality statistics, such as precision, sensitivity and specificity, of the individual deep learning modules, and (1) facilitates training given the limited data in this domain, (2) enables exploration of different deep learning structures for each module, (3) leads to developing less complex modules that are simpler to analyze, and (4) provides flexibility, permitting use of single modules within the framework or use of other modeling or machine learning applications, such as probabilistic graphical models or support vector machines. Our modular approach helps us meet the requirements of minimum accuracy levels that are demanded by the context of different decision points within a multi-class classification scheme. Convolutional Neural Networks are trained for each module for each sub-task with more than 90% classification accuracies on validation data set, and achieved classification accuracy of 96% for the task of GBM vs LGG classification, 71% for further identifying the grade of LGG into Grade II or Grade III on independent data set coming from new patients from the multi-institutional repository.

Automated Grading of Gliomas using Deep Learning in Digital Pathology Images: A modular approach with ensemble of convolutional neural networks

PubMed Central

Ertosun, Mehmet Günhan; Rubin, Daniel L.

2015-01-01

Brain glioma is the most common primary malignant brain tumors in adults with different pathologic subtypes: Lower Grade Glioma (LGG) Grade II, Lower Grade Glioma (LGG) Grade III, and Glioblastoma Multiforme (GBM) Grade IV. The survival and treatment options are highly dependent of this glioma grade. We propose a deep learning-based, modular classification pipeline for automated grading of gliomas using digital pathology images. Whole tissue digitized images of pathology slides obtained from The Cancer Genome Atlas (TCGA) were used to train our deep learning modules. Our modular pipeline provides diagnostic quality statistics, such as precision, sensitivity and specificity, of the individual deep learning modules, and (1) facilitates training given the limited data in this domain, (2) enables exploration of different deep learning structures for each module, (3) leads to developing less complex modules that are simpler to analyze, and (4) provides flexibility, permitting use of single modules within the framework or use of other modeling or machine learning applications, such as probabilistic graphical models or support vector machines. Our modular approach helps us meet the requirements of minimum accuracy levels that are demanded by the context of different decision points within a multi-class classification scheme. Convolutional Neural Networks are trained for each module for each sub-task with more than 90% classification accuracies on validation data set, and achieved classification accuracy of 96% for the task of GBM vs LGG classification, 71% for further identifying the grade of LGG into Grade II or Grade III on independent data set coming from new patients from the multi-institutional repository. PMID:26958289

End-to-end deep neural network for optical inversion in quantitative photoacoustic imaging.

PubMed

Cai, Chuangjian; Deng, Kexin; Ma, Cheng; Luo, Jianwen

2018-06-15

An end-to-end deep neural network, ResU-net, is developed for quantitative photoacoustic imaging. A residual learning framework is used to facilitate optimization and to gain better accuracy from considerably increased network depth. The contracting and expanding paths enable ResU-net to extract comprehensive context information from multispectral initial pressure images and, subsequently, to infer a quantitative image of chromophore concentration or oxygen saturation (sO 2 ). According to our numerical experiments, the estimations of sO 2 and indocyanine green concentration are accurate and robust against variations in both optical property and object geometry. An extremely short reconstruction time of 22 ms is achieved.

Deep Undercooling of Tissue Water and Winter Hardiness Limitations in Timberline Flora 1

PubMed Central

Becwar, Michael R.; Rajashekar, Channa; Bristow, Katherine J. Hansen; Burke, Michael J.

1981-01-01

Deep undercooled tissue water, which froze near −40 C, was found in winter collected stem and leaf tissue of the dominant timberline tree species of the Colorado Rocky Mountains, Engelmann spruce (Picea engelmannii (Parry) Engelm.) and subalpine fir (Abies lasiocarpa (Hook.) Nutt.), and in numerous other woody species in and below the subalpine vegetation zone. Previous work on numerous woody plants indicates that deep undercooling in xylem makes probable a −40 C winter hardiness limit in stem tissue. Visual injury determinations and electrolyte loss measurements on stem tissue revealed injury near −40 C associated with the freezing of the deep undercooled stem tissue water. These results suggest that the winter hardiness limit of this woody flora is near −40 C. The relevance of deep undercooling in relation to timberline, the upper elevational limit of the subalpine forest, is discussed. PMID:16661852

Do Facilitated Online Dual Credit Classes Result in Deep Learning?

ERIC Educational Resources Information Center

Stark Education Partnership, 2015

2015-01-01

This study, with funding from the Jennings Foundation, sought to answer the following broad research question: Do facilitated online dual credit courses result in deep learning? The answer to this question is key to addressing barriers many students face in bridging from high school to college. This report includes a descriptive case study that…

Wireless power transfer to deep-tissue microimplants

PubMed Central

Yeh, Alexander J.; Neofytou, Evgenios; Kim, Sanghoek; Tanabe, Yuji; Patlolla, Bhagat; Beygui, Ramin E.; Poon, Ada S. Y.

2014-01-01

The ability to implant electronic systems in the human body has led to many medical advances. Progress in semiconductor technology paved the way for devices at the scale of a millimeter or less (“microimplants”), but the miniaturization of the power source remains challenging. Although wireless powering has been demonstrated, energy transfer beyond superficial depths in tissue has so far been limited by large coils (at least a centimeter in diameter) unsuitable for a microimplant. Here, we show that this limitation can be overcome by a method, termed midfield powering, to create a high-energy density region deep in tissue inside of which the power-harvesting structure can be made extremely small. Unlike conventional near-field (inductively coupled) coils, for which coupling is limited by exponential field decay, a patterned metal plate is used to induce spatially confined and adaptive energy transport through propagating modes in tissue. We use this method to power a microimplant (2 mm, 70 mg) capable of closed-chest wireless control of the heart that is orders of magnitude smaller than conventional pacemakers. With exposure levels below human safety thresholds, milliwatt levels of power can be transferred to a deep-tissue (>5 cm) microimplant for both complex electronic function and physiological stimulation. The approach developed here should enable new generations of implantable systems that can be integrated into the body at minimal cost and risk. PMID:24843161

Wireless power transfer to deep-tissue microimplants.

PubMed

Ho, John S; Yeh, Alexander J; Neofytou, Evgenios; Kim, Sanghoek; Tanabe, Yuji; Patlolla, Bhagat; Beygui, Ramin E; Poon, Ada S Y

2014-06-03

The ability to implant electronic systems in the human body has led to many medical advances. Progress in semiconductor technology paved the way for devices at the scale of a millimeter or less ("microimplants"), but the miniaturization of the power source remains challenging. Although wireless powering has been demonstrated, energy transfer beyond superficial depths in tissue has so far been limited by large coils (at least a centimeter in diameter) unsuitable for a microimplant. Here, we show that this limitation can be overcome by a method, termed midfield powering, to create a high-energy density region deep in tissue inside of which the power-harvesting structure can be made extremely small. Unlike conventional near-field (inductively coupled) coils, for which coupling is limited by exponential field decay, a patterned metal plate is used to induce spatially confined and adaptive energy transport through propagating modes in tissue. We use this method to power a microimplant (2 mm, 70 mg) capable of closed-chest wireless control of the heart that is orders of magnitude smaller than conventional pacemakers. With exposure levels below human safety thresholds, milliwatt levels of power can be transferred to a deep-tissue (>5 cm) microimplant for both complex electronic function and physiological stimulation. The approach developed here should enable new generations of implantable systems that can be integrated into the body at minimal cost and risk.

Quantitative characterization of viscoelastic behavior in tissue-mimicking phantoms and ex vivo animal tissues.

PubMed

Maccabi, Ashkan; Shin, Andrew; Namiri, Nikan K; Bajwa, Neha; St John, Maie; Taylor, Zachary D; Grundfest, Warren; Saddik, George N

2018-01-01

Viscoelasticity of soft tissue is often related to pathology, and therefore, has become an important diagnostic indicator in the clinical assessment of suspect tissue. Surgeons, particularly within head and neck subsites, typically use palpation techniques for intra-operative tumor detection. This detection method, however, is highly subjective and often fails to detect small or deep abnormalities. Vibroacoustography (VA) and similar methods have previously been used to distinguish tissue with high-contrast, but a firm understanding of the main contrast mechanism has yet to be verified. The contributions of tissue mechanical properties in VA images have been difficult to verify given the limited literature on viscoelastic properties of various normal and diseased tissue. This paper aims to investigate viscoelasticity theory and present a detailed description of viscoelastic experimental results obtained in tissue-mimicking phantoms (TMPs) and ex vivo tissues to verify the main contrast mechanism in VA and similar imaging modalities. A spherical-tip micro-indentation technique was employed with the Hertzian model to acquire absolute, quantitative, point measurements of the elastic modulus (E), long term shear modulus (η), and time constant (τ) in homogeneous TMPs and ex vivo tissue in rat liver and porcine liver and gallbladder. Viscoelastic differences observed between porcine liver and gallbladder tissue suggest that imaging modalities which utilize the mechanical properties of tissue as a primary contrast mechanism can potentially be used to quantitatively differentiate between proximate organs in a clinical setting. These results may facilitate more accurate tissue modeling and add information not currently available to the field of systems characterization and biomedical research.

Quantitative characterization of viscoelastic behavior in tissue-mimicking phantoms and ex vivo animal tissues

PubMed Central

Shin, Andrew; Namiri, Nikan K.; Bajwa, Neha; St. John, Maie; Taylor, Zachary D.; Grundfest, Warren; Saddik, George N.

2018-01-01

Viscoelasticity of soft tissue is often related to pathology, and therefore, has become an important diagnostic indicator in the clinical assessment of suspect tissue. Surgeons, particularly within head and neck subsites, typically use palpation techniques for intra-operative tumor detection. This detection method, however, is highly subjective and often fails to detect small or deep abnormalities. Vibroacoustography (VA) and similar methods have previously been used to distinguish tissue with high-contrast, but a firm understanding of the main contrast mechanism has yet to be verified. The contributions of tissue mechanical properties in VA images have been difficult to verify given the limited literature on viscoelastic properties of various normal and diseased tissue. This paper aims to investigate viscoelasticity theory and present a detailed description of viscoelastic experimental results obtained in tissue-mimicking phantoms (TMPs) and ex vivo tissues to verify the main contrast mechanism in VA and similar imaging modalities. A spherical-tip micro-indentation technique was employed with the Hertzian model to acquire absolute, quantitative, point measurements of the elastic modulus (E), long term shear modulus (η), and time constant (τ) in homogeneous TMPs and ex vivo tissue in rat liver and porcine liver and gallbladder. Viscoelastic differences observed between porcine liver and gallbladder tissue suggest that imaging modalities which utilize the mechanical properties of tissue as a primary contrast mechanism can potentially be used to quantitatively differentiate between proximate organs in a clinical setting. These results may facilitate more accurate tissue modeling and add information not currently available to the field of systems characterization and biomedical research. PMID:29373598

Deep convolutional neural networks for classifying head and neck cancer using hyperspectral imaging

NASA Astrophysics Data System (ADS)

Halicek, Martin; Lu, Guolan; Little, James V.; Wang, Xu; Patel, Mihir; Griffith, Christopher C.; El-Deiry, Mark W.; Chen, Amy Y.; Fei, Baowei

2017-06-01

Surgical cancer resection requires an accurate and timely diagnosis of the cancer margins in order to achieve successful patient remission. Hyperspectral imaging (HSI) has emerged as a useful, noncontact technique for acquiring spectral and optical properties of tissue. A convolutional neural network (CNN) classifier is developed to classify excised, squamous-cell carcinoma, thyroid cancer, and normal head and neck tissue samples using HSI. The CNN classification was validated by the manual annotation of a pathologist specialized in head and neck cancer. The preliminary results of 50 patients indicate the potential of HSI and deep learning for automatic tissue-labeling of surgical specimens of head and neck patients.

Quicksilver: Fast predictive image registration - A deep learning approach.

PubMed

Yang, Xiao; Kwitt, Roland; Styner, Martin; Niethammer, Marc

2017-09-01

This paper introduces Quicksilver, a fast deformable image registration method. Quicksilver registration for image-pairs works by patch-wise prediction of a deformation model based directly on image appearance. A deep encoder-decoder network is used as the prediction model. While the prediction strategy is general, we focus on predictions for the Large Deformation Diffeomorphic Metric Mapping (LDDMM) model. Specifically, we predict the momentum-parameterization of LDDMM, which facilitates a patch-wise prediction strategy while maintaining the theoretical properties of LDDMM, such as guaranteed diffeomorphic mappings for sufficiently strong regularization. We also provide a probabilistic version of our prediction network which can be sampled during the testing time to calculate uncertainties in the predicted deformations. Finally, we introduce a new correction network which greatly increases the prediction accuracy of an already existing prediction network. We show experimental results for uni-modal atlas-to-image as well as uni-/multi-modal image-to-image registrations. These experiments demonstrate that our method accurately predicts registrations obtained by numerical optimization, is very fast, achieves state-of-the-art registration results on four standard validation datasets, and can jointly learn an image similarity measure. Quicksilver is freely available as an open-source software. Copyright © 2017 Elsevier Inc. All rights reserved.

Groth Deep Locations Image

NASA Image and Video Library

2003-07-25

NASA's Galaxy Evolution Explorer photographed this ultraviolet color blowup of the Groth Deep Image on June 22 and June 23, 2003. Hundreds of galaxies are detected in this portion of the image, and the faint red galaxies are believed to be 6 billion light years away. The white boxes show the location of these distant galaxies, of which more than a 100 can be detected in this image. NASA astronomers expect to detect 10,000 such galaxies after extrapolating to the full image at a deeper exposure level. http://photojournal.jpl.nasa.gov/catalog/PIA04626

Common-path biodynamic imaging for dynamic fluctuation spectroscopy of 3D living tissue

NASA Astrophysics Data System (ADS)

Li, Zhe; Turek, John; Nolte, David D.

2017-03-01

Biodynamic imaging is a novel 3D optical imaging technology based on short-coherence digital holography that measures intracellular motions of cells inside their natural microenvironments. Here both common-path and Mach-Zehnder designs are presented. Biological tissues such as tumor spheroids and ex vivo biopsies are used as targets, and backscattered light is collected as signal. Drugs are applied to samples, and their effects are evaluated by identifying biomarkers that capture intracellular dynamics from the reconstructed holograms. Through digital holography and coherence gating, information from different depths of the samples can be extracted, enabling the deep-tissue measurement of the responses to drugs.

Frequency-locked pulse sequencer for high-frame-rate monochromatic tissue motion imaging.

PubMed

Azar, Reza Zahiri; Baghani, Ali; Salcudean, Septimiu E; Rohling, Robert

2011-04-01

To overcome the inherent low frame rate of conventional ultrasound, we have previously presented a system that can be implemented on conventional ultrasound scanners for high-frame-rate imaging of monochromatic tissue motion. The system employs a sector subdivision technique in the sequencer to increase the acquisition rate. To eliminate the delays introduced during data acquisition, a motion phase correction algorithm has also been introduced to create in-phase displacement images. Previous experimental results from tissue- mimicking phantoms showed that the system can achieve effective frame rates of up to a few kilohertz on conventional ultrasound systems. In this short communication, we present a new pulse sequencing strategy that facilitates high-frame-rate imaging of monochromatic motion such that the acquired echo signals are inherently in-phase. The sequencer uses the knowledge of the excitation frequency to synchronize the acquisition of the entire imaging plane to that of an external exciter. This sequencing approach eliminates any need for synchronization or phase correction and has applications in tissue elastography, which we demonstrate with tissue-mimicking phantoms. © 2011 IEEE

Deep learning classifier with optical coherence tomography images for early dental caries detection

NASA Astrophysics Data System (ADS)

Karimian, Nima; Salehi, Hassan S.; Mahdian, Mina; Alnajjar, Hisham; Tadinada, Aditya

2018-02-01

Dental caries is a microbial disease that results in localized dissolution of the mineral content of dental tissue. Despite considerable decline in the incidence of dental caries, it remains a major health problem in many societies. Early detection of incipient lesions at initial stages of demineralization can result in the implementation of non-surgical preventive approaches to reverse the demineralization process. In this paper, we present a novel approach combining deep convolutional neural networks (CNN) and optical coherence tomography (OCT) imaging modality for classification of human oral tissues to detect early dental caries. OCT images of oral tissues with various densities were input to a CNN classifier to determine variations in tissue densities resembling the demineralization process. The CNN automatically learns a hierarchy of increasingly complex features and a related classifier directly from training data sets. The initial CNN layer parameters were randomly selected. The training set is split into minibatches, with 10 OCT images per batch. Given a batch of training patches, the CNN employs two convolutional and pooling layers to extract features and then classify each patch based on the probabilities from the SoftMax classification layer (output-layer). Afterward, the CNN calculates the error between the classification result and the reference label, and then utilizes the backpropagation process to fine-tune all the layer parameters to minimize this error using batch gradient descent algorithm. We validated our proposed technique on ex-vivo OCT images of human oral tissues (enamel, cortical-bone, trabecular-bone, muscular-tissue, and fatty-tissue), which attested to effectiveness of our proposed method.

Transurethral illumination probe design for deep photoacoustic imaging of prostate

NASA Astrophysics Data System (ADS)

Ai, Min; Salcudean, Tim; Rohling, Robert; Abolmaesumi, Purang; Tang, Shuo

2018-02-01

Photoacoustic (PA) imaging with internal light illumination through optical fiber could enable imaging of internal organs at deep penetration. We have developed a transurethral probe with a multimode fiber inserted in a rigid cystoscope sheath for illuminating the prostate. At the distal end, the fiber tip is processed to diffuse light circumferentially over 2 cm length. A parabolic cylinder mirror then reflects the light to form a rectangular-shaped parallel beam which has at least 1 cm2 at the probe surface. The relatively large rectangular beam size can reduce the laser fluence rate on the urethral wall and thus reduce the potential of tissue damage. A 3 cm optical penetration in chicken tissue is achieved at a fluence rate around 7 mJ/cm2 . For further validation, a prostate phantom was built with similar optical properties of the human prostate. A 1.5 cm penetration depth is achieved in the prostate mimicking phantom at 10 mJ/cm2 fluence rate. PA imaging of prostate can potentially be carried out in the future by combining a transrectal ultrasound transducer and the transurethral illumination.

Stable microwave radiometry system for long term monitoring of deep tissue temperature

NASA Astrophysics Data System (ADS)

Stauffer, Paul R.; Rodriques, Dario B.; Salahi, Sara; Topsakal, Erdem; Oliveira, Tiago R.; Prakash, Aniruddh; D'Isidoro, Fabio; Reudink, Douglas; Snow, Brent W.; Maccarini, Paolo F.

2013-02-01

Background: There are numerous clinical applications for non-invasive monitoring of deep tissue temperature. We present the design and experimental performance of a miniature radiometric thermometry system for measuring volume average temperature of tissue regions located up to 5cm deep in the body. Methods: We constructed a miniature sensor consisting of EMI-shielded log spiral microstrip antenna with high gain onaxis and integrated high-sensitivity 1.35GHz total power radiometer with 500 MHz bandwidth. We tested performance of the radiometry system in both simulated and experimental multilayer phantom models of several intended clinical measurement sites: i) brown adipose tissue (BAT) depots within 2cm of the skin surface, ii) 3-5cm deep kidney, and iii) human brain underlying intact scalp and skull. The physical models included layers of circulating tissue-mimicking liquids controlled at different temperatures to characterize our ability to quantify small changes in target temperature at depth under normothermic surface tissues. Results: We report SAR patterns that characterize the sense region of a 2.6cm diameter receive antenna, and radiometric power measurements as a function of deep tissue temperature that quantify radiometer sensitivity. The data demonstrate: i) our ability to accurately track temperature rise in realistic tissue targets such as urine refluxed from prewarmed bladder into kidney, and 10°C drop in brain temperature underlying normothermic scalp and skull, and ii) long term accuracy and stability of +0.4°C over 4.5 hours as needed for monitoring core body temperature over extended surgery or monitoring effects of brown fat metabolism over an extended sleep/wake cycle. Conclusions: A non-invasive sensor consisting of 2.6cm diameter receive antenna and integral 1.35GHz total power radiometer has demonstrated sufficient sensitivity to track clinically significant changes in temperature of deep tissue targets underlying normothermic surface

Stable Microwave Radiometry System for Long Term Monitoring of Deep Tissue Temperature.

PubMed

Stauffer, Paul R; Rodriques, Dario B; Salahi, Sara; Topsakal, Erdem; Oliveira, Tiago R; Prakash, Aniruddh; D'Isidoro, Fabio; Reudink, Douglas; Snow, Brent W; Maccarini, Paolo F

2013-02-26

There are numerous clinical applications for non-invasive monitoring of deep tissue temperature. We present the design and experimental performance of a miniature radiometric thermometry system for measuring volume average temperature of tissue regions located up to 5cm deep in the body. We constructed a miniature sensor consisting of EMI-shielded log spiral microstrip antenna with high gain on-axis and integrated high-sensitivity 1.35GHz total power radiometer with 500 MHz bandwidth. We tested performance of the radiometry system in both simulated and experimental multilayer phantom models of several intended clinical measurement sites: i) brown adipose tissue (BAT) depots within 2cm of the skin surface, ii) 3-5cm deep kidney, and iii) human brain underlying intact scalp and skull. The physical models included layers of circulating tissue-mimicking liquids controlled at different temperatures to characterize our ability to quantify small changes in target temperature at depth under normothermic surface tissues. We report SAR patterns that characterize the sense region of a 2.6cm diameter receive antenna, and radiometric power measurements as a function of deep tissue temperature that quantify radiometer sensitivity. The data demonstrate: i) our ability to accurately track temperature rise in realistic tissue targets such as urine refluxed from prewarmed bladder into kidney, and 10°C drop in brain temperature underlying normothermic scalp and skull, and ii) long term accuracy and stability of ∓0.4°C over 4.5 hours as needed for monitoring core body temperature over extended surgery or monitoring effects of brown fat metabolism over an extended sleep/wake cycle. A non-invasive sensor consisting of 2.6cm diameter receive antenna and integral 1.35GHz total power radiometer has demonstrated sufficient sensitivity to track clinically significant changes in temperature of deep tissue targets underlying normothermic surface tissues for clinical applications like the

Factors affecting the microbiological condition of the deep tissues of mechanically tenderized beef.

PubMed

Gill, C O; McGinnis, J C

2005-04-01

Whole or halved top butt prime beef cuts were treated in two types of mechanical tenderizing machines that both pierced the meat with thin blades but that used blades of different forms. Aerobes on meat surfaces and in the deep tissues of cuts after treatments were counted. When cuts were treated at a laboratory using a Lumar machine, the contamination of deep tissues increased significantly (P < 0.01) with increasing numbers of aerobic bacteria on meat surfaces and decreased significantly (P < 0.001) with increasing distance from the incised surface. However, contamination did not increase significantly (P > 0.1) with repeated incising of the meat. When halved cuts were incised one or eight times using a commercially cleaned Ross machine at a retail store, the numbers of aerobes recovered from deep tissues were similar with both treatments. When halved cuts were treated in one or other machine, deep tissue contamination was greater with the Lumar machine than with the Ross machine. Contamination of deep tissues as a result of tenderizing by piercing with thin blades can be minimized if the blades are designed to limit the number of bacteria carried into the meat and the microbiological condition of incised surface is well controlled.

Deep tissue volume imaging of birefringence through fibre-optic needle probes for the delineation of breast tumour

NASA Astrophysics Data System (ADS)

Villiger, Martin; Lorenser, Dirk; McLaughlin, Robert A.; Quirk, Bryden C.; Kirk, Rodney W.; Bouma, Brett E.; Sampson, David D.

2016-07-01

Identifying tumour margins during breast-conserving surgeries is a persistent challenge. We have previously developed miniature needle probes that could enable intraoperative volume imaging with optical coherence tomography. In many situations, however, scattering contrast alone is insufficient to clearly identify and delineate malignant regions. Additional polarization-sensitive measurements provide the means to assess birefringence, which is elevated in oriented collagen fibres and may offer an intrinsic biomarker to differentiate tumour from benign tissue. Here, we performed polarization-sensitive optical coherence tomography through miniature imaging needles and developed an algorithm to efficiently reconstruct images of the depth-resolved tissue birefringence free of artefacts. First ex vivo imaging of breast tumour samples revealed excellent contrast between lowly birefringent malignant regions, and stromal tissue, which is rich in oriented collagen and exhibits higher birefringence, as confirmed with co-located histology. The ability to clearly differentiate between tumour and uninvolved stroma based on intrinsic contrast could prove decisive for the intraoperative assessment of tumour margins.

Deep brain two-photon NIR fluorescence imaging for study of Alzheimer's disease

NASA Astrophysics Data System (ADS)

Chen, Congping; Liang, Zhuoyi; Zhou, Biao; Ip, Nancy Y.; Qu, Jianan Y.

2018-02-01

Amyloid depositions in the brain represent the characteristic hallmarks of Alzheimer's disease (AD) pathology. The abnormal accumulation of extracellular amyloid-beta (Aβ) and resulting toxic amyloid plaques are considered to be responsible for the clinical deficits including cognitive decline and memory loss. In vivo two-photon fluorescence imaging of amyloid plaques in live AD mouse model through a chronic imaging window (thinned skull or craniotomy) provides a mean to greatly facilitate the study of the pathological mechanism of AD owing to its high spatial resolution and long-term continuous monitoring. However, the imaging depth for amyloid plaques is largely limited to upper cortical layers due to the short-wavelength fluorescence emission of commonly used amyloid probes. In this work, we reported that CRANAD-3, a near-infrared (NIR) probe for amyloid species with excitation wavelength at 900 nm and emission wavelength around 650 nm, has great advantages over conventionally used probes and is well suited for twophoton deep imaging of amyloid plaques in AD mouse brain. Compared with a commonly used MeO-X04 probe, the imaging depth of CRANAD-3 is largely extended for open skull cranial window. Furthermore, by using two-photon excited fluorescence spectroscopic imaging, we characterized the intrinsic fluorescence of the "aging pigment" lipofuscin in vivo, which has distinct spectra from CRANAD-3 labeled plaques. This study reveals the unique potential of NIR probes for in vivo, high-resolution and deep imaging of brain amyloid in Alzheimer's disease.

Zero-Echo-Time and Dixon Deep Pseudo-CT (ZeDD CT): Direct Generation of Pseudo-CT Images for Pelvic PET/MRI Attenuation Correction Using Deep Convolutional Neural Networks with Multiparametric MRI.

PubMed

Leynes, Andrew P; Yang, Jaewon; Wiesinger, Florian; Kaushik, Sandeep S; Shanbhag, Dattesh D; Seo, Youngho; Hope, Thomas A; Larson, Peder E Z

2018-05-01

Accurate quantification of uptake on PET images depends on accurate attenuation correction in reconstruction. Current MR-based attenuation correction methods for body PET use a fat and water map derived from a 2-echo Dixon MRI sequence in which bone is neglected. Ultrashort-echo-time or zero-echo-time (ZTE) pulse sequences can capture bone information. We propose the use of patient-specific multiparametric MRI consisting of Dixon MRI and proton-density-weighted ZTE MRI to directly synthesize pseudo-CT images with a deep learning model: we call this method ZTE and Dixon deep pseudo-CT (ZeDD CT). Methods: Twenty-six patients were scanned using an integrated 3-T time-of-flight PET/MRI system. Helical CT images of the patients were acquired separately. A deep convolutional neural network was trained to transform ZTE and Dixon MR images into pseudo-CT images. Ten patients were used for model training, and 16 patients were used for evaluation. Bone and soft-tissue lesions were identified, and the SUV max was measured. The root-mean-squared error (RMSE) was used to compare the MR-based attenuation correction with the ground-truth CT attenuation correction. Results: In total, 30 bone lesions and 60 soft-tissue lesions were evaluated. The RMSE in PET quantification was reduced by a factor of 4 for bone lesions (10.24% for Dixon PET and 2.68% for ZeDD PET) and by a factor of 1.5 for soft-tissue lesions (6.24% for Dixon PET and 4.07% for ZeDD PET). Conclusion: ZeDD CT produces natural-looking and quantitatively accurate pseudo-CT images and reduces error in pelvic PET/MRI attenuation correction compared with standard methods. © 2018 by the Society of Nuclear Medicine and Molecular Imaging.

Imaging Strategies for Tissue Engineering Applications

PubMed Central

Nam, Seung Yun; Ricles, Laura M.; Suggs, Laura J.

2015-01-01

Tissue engineering has evolved with multifaceted research being conducted using advanced technologies, and it is progressing toward clinical applications. As tissue engineering technology significantly advances, it proceeds toward increasing sophistication, including nanoscale strategies for material construction and synergetic methods for combining with cells, growth factors, or other macromolecules. Therefore, to assess advanced tissue-engineered constructs, tissue engineers need versatile imaging methods capable of monitoring not only morphological but also functional and molecular information. However, there is no single imaging modality that is suitable for all tissue-engineered constructs. Each imaging method has its own range of applications and provides information based on the specific properties of the imaging technique. Therefore, according to the requirements of the tissue engineering studies, the most appropriate tool should be selected among a variety of imaging modalities. The goal of this review article is to describe available biomedical imaging methods to assess tissue engineering applications and to provide tissue engineers with criteria and insights for determining the best imaging strategies. Commonly used biomedical imaging modalities, including X-ray and computed tomography, positron emission tomography and single photon emission computed tomography, magnetic resonance imaging, ultrasound imaging, optical imaging, and emerging techniques and multimodal imaging, will be discussed, focusing on the latest trends of their applications in recent tissue engineering studies. PMID:25012069

Co-encapsulation of magnetic nanoparticles and doxorubicin into biodegradable microcarriers for deep tissue targeting by vascular MRI navigation.

PubMed

Pouponneau, Pierre; Leroux, Jean-Christophe; Soulez, Gilles; Gaboury, Louis; Martel, Sylvain

2011-05-01

Magnetic tumor targeting with external magnets is a promising method to increase the delivery of cytotoxic agents to tumor cells while reducing side effects. However, this approach suffers from intrinsic limitations, such as the inability to target areas within deep tissues, due mainly to a strong decrease of the magnetic field magnitude away from the magnets. Magnetic resonance navigation (MRN) involving the endovascular steering of therapeutic magnetic microcarriers (TMMC) represents a clinically viable alternative to reach deep tissues. MRN is achieved with an upgraded magnetic resonance imaging (MRI) scanner. In this proof-of-concept preclinical study, the preparation and steering of TMMC which were designed by taking into consideration the constraints of MRN and liver chemoembolization are reported. TMMC were biodegradable microparticles loaded with iron-cobalt nanoparticles and doxorubicin (DOX). These particles displayed high saturation magnetization (Ms = 72 emu g(-1)), MRI tracking compatibility (strong contrast on T2∗-weighted images), appropriate size for the blood vessel embolization (∼50 μm), and sustained release of DOX (over several days). The TMMC were successfully steered in vitro and in vivo in the rabbit model. In vivo targeting of the right or left liver lobes was achieved by MRN through the hepatic artery located 4 cm beneath the skin. Parameters such as flow velocity, TMMC release site in the artery, magnetic gradient and TMMC properties, affected the steering efficiency. These data illustrate the potential of MRN to improve drug targeting in deep tissues. Copyright © 2011 Elsevier Ltd. All rights reserved.

Deep convolutional neural networks for automatic classification of gastric carcinoma using whole slide images in digital histopathology.

PubMed

Sharma, Harshita; Zerbe, Norman; Klempert, Iris; Hellwich, Olaf; Hufnagl, Peter

2017-11-01

Deep learning using convolutional neural networks is an actively emerging field in histological image analysis. This study explores deep learning methods for computer-aided classification in H&E stained histopathological whole slide images of gastric carcinoma. An introductory convolutional neural network architecture is proposed for two computerized applications, namely, cancer classification based on immunohistochemical response and necrosis detection based on the existence of tumor necrosis in the tissue. Classification performance of the developed deep learning approach is quantitatively compared with traditional image analysis methods in digital histopathology requiring prior computation of handcrafted features, such as statistical measures using gray level co-occurrence matrix, Gabor filter-bank responses, LBP histograms, gray histograms, HSV histograms and RGB histograms, followed by random forest machine learning. Additionally, the widely known AlexNet deep convolutional framework is comparatively analyzed for the corresponding classification problems. The proposed convolutional neural network architecture reports favorable results, with an overall classification accuracy of 0.6990 for cancer classification and 0.8144 for necrosis detection. Copyright © 2017 Elsevier Ltd. All rights reserved.

Stokes-polarimetry imaging of tissue

NASA Astrophysics Data System (ADS)

Wu, Paul J.

A novel Stokes-polarimetry imaging system and technique was developed to quantify fully the polarization properties of light remitted from tissue. The uniqueness of the system and technique is established in the incident polarization. Here, the diffuse illumination is varied and controlled with the intention to improve the visibility of tissue structures. Since light retains some polarization even after multiple-scattering events, the polarization of remitted light depends upon the interactions within the material. Differentiation between tissue structures is accomplished by two-dimensional mapping of the imaged area using metrics such as the degree of linear polarization, degree of circular polarization, ellipticity, and Stokes parameters. While Stokes-polarimetry imaging can be applied to a variety of tissues and conditions, this thesis focuses on tissue types associated with the disease endometriosis. The current standard in diagnosing endometriosis is visual laparoscopy with tissue biopsy. The documented correlation between laparoscopy inspection and histological confirmation of suspected lesions was at best 67%. Endometrial lesions vary greatly in their appearance and depth of infiltration. Although laparoscopy permits tissue to be assessed by color and texture, to advance beyond the state-of-the-art, a new imaging modality involving polarized light was investigated; in particular, Stokes-polarimetry imaging was used to determine the polarization signature of light that interacted with tissue. Basic science studies were conducted on rat tails embedded within turbid gelatin. The purpose of these experiments was to determine how identification of sub-surface structures could be improved. Experimental results indicate image contrast among various structures such as tendon, soft tissue and intervertebral discs. Stokes-polarimetry imaging experiments were performed on various tissues associated with endometriosis to obtain a baseline characterization for each

Dual-modality endoscopic probe for tissue surface shape reconstruction and hyperspectral imaging enabled by deep neural networks.

PubMed

Lin, Jianyu; Clancy, Neil T; Qi, Ji; Hu, Yang; Tatla, Taran; Stoyanov, Danail; Maier-Hein, Lena; Elson, Daniel S

2018-06-15

Surgical guidance and decision making could be improved with accurate and real-time measurement of intra-operative data including shape and spectral information of the tissue surface. In this work, a dual-modality endoscopic system has been proposed to enable tissue surface shape reconstruction and hyperspectral imaging (HSI). This system centers around a probe comprised of an incoherent fiber bundle, whose fiber arrangement is different at the two ends, and miniature imaging optics. For 3D reconstruction with structured light (SL), a light pattern formed of randomly distributed spots with different colors is projected onto the tissue surface, creating artificial texture. Pattern decoding with a Convolutional Neural Network (CNN) model and a customized feature descriptor enables real-time 3D surface reconstruction at approximately 12 frames per second (FPS). In HSI mode, spatially sparse hyperspectral signals from the tissue surface can be captured with a slit hyperspectral imager in a single snapshot. A CNN based super-resolution model, namely "super-spectral-resolution" network (SSRNet), has also been developed to estimate pixel-level dense hypercubes from the endoscope cameras standard RGB images and the sparse hyperspectral signals, at approximately 2 FPS. The probe, with a 2.1 mm diameter, enables the system to be used with endoscope working channels. Furthermore, since data acquisition in both modes can be accomplished in one snapshot, operation of this system in clinical applications is minimally affected by tissue surface movement and deformation. The whole apparatus has been validated on phantoms and tissue (ex vivo and in vivo), while initial measurements on patients during laryngeal surgery show its potential in real-world clinical applications. Copyright © 2018 Elsevier B.V. All rights reserved.

Sensory processing of deep tissue nociception in the rat spinal cord and thalamic ventrobasal complex.

PubMed

Sikandar, Shafaq; West, Steven J; McMahon, Stephen B; Bennett, David L; Dickenson, Anthony H

2017-07-01

Sensory processing of deep somatic tissue constitutes an important component of the nociceptive system, yet associated central processing pathways remain poorly understood. Here, we provide a novel electrophysiological characterization and immunohistochemical analysis of neural activation in the lateral spinal nucleus (LSN). These neurons show evoked activity to deep, but not cutaneous, stimulation. The evoked responses of neurons in the LSN can be sensitized to somatosensory stimulation following intramuscular hypertonic saline, an acute model of muscle pain, suggesting this is an important spinal relay site for the processing of deep tissue nociceptive inputs. Neurons of the thalamic ventrobasal complex (VBC) mediate both cutaneous and deep tissue sensory processing, but in contrast to the lateral spinal nucleus our electrophysiological studies do not suggest the existence of a subgroup of cells that selectively process deep tissue inputs. The sensitization of polymodal and thermospecific VBC neurons to mechanical somatosensory stimulation following acute muscle stimulation with hypertonic saline suggests differential roles of thalamic subpopulations in mediating cutaneous and deep tissue nociception in pathological states. Overall, our studies at both the spinal (lateral spinal nucleus) and supraspinal (thalamic ventrobasal complex) levels suggest a convergence of cutaneous and deep somatosensory inputs onto spinothalamic pathways, which are unmasked by activation of muscle nociceptive afferents to produce consequent phenotypic alterations in spinal and thalamic neural coding of somatosensory stimulation. A better understanding of the sensory pathways involved in deep tissue nociception, as well as the degree of labeled line and convergent pathways for cutaneous and deep somatosensory inputs, is fundamental to developing targeted analgesic therapies for deep pain syndromes. © 2017 University College London. Physiological Reports published by Wiley Periodicals

High resolution and deep tissue imaging using a near infrared acoustic resolution photoacoustic microscopy

NASA Astrophysics Data System (ADS)

Moothanchery, Mohesh; Sharma, Arunima; Periyasamy, Vijitha; Pramanik, Manojit

2018-02-01

It is always a great challenge for pure optical techniques to maintain good resolution and imaging depth at the same time. Photoacoustic imaging is an emerging technique which can overcome the limitation by pulsed light illumination and acoustic detection. Here, we report a Near Infrared Acoustic-Resolution Photoacoustic Microscopy (NIR-AR-PAM) systm with 30 MHz transducer and 1064 nm illumination which can achieve a lateral resolution of around 88 μm and imaging depth of 9.2 mm. Compared to visible light NIR beam can penetrate deeper in biological tissue due to weaker optical attenuation. In this work, we also demonstrated the in vivo imaging capabilty of NIRARPAM by near infrared detection of SLN with black ink as exogenous photoacoustic contrast agent in a rodent model.

Optical coherence microscopy for deep tissue imaging of the cerebral cortex with intrinsic contrast

PubMed Central

Srinivasan, Vivek J.; Radhakrishnan, Harsha; Jiang, James Y.; Barry, Scott; Cable, Alex E.

2012-01-01

In vivo optical microscopic imaging techniques have recently emerged as important tools for the study of neurobiological development and pathophysiology. In particular, two-photon microscopy has proved to be a robust and highly flexible method for in vivo imaging in highly scattering tissue. However, two-photon imaging typically requires extrinsic dyes or contrast agents, and imaging depths are limited to a few hundred microns. Here we demonstrate Optical Coherence Microscopy (OCM) for in vivo imaging of neuronal cell bodies and cortical myelination up to depths of ~1.3 mm in the rat neocortex. Imaging does not require the administration of exogenous dyes or contrast agents, and is achieved through intrinsic scattering contrast and image processing alone. Furthermore, using OCM we demonstrate in vivo, quantitative measurements of optical properties (index of refraction and attenuation coefficient) in the cortex, and correlate these properties with laminar cellular architecture determined from the images. Lastly, we show that OCM enables direct visualization of cellular changes during cell depolarization and may therefore provide novel optical markers of cell viability. PMID:22330462

Using deep learning in image hyper spectral segmentation, classification, and detection

NASA Astrophysics Data System (ADS)

Zhao, Xiuying; Su, Zhenyu

2018-02-01

Recent years have shown that deep learning neural networks are a valuable tool in the field of computer vision. Deep learning method can be used in applications like remote sensing such as Land cover Classification, Detection of Vehicle in Satellite Images, Hyper spectral Image classification. This paper addresses the use of the deep learning artificial neural network in Satellite image segmentation. Image segmentation plays an important role in image processing. The hue of the remote sensing image often has a large hue difference, which will result in the poor display of the images in the VR environment. Image segmentation is a pre processing technique applied to the original images and splits the image into many parts which have different hue to unify the color. Several computational models based on supervised, unsupervised, parametric, probabilistic region based image segmentation techniques have been proposed. Recently, one of the machine learning technique known as, deep learning with convolution neural network has been widely used for development of efficient and automatic image segmentation models. In this paper, we focus on study of deep neural convolution network and its variants for automatic image segmentation rather than traditional image segmentation strategies.

Magnetic Resonance Imaging of Human Tissue-Engineered Adipose Substitutes

PubMed Central

Proulx, Maryse; Aubin, Kim; Lagueux, Jean; Audet, Pierre; Auger, Michèle

2015-01-01

Adipose tissue (AT) substitutes are being developed to answer the strong demand in reconstructive surgery. To facilitate the validation of their functional performance in vivo, and to avoid resorting to excessive number of animals, it is crucial at this stage to develop biomedical imaging methodologies, enabling the follow-up of reconstructed AT substitutes. Until now, biomedical imaging of AT substitutes has scarcely been reported in the literature. Therefore, the optimal parameters enabling good resolution, appropriate contrast, and graft delineation, as well as blood perfusion validation, must be studied and reported. In this study, human adipose substitutes produced from adipose-derived stem/stromal cells using the self-assembly approach of tissue engineering were implanted into athymic mice. The fate of the reconstructed AT substitutes implanted in vivo was successfully followed by magnetic resonance imaging (MRI), which is the imaging modality of choice for visualizing soft ATs. T1-weighted images allowed clear delineation of the grafts, followed by volume integration. The magnetic resonance (MR) signal of reconstructed AT was studied in vitro by proton nuclear magnetic resonance (1H-NMR). This confirmed the presence of a strong triglyceride peak of short longitudinal proton relaxation time (T1) values (200±53 ms) in reconstructed AT substitutes (total T1=813±76 ms), which establishes a clear signal difference between adjacent muscle, connective tissue, and native fat (total T1 ∼300 ms). Graft volume retention was followed up to 6 weeks after implantation, revealing a gradual resorption rate averaging at 44% of initial substitute's volume. In addition, vascular perfusion measured by dynamic contrast-enhanced-MRI confirmed the graft's vascularization postimplantation (14 and 21 days after grafting). Histological analysis of the grafted tissues revealed the persistence of numerous adipocytes without evidence of cysts or tissue necrosis. This study

Swept-source optical coherence tomography powered by a 1.3-μm vertical cavity surface emitting laser enables 2.3-mm-deep brain imaging in mice in vivo

NASA Astrophysics Data System (ADS)

Choi, Woo June; Wang, Ruikang K.

2015-10-01

We report noninvasive, in vivo optical imaging deep within a mouse brain by swept-source optical coherence tomography (SS-OCT), enabled by a 1.3-μm vertical cavity surface emitting laser (VCSEL). VCSEL SS-OCT offers a constant signal sensitivity of 105 dB throughout an entire depth of 4.25 mm in air, ensuring an extended usable imaging depth range of more than 2 mm in turbid biological tissue. Using this approach, we show deep brain imaging in mice with an open-skull cranial window preparation, revealing intact mouse brain anatomy from the superficial cerebral cortex to the deep hippocampus. VCSEL SS-OCT would be applicable to small animal studies for the investigation of deep tissue compartments in living brains where diseases such as dementia and tumor can take their toll.

NiftyNet: a deep-learning platform for medical imaging.

PubMed

Gibson, Eli; Li, Wenqi; Sudre, Carole; Fidon, Lucas; Shakir, Dzhoshkun I; Wang, Guotai; Eaton-Rosen, Zach; Gray, Robert; Doel, Tom; Hu, Yipeng; Whyntie, Tom; Nachev, Parashkev; Modat, Marc; Barratt, Dean C; Ourselin, Sébastien; Cardoso, M Jorge; Vercauteren, Tom

2018-05-01

Medical image analysis and computer-assisted intervention problems are increasingly being addressed with deep-learning-based solutions. Established deep-learning platforms are flexible but do not provide specific functionality for medical image analysis and adapting them for this domain of application requires substantial implementation effort. Consequently, there has been substantial duplication of effort and incompatible infrastructure developed across many research groups. This work presents the open-source NiftyNet platform for deep learning in medical imaging. The ambition of NiftyNet is to accelerate and simplify the development of these solutions, and to provide a common mechanism for disseminating research outputs for the community to use, adapt and build upon. The NiftyNet infrastructure provides a modular deep-learning pipeline for a range of medical imaging applications including segmentation, regression, image generation and representation learning applications. Components of the NiftyNet pipeline including data loading, data augmentation, network architectures, loss functions and evaluation metrics are tailored to, and take advantage of, the idiosyncracies of medical image analysis and computer-assisted intervention. NiftyNet is built on the TensorFlow framework and supports features such as TensorBoard visualization of 2D and 3D images and computational graphs by default. We present three illustrative medical image analysis applications built using NiftyNet infrastructure: (1) segmentation of multiple abdominal organs from computed tomography; (2) image regression to predict computed tomography attenuation maps from brain magnetic resonance images; and (3) generation of simulated ultrasound images for specified anatomical poses. The NiftyNet infrastructure enables researchers to rapidly develop and distribute deep learning solutions for segmentation, regression, image generation and representation learning applications, or extend the platform to new

Nonlinear Deep Kernel Learning for Image Annotation.

PubMed

Jiu, Mingyuan; Sahbi, Hichem

2017-02-08

Multiple kernel learning (MKL) is a widely used technique for kernel design. Its principle consists in learning, for a given support vector classifier, the most suitable convex (or sparse) linear combination of standard elementary kernels. However, these combinations are shallow and often powerless to capture the actual similarity between highly semantic data, especially for challenging classification tasks such as image annotation. In this paper, we redefine multiple kernels using deep multi-layer networks. In this new contribution, a deep multiple kernel is recursively defined as a multi-layered combination of nonlinear activation functions, each one involves a combination of several elementary or intermediate kernels, and results into a positive semi-definite deep kernel. We propose four different frameworks in order to learn the weights of these networks: supervised, unsupervised, kernel-based semisupervised and Laplacian-based semi-supervised. When plugged into support vector machines (SVMs), the resulting deep kernel networks show clear gain, compared to several shallow kernels for the task of image annotation. Extensive experiments and analysis on the challenging ImageCLEF photo annotation benchmark, the COREL5k database and the Banana dataset validate the effectiveness of the proposed method.

Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning.

PubMed

Kermany, Daniel S; Goldbaum, Michael; Cai, Wenjia; Valentim, Carolina C S; Liang, Huiying; Baxter, Sally L; McKeown, Alex; Yang, Ge; Wu, Xiaokang; Yan, Fangbing; Dong, Justin; Prasadha, Made K; Pei, Jacqueline; Ting, Magdalene Y L; Zhu, Jie; Li, Christina; Hewett, Sierra; Dong, Jason; Ziyar, Ian; Shi, Alexander; Zhang, Runze; Zheng, Lianghong; Hou, Rui; Shi, William; Fu, Xin; Duan, Yaou; Huu, Viet A N; Wen, Cindy; Zhang, Edward D; Zhang, Charlotte L; Li, Oulan; Wang, Xiaobo; Singer, Michael A; Sun, Xiaodong; Xu, Jie; Tafreshi, Ali; Lewis, M Anthony; Xia, Huimin; Zhang, Kang

2018-02-22

The implementation of clinical-decision support algorithms for medical imaging faces challenges with reliability and interpretability. Here, we establish a diagnostic tool based on a deep-learning framework for the screening of patients with common treatable blinding retinal diseases. Our framework utilizes transfer learning, which trains a neural network with a fraction of the data of conventional approaches. Applying this approach to a dataset of optical coherence tomography images, we demonstrate performance comparable to that of human experts in classifying age-related macular degeneration and diabetic macular edema. We also provide a more transparent and interpretable diagnosis by highlighting the regions recognized by the neural network. We further demonstrate the general applicability of our AI system for diagnosis of pediatric pneumonia using chest X-ray images. This tool may ultimately aid in expediting the diagnosis and referral of these treatable conditions, thereby facilitating earlier treatment, resulting in improved clinical outcomes. VIDEO ABSTRACT. Copyright © 2018 Elsevier Inc. All rights reserved.

Microscopic medical image classification framework via deep learning and shearlet transform.

PubMed

Rezaeilouyeh, Hadi; Mollahosseini, Ali; Mahoor, Mohammad H

2016-10-01

Cancer is the second leading cause of death in US after cardiovascular disease. Image-based computer-aided diagnosis can assist physicians to efficiently diagnose cancers in early stages. Existing computer-aided algorithms use hand-crafted features such as wavelet coefficients, co-occurrence matrix features, and recently, histogram of shearlet coefficients for classification of cancerous tissues and cells in images. These hand-crafted features often lack generalizability since every cancerous tissue and cell has a specific texture, structure, and shape. An alternative approach is to use convolutional neural networks (CNNs) to learn the most appropriate feature abstractions directly from the data and handle the limitations of hand-crafted features. A framework for breast cancer detection and prostate Gleason grading using CNN trained on images along with the magnitude and phase of shearlet coefficients is presented. Particularly, we apply shearlet transform on images and extract the magnitude and phase of shearlet coefficients. Then we feed shearlet features along with the original images to our CNN consisting of multiple layers of convolution, max pooling, and fully connected layers. Our experiments show that using the magnitude and phase of shearlet coefficients as extra information to the network can improve the accuracy of detection and generalize better compared to the state-of-the-art methods that rely on hand-crafted features. This study expands the application of deep neural networks into the field of medical image analysis, which is a difficult domain considering the limited medical data available for such analysis.

Enhanced detection of myeloperoxidase activity in deep tissues through luminescent excitation of near-infrared nanoparticles.

PubMed

Zhang, Ning; Francis, Kevin P; Prakash, Arun; Ansaldi, Daniel

2013-04-01

A previous study reported the use of luminol for the detection of myeloperoxidase (MPO) activity using optical imaging in infiltrating neutrophils under inflammatory disease conditions. The detection is based on a photon-emitting reaction between luminol and an MPO metabolite. Because of tissue absorption and scattering, however, luminol-emitted blue light can be efficiently detected from superficial inflammatory foci only. In this study we report a chemiluminescence resonance energy transfer (CRET) methodology in which luminol-generated blue light excites nanoparticles to emit light in the near-infrared spectral range, resulting in remarkable improvement of MPO detectability in vivo. CRET caused a 37-fold increase in luminescence emission over luminol alone in detecting MPO activity in lung tissues after lipopolysaccharide challenge. We demonstrated a dependence of the chemiluminescent signal on MPO activity using MPO-deficient mice. In addition, co-administration of 4-aminobenzoic acid hydrazide (4-ABAH), an irreversible inhibitor of MPO, significantly attenuated luminescent emission from inflamed lungs. Inhibition of nitric oxide synthase with a nonspecific inhibitor, L-NAME, had no effect on luminol-mediated chemiluminescence production. Pretreatment of mice with MLN120B, a selective inhibitor of IKK-2, resulted in suppression of neutrophil infiltration to the lung tissues and reduction of MPO activity. We also demonstrated that CRET can effectively detect MPO activity at deep tissue tumor foci due to tumor development-associated neutrophil infiltration. We developed a sensitive MPO detection methodology that provides a means for visualizing and quantifying oxidative stress in deep tissue. This method is amenable to rapid evaluation of anti-inflammatory agents in animal models.

Photoacoustic and ultrasound imaging of cancellous bone tissue.

PubMed

Yang, Lifeng; Lashkari, Bahman; Tan, Joel W Y; Mandelis, Andreas

2015-07-01

We used ultrasound (US) and photoacoustic (PA) imaging modalities to characterize cattle trabecular bones. The PA signals were generated with an 805-nm continuous wave laser used for optimally deep optical penetration depth. The detector for both modalities was a 2.25-MHz US transducer with a lateral resolution of ~1 mm at its focal point. Using a lateral pixel size much larger than the size of the trabeculae, raster scanning generated PA images related to the averaged values of the optical and thermoelastic properties, as well as density measurements in the focal volume. US backscatter yielded images related to mechanical properties and density in the focal volume. The depth of interest was selected by time-gating the signals for both modalities. The raster scanned PA and US images were compared with microcomputed tomography (μCT) images averaged over the same volume to generate similar spatial resolution as US and PA. The comparison revealed correlations between PA and US modalities with the mineral volume fraction of the bone tissue. Various features and properties of these modalities such as detectable depth, resolution, and sensitivity are discussed.

Biodynamic Doppler imaging of subcellular motion inside 3D living tissue culture and biopsies (Conference Presentation)

NASA Astrophysics Data System (ADS)

Nolte, David D.

2016-03-01

Biodynamic imaging is an emerging 3D optical imaging technology that probes up to 1 mm deep inside three-dimensional living tissue using short-coherence dynamic light scattering to measure the intracellular motions of cells inside their natural microenvironments. Biodynamic imaging is label-free and non-invasive. The information content of biodynamic imaging is captured through tissue dynamics spectroscopy that displays the changes in the Doppler signatures from intracellular constituents in response to applied compounds. The affected dynamic intracellular mechanisms include organelle transport, membrane undulations, cytoskeletal restructuring, strain at cellular adhesions, cytokinesis, mitosis, exo- and endo-cytosis among others. The development of 3D high-content assays such as biodynamic profiling can become a critical new tool for assessing efficacy of drugs and the suitability of specific types of tissue growth for drug discovery and development. The use of biodynamic profiling to predict clinical outcome of living biopsies to cancer therapeutics can be developed into a phenotypic companion diagnostic, as well as a new tool for therapy selection in personalized medicine. This invited talk will present an overview of the optical, physical and physiological processes involved in biodynamic imaging. Several different biodynamic imaging modalities include motility contrast imaging (MCI), tissue-dynamics spectroscopy (TDS) and tissue-dynamics imaging (TDI). A wide range of potential applications will be described that include process monitoring for 3D tissue culture, drug discovery and development, cancer therapy selection, embryo assessment for in-vitro fertilization and artificial reproductive technologies, among others.

3D Printer Generated Tissue iMolds for Cleared Tissue Using Single- and Multi-Photon Microscopy for Deep Tissue Evaluation.

PubMed

Miller, Sean J; Rothstein, Jeffrey D

2017-01-01

Pathological analyses and methodology has recently undergone a dramatic revolution. With the creation of tissue clearing methods such as CLARITY and CUBIC, groups can now achieve complete transparency in tissue samples in nano-porous hydrogels. Cleared tissue is then imagined in a semi-aqueous medium that matches the refractive index of the objective being used. However, one major challenge is the ability to control tissue movement during imaging and to relocate precise locations post sequential clearing and re-staining. Using 3D printers, we designed tissue molds that fit precisely around the specimen being imaged. First, images are taken of the specimen, followed by importing and design of a structural mold, then printed with affordable plastics by a 3D printer. With our novel design, we have innovated tissue molds called innovative molds (iMolds) that can be generated in any laboratory and are customized for any organ, tissue, or bone matter being imaged. Furthermore, the inexpensive and reusable tissue molds are made compatible for any microscope such as single and multi-photon confocal with varying stage dimensions. Excitingly, iMolds can also be generated to hold multiple organs in one mold, making reconstruction and imaging much easier. Taken together, with iMolds it is now possible to image cleared tissue in clearing medium while limiting movement and being able to relocate precise anatomical and cellular locations on sequential imaging events in any basic laboratory. This system provides great potential for screening widespread effects of therapeutics and disease across entire organ systems.

Flexible needle with integrated optical coherence tomography probe for imaging during transbronchial tissue aspiration

NASA Astrophysics Data System (ADS)

Li, Jiawen; Quirk, Bryden C.; Noble, Peter B.; Kirk, Rodney W.; Sampson, David D.; McLaughlin, Robert A.

2017-10-01

Transbronchial needle aspiration (TBNA) of small lesions or lymph nodes in the lung may result in nondiagnostic tissue samples. We demonstrate the integration of an optical coherence tomography (OCT) probe into a 19-gauge flexible needle for lung tissue aspiration. This probe allows simultaneous visualization and aspiration of the tissue. By eliminating the need for insertion and withdrawal of a separate imaging probe, this integrated design minimizes the risk of dislodging the needle from the lesion prior to aspiration and may facilitate more accurate placement of the needle. Results from in situ imaging in a sheep lung show clear distinction between solid tissue and two typical constituents of nondiagnostic samples (adipose and lung parenchyma). Clinical translation of this OCT-guided aspiration needle holds promise for improving the diagnostic yield of TBNA.

Image-guided tissue engineering

PubMed Central

Ballyns, Jeffrey J; Bonassar, Lawrence J

2009-01-01

Replication of anatomic shape is a significant challenge in developing implants for regenerative medicine. This has lead to significant interest in using medical imaging techniques such as magnetic resonance imaging and computed tomography to design tissue engineered constructs. Implementation of medical imaging and computer aided design in combination with technologies for rapid prototyping of living implants enables the generation of highly reproducible constructs with spatial resolution up to 25 μm. In this paper, we review the medical imaging modalities available and a paradigm for choosing a particular imaging technique. We also present fabrication techniques and methodologies for producing cellular engineered constructs. Finally, we comment on future challenges involved with image guided tissue engineering and efforts to generate engineered constructs ready for implantation. PMID:19583811

Improved QD-BRET conjugates for detection and imaging

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

Xing Yun; So, Min-kyung; Koh, Ai Leen

2008-08-01

Self-illuminating quantum dots, also known as QD-BRET conjugates, are a new class of quantum dot bioconjugates which do not need external light for excitation. Instead, light emission relies on the bioluminescence resonance energy transfer from the attached Renilla luciferase enzyme, which emits light upon the oxidation of its substrate. QD-BRET combines the advantages of the QDs (such as superior brightness and photostability, tunable emission, multiplexing) as well as the high sensitivity of bioluminescence imaging, thus holding the promise for improved deep tissue in vivo imaging. Although studies have demonstrated the superior sensitivity and deep tissue imaging potential, the stability ofmore » the QD-BRET conjugates in biological environment needs to be improved for long-term imaging studies such as in vivo cell tracking. In this study, we seek to improve the stability of QD-BRET probes through polymeric encapsulation with a polyacrylamide gel. Results show that encapsulation caused some activity loss, but significantly improved both the in vitro serum stability and in vivo stability when subcutaneously injected into the animal. Stable QD-BRET probes should further facilitate their applications for both in vitro testing as well as in vivo cell tracking studies.« less

Tissue microstructure estimation using a deep network inspired by a dictionary-based framework.

PubMed

Ye, Chuyang

2017-12-01

Diffusion magnetic resonance imaging (dMRI) captures the anisotropic pattern of water displacement in the neuronal tissue and allows noninvasive investigation of the complex tissue microstructure. A number of biophysical models have been proposed to relate the tissue organization with the observed diffusion signals, so that the tissue microstructure can be inferred. The Neurite Orientation Dispersion and Density Imaging (NODDI) model has been a popular choice and has been widely used for many neuroscientific studies. It models the diffusion signal with three compartments that are characterized by distinct diffusion properties, and the parameters in the model describe tissue microstructure. In NODDI, these parameters are estimated in a maximum likelihood framework, where the nonlinear model fitting is computationally intensive. Therefore, efforts have been made to develop efficient and accurate algorithms for NODDI microstructure estimation, which is still an open problem. In this work, we propose a deep network based approach that performs end-to-end estimation of NODDI microstructure, which is named Microstructure Estimation using a Deep Network (MEDN). MEDN comprises two cascaded stages and is motivated by the AMICO algorithm, where the NODDI microstructure estimation is formulated in a dictionary-based framework. The first stage computes the coefficients of the dictionary. It resembles the solution to a sparse reconstruction problem, where the iterative process in conventional estimation approaches is unfolded and truncated, and the weights are learned instead of predetermined by the dictionary. In the second stage, microstructure properties are computed from the output of the first stage, which resembles the weighted sum of normalized dictionary coefficients in AMICO, and the weights are also learned. Because spatial consistency of diffusion signals can be used to reduce the effect of noise, we also propose MEDN+, which is an extended version of MEDN. MEDN

Deep Learning in Nuclear Medicine and Molecular Imaging: Current Perspectives and Future Directions.

PubMed

Choi, Hongyoon

2018-04-01

Recent advances in deep learning have impacted various scientific and industrial fields. Due to the rapid application of deep learning in biomedical data, molecular imaging has also started to adopt this technique. In this regard, it is expected that deep learning will potentially affect the roles of molecular imaging experts as well as clinical decision making. This review firstly offers a basic overview of deep learning particularly for image data analysis to give knowledge to nuclear medicine physicians and researchers. Because of the unique characteristics and distinctive aims of various types of molecular imaging, deep learning applications can be different from other fields. In this context, the review deals with current perspectives of deep learning in molecular imaging particularly in terms of development of biomarkers. Finally, future challenges of deep learning application for molecular imaging and future roles of experts in molecular imaging will be discussed.

Distributed deep learning networks among institutions for medical imaging.

PubMed

Chang, Ken; Balachandar, Niranjan; Lam, Carson; Yi, Darvin; Brown, James; Beers, Andrew; Rosen, Bruce; Rubin, Daniel L; Kalpathy-Cramer, Jayashree

2018-03-29

Deep learning has become a promising approach for automated support for clinical diagnosis. When medical data samples are limited, collaboration among multiple institutions is necessary to achieve high algorithm performance. However, sharing patient data often has limitations due to technical, legal, or ethical concerns. In this study, we propose methods of distributing deep learning models as an attractive alternative to sharing patient data. We simulate the distribution of deep learning models across 4 institutions using various training heuristics and compare the results with a deep learning model trained on centrally hosted patient data. The training heuristics investigated include ensembling single institution models, single weight transfer, and cyclical weight transfer. We evaluated these approaches for image classification in 3 independent image collections (retinal fundus photos, mammography, and ImageNet). We find that cyclical weight transfer resulted in a performance that was comparable to that of centrally hosted patient data. We also found that there is an improvement in the performance of cyclical weight transfer heuristic with a high frequency of weight transfer. We show that distributing deep learning models is an effective alternative to sharing patient data. This finding has implications for any collaborative deep learning study.

Nonlinear spectral imaging of biological tissues

NASA Astrophysics Data System (ADS)

Palero, J. A.

2007-07-01

The work presented in this thesis demonstrates live high resolution 3D imaging of tissue in its native state and environment. The nonlinear interaction between focussed femtosecond light pulses and the biological tissue results in the emission of natural autofluorescence and second-harmonic signal. Because biological intrinsic emission is generally very weak and extends from the ultraviolet to the visible spectral range, a broad-spectral range and high sensitivity 3D spectral imaging system is developed. Imaging the spectral characteristics of the biological intrinsic emission reveals the structure and biochemistry of the cells and extra-cellular components. By using different methods in visualizing the spectral images, discrimination between different tissue structures is achieved without the use of any stain or fluorescent label. For instance, RGB real color spectral images of the intrinsic emission of mouse skin tissues show blue cells, green hair follicles, and purple collagen fibers. The color signature of each tissue component is directly related to its characteristic emission spectrum. The results of this study show that skin tissue nonlinear intrinsic emission is mainly due to the autofluorescence of reduced nicotinamide adenine dinucleotide (phosphate), flavins, keratin, melanin, phospholipids, elastin and collagen and nonlinear Raman scattering and second-harmonic generation in Type I collagen. In vivo time-lapse spectral imaging is implemented to study metabolic changes in epidermal cells in tissues. Optical scattering in tissues, a key factor in determining the maximum achievable imaging depth, is also investigated in this work.

A deep-learning classifier identifies patients with clinical heart failure using whole-slide images of H&E tissue

PubMed Central

Peyster, Eliot G.; Frank, Renee; Margulies, Kenneth B.; Feldman, Michael D.

2018-01-01

Over 26 million people worldwide suffer from heart failure annually. When the cause of heart failure cannot be identified, endomyocardial biopsy (EMB) represents the gold-standard for the evaluation of disease. However, manual EMB interpretation has high inter-rater variability. Deep convolutional neural networks (CNNs) have been successfully applied to detect cancer, diabetic retinopathy, and dermatologic lesions from images. In this study, we develop a CNN classifier to detect clinical heart failure from H&E stained whole-slide images from a total of 209 patients, 104 patients were used for training and the remaining 105 patients for independent testing. The CNN was able to identify patients with heart failure or severe pathology with a 99% sensitivity and 94% specificity on the test set, outperforming conventional feature-engineering approaches. Importantly, the CNN outperformed two expert pathologists by nearly 20%. Our results suggest that deep learning analytics of EMB can be used to predict cardiac outcome. PMID:29614076

A deep-learning classifier identifies patients with clinical heart failure using whole-slide images of H&E tissue.

PubMed

Nirschl, Jeffrey J; Janowczyk, Andrew; Peyster, Eliot G; Frank, Renee; Margulies, Kenneth B; Feldman, Michael D; Madabhushi, Anant

2018-01-01

Over 26 million people worldwide suffer from heart failure annually. When the cause of heart failure cannot be identified, endomyocardial biopsy (EMB) represents the gold-standard for the evaluation of disease. However, manual EMB interpretation has high inter-rater variability. Deep convolutional neural networks (CNNs) have been successfully applied to detect cancer, diabetic retinopathy, and dermatologic lesions from images. In this study, we develop a CNN classifier to detect clinical heart failure from H&E stained whole-slide images from a total of 209 patients, 104 patients were used for training and the remaining 105 patients for independent testing. The CNN was able to identify patients with heart failure or severe pathology with a 99% sensitivity and 94% specificity on the test set, outperforming conventional feature-engineering approaches. Importantly, the CNN outperformed two expert pathologists by nearly 20%. Our results suggest that deep learning analytics of EMB can be used to predict cardiac outcome.

3D printing facilitated scaffold-free tissue unit fabrication.

PubMed

Tan, Yu; Richards, Dylan J; Trusk, Thomas C; Visconti, Richard P; Yost, Michael J; Kindy, Mark S; Drake, Christopher J; Argraves, William Scott; Markwald, Roger R; Mei, Ying

2014-06-01

Tissue spheroids hold great potential in tissue engineering as building blocks to assemble into functional tissues. To date, agarose molds have been extensively used to facilitate fusion process of tissue spheroids. As a molding material, agarose typically requires low temperature plates for gelation and/or heated dispenser units. Here, we proposed and developed an alginate-based, direct 3D mold-printing technology: 3D printing microdroplets of alginate solution into biocompatible, bio-inert alginate hydrogel molds for the fabrication of scaffold-free tissue engineering constructs. Specifically, we developed a 3D printing technology to deposit microdroplets of alginate solution on calcium containing substrates in a layer-by-layer fashion to prepare ring-shaped 3D hydrogel molds. Tissue spheroids composed of 50% endothelial cells and 50% smooth muscle cells were robotically placed into the 3D printed alginate molds using a 3D printer, and were found to rapidly fuse into toroid-shaped tissue units. Histological and immunofluorescence analysis indicated that the cells secreted collagen type I playing a critical role in promoting cell-cell adhesion, tissue formation and maturation.

A survey on deep learning in medical image analysis.

PubMed

Litjens, Geert; Kooi, Thijs; Bejnordi, Babak Ehteshami; Setio, Arnaud Arindra Adiyoso; Ciompi, Francesco; Ghafoorian, Mohsen; van der Laak, Jeroen A W M; van Ginneken, Bram; Sánchez, Clara I

2017-12-01

Deep learning algorithms, in particular convolutional networks, have rapidly become a methodology of choice for analyzing medical images. This paper reviews the major deep learning concepts pertinent to medical image analysis and summarizes over 300 contributions to the field, most of which appeared in the last year. We survey the use of deep learning for image classification, object detection, segmentation, registration, and other tasks. Concise overviews are provided of studies per application area: neuro, retinal, pulmonary, digital pathology, breast, cardiac, abdominal, musculoskeletal. We end with a summary of the current state-of-the-art, a critical discussion of open challenges and directions for future research. Copyright © 2017 Elsevier B.V. All rights reserved.

Distribution, composition and functions of gelatinous tissues in deep-sea fishes.

PubMed

Gerringer, Mackenzie E; Drazen, Jeffrey C; Linley, Thomas D; Summers, Adam P; Jamieson, Alan J; Yancey, Paul H

2017-12-01

Many deep-sea fishes have a gelatinous layer, or subdermal extracellular matrix, below the skin or around the spine. We document the distribution of gelatinous tissues across fish families (approx. 200 species in ten orders), then review and investigate their composition and function. Gelatinous tissues from nine species were analysed for water content (96.53 ± 1.78% s.d.), ionic composition, osmolality, protein (0.39 ± 0.23%), lipid (0.69 ± 0.56%) and carbohydrate (0.61 ± 0.28%). Results suggest that gelatinous tissues are mostly extracellular fluid, which may allow animals to grow inexpensively. Further, almost all gelatinous tissues floated in cold seawater, thus their lower density than seawater may contribute to buoyancy in some species. We also propose a new hypothesis: gelatinous tissues, which are inexpensive to grow, may sometimes be a method to increase swimming efficiency by fairing the transition from trunk to tail. Such a layer is particularly prominent in hadal snailfishes (Liparidae); therefore, a robotic snailfish model was designed and constructed to analyse the influence of gelatinous tissues on locomotory performance. The model swam faster with a watery layer, representing gelatinous tissue, around the tail than without. Results suggest that the tissues may, in addition to providing buoyancy and low-cost growth, aid deep-sea fish locomotion.

Distribution, composition and functions of gelatinous tissues in deep-sea fishes

PubMed Central

Drazen, Jeffrey C.; Linley, Thomas D.; Summers, Adam P.; Jamieson, Alan J.; Yancey, Paul H.

2017-01-01

Many deep-sea fishes have a gelatinous layer, or subdermal extracellular matrix, below the skin or around the spine. We document the distribution of gelatinous tissues across fish families (approx. 200 species in ten orders), then review and investigate their composition and function. Gelatinous tissues from nine species were analysed for water content (96.53 ± 1.78% s.d.), ionic composition, osmolality, protein (0.39 ± 0.23%), lipid (0.69 ± 0.56%) and carbohydrate (0.61 ± 0.28%). Results suggest that gelatinous tissues are mostly extracellular fluid, which may allow animals to grow inexpensively. Further, almost all gelatinous tissues floated in cold seawater, thus their lower density than seawater may contribute to buoyancy in some species. We also propose a new hypothesis: gelatinous tissues, which are inexpensive to grow, may sometimes be a method to increase swimming efficiency by fairing the transition from trunk to tail. Such a layer is particularly prominent in hadal snailfishes (Liparidae); therefore, a robotic snailfish model was designed and constructed to analyse the influence of gelatinous tissues on locomotory performance. The model swam faster with a watery layer, representing gelatinous tissue, around the tail than without. Results suggest that the tissues may, in addition to providing buoyancy and low-cost growth, aid deep-sea fish locomotion. PMID:29308245

Iterative deep convolutional encoder-decoder network for medical image segmentation.

PubMed

Jung Uk Kim; Hak Gu Kim; Yong Man Ro

2017-07-01

In this paper, we propose a novel medical image segmentation using iterative deep learning framework. We have combined an iterative learning approach and an encoder-decoder network to improve segmentation results, which enables to precisely localize the regions of interest (ROIs) including complex shapes or detailed textures of medical images in an iterative manner. The proposed iterative deep convolutional encoder-decoder network consists of two main paths: convolutional encoder path and convolutional decoder path with iterative learning. Experimental results show that the proposed iterative deep learning framework is able to yield excellent medical image segmentation performances for various medical images. The effectiveness of the proposed method has been proved by comparing with other state-of-the-art medical image segmentation methods.

Radiometric Thermometry for Wearable Deep Tissue Monitoring

NASA Astrophysics Data System (ADS)

Momenroodaki, Parisa

Microwave thermometry is an attractive non-invasive method for measuring internal body temperature. This approach has the potential of enabling a wearable device that can continuously monitor core body temperature. There are a number of health-related applications in both diagnostics and therapy, which can benefit from the knowledge of core body temperature. However,there are a limited number of device solutions, which are usually not wearable or cannot continuously monitor internal body temperature non-invasively. In this thesis, a possible path toward implementing such a thermometer is presented. The device operates in the "quiet" frequency band of 1.4 GHz which is chosen as a compromise between sensing depth and radio frequency interference (RFI). A major challenge in microwave thermometry is detecting small temperature variations of deep tissue layers from surface (skin) measurements. The type and thickness of tissue materials significantly affect the design of the probe, which has the function of receiving black-body radiation from tissues beneath it and coupling the power to a sensitive radiometric receiver. High dielectric constant contrast between skin, fat (/bone), and muscle layers suggests structures with dominant tangential component of the electric field, such as a patch or slot. Adding a layer of low-loss,low-dielectric constant superstrate can further reduce the contribution of superficial tissue layers in the received thermal noise. Several probe types are designed using a full-wave electromagnetic simulator, with a goal of maximizing the power reception from deep tissue layers. The designs are validated with a second software tool and various measurements. A stable, narrow-band, and highly sensitive radiometer is developed, enabling the device to operate in a non-shielded RF environment.To use the microwave thermometer in a RF congested environment, not only narrow-band probe and radiometers are used but an additional probe is introduced for

Use of kurtosis for locating deep blood vessels in raw speckle imaging using a homogeneity representation.

PubMed

Peregrina-Barreto, Hayde; Perez-Corona, Elizabeth; Rangel-Magdaleno, Jose; Ramos-Garcia, Ruben; Chiu, Roger; Ramirez-San-Juan, Julio C

2017-06-01

Visualization of deep blood vessels in speckle images is an important task as it is used to analyze the dynamics of the blood flow and the health status of biological tissue. Laser speckle imaging is a wide-field optical technique to measure relative blood flow speed based on the local speckle contrast analysis. However, it has been reported that this technique is limited to certain deep blood vessels (about ? = 300 ?? ? m ) because of the high scattering of the sample; beyond this depth, the quality of the vessel’s image decreases. The use of a representation based on homogeneity values, computed from the co-occurrence matrix, is proposed as it provides an improved vessel definition and its corresponding diameter. Moreover, a methodology is proposed for automatic blood vessel location based on the kurtosis analysis. Results were obtained from the different skin phantoms, showing that it is possible to identify the vessel region for different morphologies, even up to 900 ?? ? m in depth.

Characterization of water molecular state in in-vivo thick tissues using diffuse optical spectroscopic imaging

NASA Astrophysics Data System (ADS)

Chung, So Hyun

Structural changes in water molecules are related to physiological, anatomical and pathological properties of tissues. Near infrared (NIR) optical absorption methods are sensitive to water; however, detailed characterization of water in thick tissues is difficult to achieve because subtle spectral shifts can be obscured by multiple light scattering. In the NIR, a water absorption peak is observed around 975 nm. The precise NIR peak's shape and position are highly sensitive to water molecular disposition. A bound water index (BWI) was developed that quantifies the spectral shift and shape changes observed in tissue water absorption spectra measured by broadband diffuse optical spectroscopic imaging (DOSI). DOSI quantitatively measures light absorption and scattering spectra in cm-deep tissues and therefore reveals bound water spectral shifts. BWI as a water state index was validated by comparing broadband DOSI to MRI and a conductivity cell using bound water phantoms. Non-invasive BWI measurements of malignant and normal tissues in 18 subjects showed a significantly higher fraction of free water in malignant tissues (p<0.0001) compared to normal tissues. BWI showed potential as a prognostic index based on high correlations with tumor grade and size. An algorithm for absolute temperature measurements in deep tissues was developed based on resolving opposing effects of water vibrational frequency shifts due to macromolecular binding. DOSI measures absolute temperature with a difference of 1.1+/-0.91°C from a thermistor. Deep tissue temperature measured in forearms during cold-stress was consistent with previously reported invasively-measured deep tissue temperature. Finally, the BWI was compared to Apparent Diffusion Coefficient (ADC) of diffusion weighted MRI in 9 breast cancer patients. The BWI and ADC correlated (R=0.8, p=<0.01) and both parameters decreased with increasing bulk water content in cancer tissues. Although BWI and ADC are positively correlated in

Deep kernel learning method for SAR image target recognition

NASA Astrophysics Data System (ADS)

Chen, Xiuyuan; Peng, Xiyuan; Duan, Ran; Li, Junbao

2017-10-01

With the development of deep learning, research on image target recognition has made great progress in recent years. Remote sensing detection urgently requires target recognition for military, geographic, and other scientific research. This paper aims to solve the synthetic aperture radar image target recognition problem by combining deep and kernel learning. The model, which has a multilayer multiple kernel structure, is optimized layer by layer with the parameters of Support Vector Machine and a gradient descent algorithm. This new deep kernel learning method improves accuracy and achieves competitive recognition results compared with other learning methods.

Image quality assessment using deep convolutional networks

NASA Astrophysics Data System (ADS)

Li, Yezhou; Ye, Xiang; Li, Yong

2017-12-01

This paper proposes a method of accurately assessing image quality without a reference image by using a deep convolutional neural network. Existing training based methods usually utilize a compact set of linear filters for learning features of images captured by different sensors to assess their quality. These methods may not be able to learn the semantic features that are intimately related with the features used in human subject assessment. Observing this drawback, this work proposes training a deep convolutional neural network (CNN) with labelled images for image quality assessment. The ReLU in the CNN allows non-linear transformations for extracting high-level image features, providing a more reliable assessment of image quality than linear filters. To enable the neural network to take images of any arbitrary size as input, the spatial pyramid pooling (SPP) is introduced connecting the top convolutional layer and the fully-connected layer. In addition, the SPP makes the CNN robust to object deformations to a certain extent. The proposed method taking an image as input carries out an end-to-end learning process, and outputs the quality of the image. It is tested on public datasets. Experimental results show that it outperforms existing methods by a large margin and can accurately assess the image quality on images taken by different sensors of varying sizes.

Application of deep learning to the classification of images from colposcopy.

PubMed

Sato, Masakazu; Horie, Koji; Hara, Aki; Miyamoto, Yuichiro; Kurihara, Kazuko; Tomio, Kensuke; Yokota, Harushige

2018-03-01

The objective of the present study was to investigate whether deep learning could be applied successfully to the classification of images from colposcopy. For this purpose, a total of 158 patients who underwent conization were enrolled, and medical records and data from the gynecological oncology database were retrospectively reviewed. Deep learning was performed with the Keras neural network and TensorFlow libraries. Using preoperative images from colposcopy as the input data and deep learning technology, the patients were classified into three groups [severe dysplasia, carcinoma in situ (CIS) and invasive cancer (IC)]. A total of 485 images were obtained for the analysis, of which 142 images were of severe dysplasia (2.9 images/patient), 257 were of CIS (3.3 images/patient), and 86 were of IC (4.1 images/patient). Of these, 233 images were captured with a green filter, and the remaining 252 were captured without a green filter. Following the application of L2 regularization, L1 regularization, dropout and data augmentation, the accuracy of the validation dataset was ~50%. Although the present study is preliminary, the results indicated that deep learning may be applied to classify colposcopy images.

Application of deep learning to the classification of images from colposcopy

PubMed Central

Sato, Masakazu; Horie, Koji; Hara, Aki; Miyamoto, Yuichiro; Kurihara, Kazuko; Tomio, Kensuke; Yokota, Harushige

2018-01-01

The objective of the present study was to investigate whether deep learning could be applied successfully to the classification of images from colposcopy. For this purpose, a total of 158 patients who underwent conization were enrolled, and medical records and data from the gynecological oncology database were retrospectively reviewed. Deep learning was performed with the Keras neural network and TensorFlow libraries. Using preoperative images from colposcopy as the input data and deep learning technology, the patients were classified into three groups [severe dysplasia, carcinoma in situ (CIS) and invasive cancer (IC)]. A total of 485 images were obtained for the analysis, of which 142 images were of severe dysplasia (2.9 images/patient), 257 were of CIS (3.3 images/patient), and 86 were of IC (4.1 images/patient). Of these, 233 images were captured with a green filter, and the remaining 252 were captured without a green filter. Following the application of L2 regularization, L1 regularization, dropout and data augmentation, the accuracy of the validation dataset was ~50%. Although the present study is preliminary, the results indicated that deep learning may be applied to classify colposcopy images. PMID:29456725

Photoacoustic image reconstruction via deep learning

NASA Astrophysics Data System (ADS)

Antholzer, Stephan; Haltmeier, Markus; Nuster, Robert; Schwab, Johannes

2018-02-01

Applying standard algorithms to sparse data problems in photoacoustic tomography (PAT) yields low-quality images containing severe under-sampling artifacts. To some extent, these artifacts can be reduced by iterative image reconstruction algorithms which allow to include prior knowledge such as smoothness, total variation (TV) or sparsity constraints. These algorithms tend to be time consuming as the forward and adjoint problems have to be solved repeatedly. Further, iterative algorithms have additional drawbacks. For example, the reconstruction quality strongly depends on a-priori model assumptions about the objects to be recovered, which are often not strictly satisfied in practical applications. To overcome these issues, in this paper, we develop direct and efficient reconstruction algorithms based on deep learning. As opposed to iterative algorithms, we apply a convolutional neural network, whose parameters are trained before the reconstruction process based on a set of training data. For actual image reconstruction, a single evaluation of the trained network yields the desired result. Our presented numerical results (using two different network architectures) demonstrate that the proposed deep learning approach reconstructs images with a quality comparable to state of the art iterative reconstruction methods.

Jet-images — deep learning edition

DOE PAGES

de Oliveira, Luke; Kagan, Michael; Mackey, Lester; ...

2016-07-13

Building on the notion of a particle physics detector as a camera and the collimated streams of high energy particles, or jets, it measures as an image, we investigate the potential of machine learning techniques based on deep learning architectures to identify highly boosted W bosons. Modern deep learning algorithms trained on jet images can out-perform standard physically-motivated feature driven approaches to jet tagging. We develop techniques for visualizing how these features are learned by the network and what additional information is used to improve performance. Finally, this interplay between physically-motivated feature driven tools and supervised learning algorithms is generalmore » and can be used to significantly increase the sensitivity to discover new particles and new forces, and gain a deeper understanding of the physics within jets.« less

Jet-images — deep learning edition

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

de Oliveira, Luke; Kagan, Michael; Mackey, Lester

Building on the notion of a particle physics detector as a camera and the collimated streams of high energy particles, or jets, it measures as an image, we investigate the potential of machine learning techniques based on deep learning architectures to identify highly boosted W bosons. Modern deep learning algorithms trained on jet images can out-perform standard physically-motivated feature driven approaches to jet tagging. We develop techniques for visualizing how these features are learned by the network and what additional information is used to improve performance. Finally, this interplay between physically-motivated feature driven tools and supervised learning algorithms is generalmore » and can be used to significantly increase the sensitivity to discover new particles and new forces, and gain a deeper understanding of the physics within jets.« less

Imaging of Alkaline Phosphatase Activity in Bone Tissue

PubMed Central

Gade, Terence P.; Motley, Matthew W.; Beattie, Bradley J.; Bhakta, Roshni; Boskey, Adele L.; Koutcher, Jason A.; Mayer-Kuckuk, Philipp

2011-01-01

The purpose of this study was to develop a paradigm for quantitative molecular imaging of bone cell activity. We hypothesized the feasibility of non-invasive imaging of the osteoblast enzyme alkaline phosphatase (ALP) using a small imaging molecule in combination with 19Flourine magnetic resonance spectroscopic imaging (19FMRSI). 6, 8-difluoro-4-methylumbelliferyl phosphate (DiFMUP), a fluorinated ALP substrate that is activatable to a fluorescent hydrolysis product was utilized as a prototype small imaging molecule. The molecular structure of DiFMUP includes two Fluorine atoms adjacent to a phosphate group allowing it and its hydrolysis product to be distinguished using 19Fluorine magnetic resonance spectroscopy (19FMRS) and 19FMRSI. ALP-mediated hydrolysis of DiFMUP was tested on osteoblastic cells and bone tissue, using serial measurements of fluorescence activity. Extracellular activation of DiFMUP on ALP-positive mouse bone precursor cells was observed. Concurringly, DiFMUP was also activated on bone derived from rat tibia. Marked inhibition of the cell and tissue activation of DiFMUP was detected after the addition of the ALP inhibitor levamisole. 19FMRS and 19FMRSI were applied for the non-invasive measurement of DiFMUP hydrolysis. 19FMRS revealed a two-peak spectrum representing DiFMUP with an associated chemical shift for the hydrolysis product. Activation of DiFMUP by ALP yielded a characteristic pharmacokinetic profile, which was quantifiable using non-localized 19FMRS and enabled the development of a pharmacokinetic model of ALP activity. Application of 19FMRSI facilitated anatomically accurate, non-invasive imaging of ALP concentration and activity in rat bone. Thus, 19FMRSI represents a promising approach for the quantitative imaging of bone cell activity during bone formation with potential for both preclinical and clinical applications. PMID:21799916

Three Dimensional Imaging of Paraffin Embedded Human Lung Tissue Samples by Micro-Computed Tomography

PubMed Central

Scott, Anna E.; Vasilescu, Dragos M.; Seal, Katherine A. D.; Keyes, Samuel D.; Mavrogordato, Mark N.; Hogg, James C.; Sinclair, Ian; Warner, Jane A.; Hackett, Tillie-Louise; Lackie, Peter M.

2015-01-01

Background Understanding the three-dimensional (3-D) micro-architecture of lung tissue can provide insights into the pathology of lung disease. Micro computed tomography (µCT) has previously been used to elucidate lung 3D histology and morphometry in fixed samples that have been stained with contrast agents or air inflated and dried. However, non-destructive microstructural 3D imaging of formalin-fixed paraffin embedded (FFPE) tissues would facilitate retrospective analysis of extensive tissue archives of lung FFPE lung samples with linked clinical data. Methods FFPE human lung tissue samples (n = 4) were scanned using a Nikon metrology µCT scanner. Semi-automatic techniques were used to segment the 3D structure of airways and blood vessels. Airspace size (mean linear intercept, Lm) was measured on µCT images and on matched histological sections from the same FFPE samples imaged by light microscopy to validate µCT imaging. Results The µCT imaging protocol provided contrast between tissue and paraffin in FFPE samples (15mm x 7mm). Resolution (voxel size 6.7 µm) in the reconstructed images was sufficient for semi-automatic image segmentation of airways and blood vessels as well as quantitative airspace analysis. The scans were also used to scout for regions of interest, enabling time-efficient preparation of conventional histological sections. The Lm measurements from µCT images were not significantly different to those from matched histological sections. Conclusion We demonstrated how non-destructive imaging of routinely prepared FFPE samples by laboratory µCT can be used to visualize and assess the 3D morphology of the lung including by morphometric analysis. PMID:26030902

Deep learning of symmetrical discrepancies for computer-aided detection of mammographic masses

NASA Astrophysics Data System (ADS)

Kooi, Thijs; Karssemeijer, Nico

2017-03-01

When humans identify objects in images, context is an important cue; a cheetah is more likely to be a domestic cat when a television set is recognised in the background. Similar principles apply to the analysis of medical images. The detection of diseases that manifest unilaterally in symmetrical organs or organ pairs can in part be facilitated by a search for symmetrical discrepancies in or between the organs in question. During a mammographic exam, images are recorded of each breast and absence of a certain structure around the same location in the contralateral image will render the area under scrutiny more suspicious and conversely, the presence of similar tissue less so. In this paper, we present a fusion scheme for a deep Convolutional Neural Network (CNN) architecture with the goal to optimally capture such asymmetries. The method is applied to the domain of mammography CAD, but can be relevant to other medical image analysis tasks where symmetry is important such as lung, prostate or brain images.

Ontology-based, Tissue MicroArray oriented, image centered tissue bank

PubMed Central

Viti, Federica; Merelli, Ivan; Caprera, Andrea; Lazzari, Barbara; Stella, Alessandra; Milanesi, Luciano

2008-01-01

Background Tissue MicroArray technique is becoming increasingly important in pathology for the validation of experimental data from transcriptomic analysis. This approach produces many images which need to be properly managed, if possible with an infrastructure able to support tissue sharing between institutes. Moreover, the available frameworks oriented to Tissue MicroArray provide good storage for clinical patient, sample treatment and block construction information, but their utility is limited by the lack of data integration with biomolecular information. Results In this work we propose a Tissue MicroArray web oriented system to support researchers in managing bio-samples and, through the use of ontologies, enables tissue sharing aimed at the design of Tissue MicroArray experiments and results evaluation. Indeed, our system provides ontological description both for pre-analysis tissue images and for post-process analysis image results, which is crucial for information exchange. Moreover, working on well-defined terms it is then possible to query web resources for literature articles to integrate both pathology and bioinformatics data. Conclusions Using this system, users associate an ontology-based description to each image uploaded into the database and also integrate results with the ontological description of biosequences identified in every tissue. Moreover, it is possible to integrate the ontological description provided by the user with a full compliant gene ontology definition, enabling statistical studies about correlation between the analyzed pathology and the most commonly related biological processes. PMID:18460177

Deep Learning in Medical Imaging: General Overview

PubMed Central

Lee, June-Goo; Jun, Sanghoon; Cho, Young-Won; Lee, Hyunna; Kim, Guk Bae

2017-01-01

The artificial neural network (ANN)–a machine learning technique inspired by the human neuronal synapse system–was introduced in the 1950s. However, the ANN was previously limited in its ability to solve actual problems, due to the vanishing gradient and overfitting problems with training of deep architecture, lack of computing power, and primarily the absence of sufficient data to train the computer system. Interest in this concept has lately resurfaced, due to the availability of big data, enhanced computing power with the current graphics processing units, and novel algorithms to train the deep neural network. Recent studies on this technology suggest its potentially to perform better than humans in some visual and auditory recognition tasks, which may portend its applications in medicine and healthcare, especially in medical imaging, in the foreseeable future. This review article offers perspectives on the history, development, and applications of deep learning technology, particularly regarding its applications in medical imaging. PMID:28670152

Deep Learning in Medical Imaging: General Overview.

PubMed

Lee, June-Goo; Jun, Sanghoon; Cho, Young-Won; Lee, Hyunna; Kim, Guk Bae; Seo, Joon Beom; Kim, Namkug

2017-01-01

The artificial neural network (ANN)-a machine learning technique inspired by the human neuronal synapse system-was introduced in the 1950s. However, the ANN was previously limited in its ability to solve actual problems, due to the vanishing gradient and overfitting problems with training of deep architecture, lack of computing power, and primarily the absence of sufficient data to train the computer system. Interest in this concept has lately resurfaced, due to the availability of big data, enhanced computing power with the current graphics processing units, and novel algorithms to train the deep neural network. Recent studies on this technology suggest its potentially to perform better than humans in some visual and auditory recognition tasks, which may portend its applications in medicine and healthcare, especially in medical imaging, in the foreseeable future. This review article offers perspectives on the history, development, and applications of deep learning technology, particularly regarding its applications in medical imaging.

Photoacoustic tomography of foreign bodies in soft biological tissue.

PubMed

Cai, Xin; Kim, Chulhong; Pramanik, Manojit; Wang, Lihong V

2011-04-01

In detecting small foreign bodies in soft biological tissue, ultrasound imaging suffers from poor sensitivity (52.6%) and specificity (47.2%). Hence, alternative imaging methods are needed. Photoacoustic (PA) imaging takes advantage of strong optical absorption contrast and high ultrasonic resolution. A PA imaging system is employed to detect foreign bodies in biological tissues. To achieve deep penetration, we use near-infrared light ranging from 750 to 800 nm and a 5-MHz spherically focused ultrasonic transducer. PA images were obtained from various targets including glass, wood, cloth, plastic, and metal embedded more than 1 cm deep in chicken tissue. The locations and sizes of the targets from the PA images agreed well with those of the actual samples. Spectroscopic PA imaging was also performed on the objects. These results suggest that PA imaging can potentially be a useful intraoperative imaging tool to identify foreign bodies.

Classification of breast MRI lesions using small-size training sets: comparison of deep learning approaches

NASA Astrophysics Data System (ADS)

Amit, Guy; Ben-Ari, Rami; Hadad, Omer; Monovich, Einat; Granot, Noa; Hashoul, Sharbell

2017-03-01

Diagnostic interpretation of breast MRI studies requires meticulous work and a high level of expertise. Computerized algorithms can assist radiologists by automatically characterizing the detected lesions. Deep learning approaches have shown promising results in natural image classification, but their applicability to medical imaging is limited by the shortage of large annotated training sets. In this work, we address automatic classification of breast MRI lesions using two different deep learning approaches. We propose a novel image representation for dynamic contrast enhanced (DCE) breast MRI lesions, which combines the morphological and kinetics information in a single multi-channel image. We compare two classification approaches for discriminating between benign and malignant lesions: training a designated convolutional neural network and using a pre-trained deep network to extract features for a shallow classifier. The domain-specific trained network provided higher classification accuracy, compared to the pre-trained model, with an area under the ROC curve of 0.91 versus 0.81, and an accuracy of 0.83 versus 0.71. Similar accuracy was achieved in classifying benign lesions, malignant lesions, and normal tissue images. The trained network was able to improve accuracy by using the multi-channel image representation, and was more robust to reductions in the size of the training set. A small-size convolutional neural network can learn to accurately classify findings in medical images using only a few hundred images from a few dozen patients. With sufficient data augmentation, such a network can be trained to outperform a pre-trained out-of-domain classifier. Developing domain-specific deep-learning models for medical imaging can facilitate technological advancements in computer-aided diagnosis.

Scratching the surface: the processing of pain from deep tissues.

PubMed

Sikandar, Shafaq; Aasvang, Eske Kvanner; Dickenson, Anthony H

2016-04-01

Although most pain research focuses on skin, muscles, joints and viscerae are major sources of pain. We discuss the mechanisms of deep pains arising from somatic and visceral structures and how this can lead to widespread manifestations and chronification. We include how both altered peripheral and central sensory neurotransmission lead to deep pain states and comment on key areas such as top-down modulation where little is known. It is vital that the clinical characterization of deep pain in patients is improved to allow for back translation to preclinical models so that the missing links can be ascertained. The contribution of deeper somatic and visceral tissues to various chronic pain syndromes is common but there is much we need to know.

In vivo multiphoton imaging of a diverse array of fluorophores to investigate deep neurovascular structure

PubMed Central

Miller, David R.; Hassan, Ahmed M.; Jarrett, Jeremy W.; Medina, Flor A.; Perillo, Evan P.; Hagan, Kristen; Shams Kazmi, S. M.; Clark, Taylor A.; Sullender, Colin T.; Jones, Theresa A.; Zemelman, Boris V.; Dunn, Andrew K.

2017-01-01

We perform high-resolution, non-invasive, in vivo deep-tissue imaging of the mouse neocortex using multiphoton microscopy with a high repetition rate optical parametric amplifier laser source tunable between λ=1,100 and 1,400 nm. By combining the high repetition rate (511 kHz) and high pulse energy (400 nJ) of our amplifier laser system, we demonstrate imaging of vasculature labeled with Texas Red and Indocyanine Green, and neurons expressing tdTomato and yellow fluorescent protein. We measure the blood flow speed of a single capillary at a depth of 1.2 mm, and image vasculature to a depth of 1.53 mm with fine axial steps (5 μm) and reasonable acquisition times. The high image quality enabled analysis of vascular morphology at depths to 1.45 mm. PMID:28717582

Speckle contrast optical tomography: A new method for deep tissue three-dimensional tomography of blood flow

PubMed Central

Varma, Hari M.; Valdes, Claudia P.; Kristoffersen, Anna K.; Culver, Joseph P.; Durduran, Turgut

2014-01-01

A novel tomographic method based on the laser speckle contrast, speckle contrast optical tomography (SCOT) is introduced that allows us to reconstruct three dimensional distribution of blood flow in deep tissues. This method is analogous to the diffuse optical tomography (DOT) but for deep tissue blood flow. We develop a reconstruction algorithm based on first Born approximation to generate three dimensional distribution of flow using the experimental data obtained from tissue simulating phantoms. PMID:24761306

Tissues segmentation based on multi spectral medical images

NASA Astrophysics Data System (ADS)

Li, Ya; Wang, Ying

2017-11-01

Each band image contains the most obvious tissue feature according to the optical characteristics of different tissues in different specific bands for multispectral medical images. In this paper, the tissues were segmented by their spectral information at each multispectral medical images. Four Local Binary Patter descriptors were constructed to extract blood vessels based on the gray difference between the blood vessels and their neighbors. The segmented tissue in each band image was merged to a clear image.

Near-infrared spectroscopic tissue imaging for medical applications

DOEpatents

Demos,; Stavros, Staggs [Livermore, CA; Michael, C [Tracy, CA

2006-03-21

Near infrared imaging using elastic light scattering and tissue autofluorescence are explored for medical applications. The approach involves imaging using cross-polarized elastic light scattering and tissue autofluorescence in the Near Infra-Red (NIR) coupled with image processing and inter-image operations to differentiate human tissue components.

Near-infrared spectroscopic tissue imaging for medical applications

DOEpatents

Demos, Stavros [Livermore, CA; Staggs, Michael C [Tracy, CA

2006-12-12

Near infrared imaging using elastic light scattering and tissue autofluorescence are explored for medical applications. The approach involves imaging using cross-polarized elastic light scattering and tissue autofluorescence in the Near Infra-Red (NIR) coupled with image processing and inter-image operations to differentiate human tissue components.

Penetration depth of photons in biological tissues from hyperspectral imaging in shortwave infrared in transmission and reflection geometries.

PubMed

Zhang, Hairong; Salo, Daniel; Kim, David M; Komarov, Sergey; Tai, Yuan-Chuan; Berezin, Mikhail Y

2016-12-01

Measurement of photon penetration in biological tissues is a central theme in optical imaging. A great number of endogenous tissue factors such as absorption, scattering, and anisotropy affect the path of photons in tissue, making it difficult to predict the penetration depth at different wavelengths. Traditional studies evaluating photon penetration at different wavelengths are focused on tissue spectroscopy that does not take into account the heterogeneity within the sample. This is especially critical in shortwave infrared where the individual vibration-based absorption properties of the tissue molecules are affected by nearby tissue components. We have explored the depth penetration in biological tissues from 900 to 1650 nm using Monte–Carlo simulation and a hyperspectral imaging system with Michelson spatial contrast as a metric of light penetration. Chromatic aberration-free hyperspectral images in transmission and reflection geometries were collected with a spectral resolution of 5.27 nm and a total acquisition time of 3 min. Relatively short recording time minimized artifacts from sample drying. Results from both transmission and reflection geometries consistently revealed that the highest spatial contrast in the wavelength range for deep tissue lies within 1300 to 1375 nm; however, in heavily pigmented tissue such as the liver, the range 1550 to 1600 nm is also prominent.

Deep Learning and Its Applications in Biomedicine.

PubMed

Cao, Chensi; Liu, Feng; Tan, Hai; Song, Deshou; Shu, Wenjie; Li, Weizhong; Zhou, Yiming; Bo, Xiaochen; Xie, Zhi

2018-02-01

Advances in biological and medical technologies have been providing us explosive volumes of biological and physiological data, such as medical images, electroencephalography, genomic and protein sequences. Learning from these data facilitates the understanding of human health and disease. Developed from artificial neural networks, deep learning-based algorithms show great promise in extracting features and learning patterns from complex data. The aim of this paper is to provide an overview of deep learning techniques and some of the state-of-the-art applications in the biomedical field. We first introduce the development of artificial neural network and deep learning. We then describe two main components of deep learning, i.e., deep learning architectures and model optimization. Subsequently, some examples are demonstrated for deep learning applications, including medical image classification, genomic sequence analysis, as well as protein structure classification and prediction. Finally, we offer our perspectives for the future directions in the field of deep learning. Copyright © 2018. Production and hosting by Elsevier B.V.

Coherence-Gated Sensorless Adaptive Optics Multiphoton Retinal Imaging.

PubMed

Cua, Michelle; Wahl, Daniel J; Zhao, Yuan; Lee, Sujin; Bonora, Stefano; Zawadzki, Robert J; Jian, Yifan; Sarunic, Marinko V

2016-09-07

Multiphoton microscopy enables imaging deep into scattering tissues. The efficient generation of non-linear optical effects is related to both the pulse duration (typically on the order of femtoseconds) and the size of the focused spot. Aberrations introduced by refractive index inhomogeneity in the sample distort the wavefront and enlarge the focal spot, which reduces the multiphoton signal. Traditional approaches to adaptive optics wavefront correction are not effective in thick or multi-layered scattering media. In this report, we present sensorless adaptive optics (SAO) using low-coherence interferometric detection of the excitation light for depth-resolved aberration correction of two-photon excited fluorescence (TPEF) in biological tissue. We demonstrate coherence-gated SAO TPEF using a transmissive multi-actuator adaptive lens for in vivo imaging in a mouse retina. This configuration has significant potential for reducing the laser power required for adaptive optics multiphoton imaging, and for facilitating integration with existing systems.

Noncontact diffuse correlation spectroscopy for noninvasive deep tissue blood flow measurement

NASA Astrophysics Data System (ADS)

Lin, Yu; He, Lian; Shang, Yu; Yu, Guoqiang

2012-01-01

A noncontact diffuse correlation spectroscopy (DCS) probe has been developed using two separated optical paths for the source and detector. This unique design avoids the interference between the source and detector and allows large source-detector separations for deep tissue blood flow measurements. The noncontact probe has been calibrated against a contact probe in a tissue-like phantom solution and human muscle tissues; flow changes concurrently measured by the two probes are highly correlated in both phantom (R2=0.89, p<10-5) and real-tissue (R2=0.77, p<10-5, n=9) tests. The noncontact DCS holds promise for measuring blood flow in vulnerable (e.g., pressure ulcer) and soft (e.g., breast) tissues without distorting tissue hemodynamic properties.

Nondestructive cryomicro-CT imaging enables structural and molecular analysis of human lung tissue.

PubMed

Vasilescu, Dragoş M; Phillion, André B; Tanabe, Naoya; Kinose, Daisuke; Paige, David F; Kantrowitz, Jacob J; Liu, Gang; Liu, Hanqiao; Fishbane, Nick; Verleden, Stijn E; Vanaudenaerde, Bart M; Lenburg, Marc; Stevenson, Christopher S; Spira, Avrum; Cooper, Joel D; Hackett, Tillie-Louise; Hogg, James C

2017-01-01

Micro-computed tomography (CT) enables three-dimensional (3D) imaging of complex soft tissue structures, but current protocols used to achieve this goal preclude cellular and molecular phenotyping of the tissue. Here we describe a radiolucent cryostage that permits micro-CT imaging of unfixed frozen human lung samples at an isotropic voxel size of (11 µm) 3 under conditions where the sample is maintained frozen at -30°C during imaging. The cryostage was tested for thermal stability to maintain samples frozen up to 8 h. This report describes the methods used to choose the materials required for cryostage construction and demonstrates that whole genome mRNA integrity and expression are not compromised by exposure to micro-CT radiation and that the tissue can be used for immunohistochemistry. The new cryostage provides a novel method enabling integration of 3D tissue structure with cellular and molecular analysis to facilitate the identification of molecular determinants of disease. The described micro-CT cryostage provides a novel way to study the three-dimensional lung structure preserved without the effects of fixatives while enabling subsequent studies of the cellular matrix composition and gene expression. This approach will, for the first time, enable researchers to study structural changes of lung tissues that occur with disease and correlate them with changes in gene or protein signatures. Copyright © 2017 the American Physiological Society.

Study of CT image texture using deep learning techniques

NASA Astrophysics Data System (ADS)

Dutta, Sandeep; Fan, Jiahua; Chevalier, David

2018-03-01

For CT imaging, reduction of radiation dose while improving or maintaining image quality (IQ) is currently a very active research and development topic. Iterative Reconstruction (IR) approaches have been suggested to be able to offer better IQ to dose ratio compared to the conventional Filtered Back Projection (FBP) reconstruction. However, it has been widely reported that often CT image texture from IR is different compared to that from FBP. Researchers have proposed different figure of metrics to quantitate the texture from different reconstruction methods. But there is still a lack of practical and robust method in the field for texture description. This work applied deep learning method for CT image texture study. Multiple dose scans of a 20cm diameter cylindrical water phantom was performed on Revolution CT scanner (GE Healthcare, Waukesha) and the images were reconstructed with FBP and four different IR reconstruction settings. The training images generated were randomly allotted (80:20) to a training and validation set. An independent test set of 256-512 images/class were collected with the same scan and reconstruction settings. Multiple deep learning (DL) networks with Convolution, RELU activation, max-pooling, fully-connected, global average pooling and softmax activation layers were investigated. Impact of different image patch size for training was investigated. Original pixel data as well as normalized image data were evaluated. DL models were reliably able to classify CT image texture with accuracy up to 99%. Results show that the deep learning techniques suggest that CT IR techniques may help lower the radiation dose compared to FBP.

Deep Hashing for Scalable Image Search.

PubMed

Lu, Jiwen; Liong, Venice Erin; Zhou, Jie

2017-05-01

In this paper, we propose a new deep hashing (DH) approach to learn compact binary codes for scalable image search. Unlike most existing binary codes learning methods, which usually seek a single linear projection to map each sample into a binary feature vector, we develop a deep neural network to seek multiple hierarchical non-linear transformations to learn these binary codes, so that the non-linear relationship of samples can be well exploited. Our model is learned under three constraints at the top layer of the developed deep network: 1) the loss between the compact real-valued code and the learned binary vector is minimized, 2) the binary codes distribute evenly on each bit, and 3) different bits are as independent as possible. To further improve the discriminative power of the learned binary codes, we extend DH into supervised DH (SDH) and multi-label SDH by including a discriminative term into the objective function of DH, which simultaneously maximizes the inter-class variations and minimizes the intra-class variations of the learned binary codes with the single-label and multi-label settings, respectively. Extensive experimental results on eight widely used image search data sets show that our proposed methods achieve very competitive results with the state-of-the-arts.

Single-shot T2 mapping using overlapping-echo detachment planar imaging and a deep convolutional neural network.

PubMed

Cai, Congbo; Wang, Chao; Zeng, Yiqing; Cai, Shuhui; Liang, Dong; Wu, Yawen; Chen, Zhong; Ding, Xinghao; Zhong, Jianhui

2018-04-24

An end-to-end deep convolutional neural network (CNN) based on deep residual network (ResNet) was proposed to efficiently reconstruct reliable T 2 mapping from single-shot overlapping-echo detachment (OLED) planar imaging. The training dataset was obtained from simulations that were carried out on SPROM (Simulation with PRoduct Operator Matrix) software developed by our group. The relationship between the original OLED image containing two echo signals and the corresponding T 2 mapping was learned by ResNet training. After the ResNet was trained, it was applied to reconstruct the T 2 mapping from simulation and in vivo human brain data. Although the ResNet was trained entirely on simulated data, the trained network was generalized well to real human brain data. The results from simulation and in vivo human brain experiments show that the proposed method significantly outperforms the echo-detachment-based method. Reliable T 2 mapping with higher accuracy is achieved within 30 ms after the network has been trained, while the echo-detachment-based OLED reconstruction method took approximately 2 min. The proposed method will facilitate real-time dynamic and quantitative MR imaging via OLED sequence, and deep convolutional neural network has the potential to reconstruct maps from complex MRI sequences efficiently. © 2018 International Society for Magnetic Resonance in Medicine.

Using deep learning for content-based medical image retrieval

NASA Astrophysics Data System (ADS)

Sun, Qinpei; Yang, Yuanyuan; Sun, Jianyong; Yang, Zhiming; Zhang, Jianguo

2017-03-01

Content-Based medical image retrieval (CBMIR) is been highly active research area from past few years. The retrieval performance of a CBMIR system crucially depends on the feature representation, which have been extensively studied by researchers for decades. Although a variety of techniques have been proposed, it remains one of the most challenging problems in current CBMIR research, which is mainly due to the well-known "semantic gap" issue that exists between low-level image pixels captured by machines and high-level semantic concepts perceived by human[1]. Recent years have witnessed some important advances of new techniques in machine learning. One important breakthrough technique is known as "deep learning". Unlike conventional machine learning methods that are often using "shallow" architectures, deep learning mimics the human brain that is organized in a deep architecture and processes information through multiple stages of transformation and representation. This means that we do not need to spend enormous energy to extract features manually. In this presentation, we propose a novel framework which uses deep learning to retrieval the medical image to improve the accuracy and speed of a CBIR in integrated RIS/PACS.

3D Deep Learning Angiography (3D-DLA) from C-arm Conebeam CT.

PubMed

Montoya, J C; Li, Y; Strother, C; Chen, G-H

2018-05-01

Deep learning is a branch of artificial intelligence that has demonstrated unprecedented performance in many medical imaging applications. Our purpose was to develop a deep learning angiography method to generate 3D cerebral angiograms from a single contrast-enhanced C-arm conebeam CT acquisition in order to reduce image artifacts and radiation dose. A set of 105 3D rotational angiography examinations were randomly selected from an internal data base. All were acquired using a clinical system in conjunction with a standard injection protocol. More than 150 million labeled voxels from 35 subjects were used for training. A deep convolutional neural network was trained to classify each image voxel into 3 tissue types (vasculature, bone, and soft tissue). The trained deep learning angiography model was then applied for tissue classification into a validation cohort of 8 subjects and a final testing cohort of the remaining 62 subjects. The final vasculature tissue class was used to generate the 3D deep learning angiography images. To quantify the generalization error of the trained model, we calculated the accuracy, sensitivity, precision, and Dice similarity coefficients for vasculature classification in relevant anatomy. The 3D deep learning angiography and clinical 3D rotational angiography images were subjected to a qualitative assessment for the presence of intersweep motion artifacts. Vasculature classification accuracy and 95% CI in the testing dataset were 98.7% (98.3%-99.1%). No residual signal from osseous structures was observed for any 3D deep learning angiography testing cases except for small regions in the otic capsule and nasal cavity compared with 37% (23/62) of the 3D rotational angiographies. Deep learning angiography accurately recreated the vascular anatomy of the 3D rotational angiography reconstructions without a mask. Deep learning angiography reduced misregistration artifacts induced by intersweep motion, and it reduced radiation exposure

Deep UV Native Fluorescence Imaging of Antarctic Cryptoendolithic Communities

NASA Technical Reports Server (NTRS)

Storrie-Lombardi, M. C.; Douglas, S.; Sun, H.; McDonald, G. D.; Bhartia, R.; Nealson, K. H.; Hug, W. F.

2001-01-01

An interdisciplinary team at the Jet Propulsion Laboratory Center for Life Detection has embarked on a project to provide in situ chemical and morphological characterization of Antarctic cryptoendolithic microbial communities. We present here in situ deep ultraviolet (UV) native fluorescence and environmental scanning electron microscopy images transiting 8.5 mm into a sandstone sample from the Antarctic Dry Valleys. The deep ultraviolet imaging system employs 224.3, 248.6, and 325 nm lasers to elicit differential fluorescence and resonance Raman responses from biomolecules and minerals. The 224.3 and 248.6 nm lasers elicit a fluorescence response from the aromatic amino and nucleic acids. Excitation at 325 nm may elicit activity from a variety of biomolecules, but is more likely to elicit mineral fluorescence. The resultant fluorescence images provide in situ chemical and morphological maps of microorganisms and the associated organic matrix. Visible broadband reflectance images provide orientation against the mineral background. Environmental scanning electron micrographs provided detailed morphological information. The technique has made possible the construction of detailed fluorescent maps extending from the surface of an Antarctic sandstone sample to a depth of 8.5 mm. The images detect no evidence of microbial life in the superficial 0.2 mm crustal layer. The black lichen component between 0.3 and 0.5 mm deep absorbs all wavelengths of both laser and broadband illumination. Filamentous deep ultraviolet native fluorescent activity dominates in the white layer between 0.6 mm and 5.0 mm from the surface. These filamentous forms are fungi that continue into the red (iron-rich) region of the sample extending from 5.0 to 8.5 mm. Using differential image subtraction techniques it is possible to identify fungal nuclei. The ultraviolet response is markedly attenuated in this region, apparently from the absorption of ultraviolet light by iron-rich particles coating

Classification of radiolarian images with hand-crafted and deep features

NASA Astrophysics Data System (ADS)

Keçeli, Ali Seydi; Kaya, Aydın; Keçeli, Seda Uzunçimen

2017-12-01

Radiolarians are planktonic protozoa and are important biostratigraphic and paleoenvironmental indicators for paleogeographic reconstructions. Radiolarian paleontology still remains as a low cost and the one of the most convenient way to obtain dating of deep ocean sediments. Traditional methods for identifying radiolarians are time-consuming and cannot scale to the granularity or scope necessary for large-scale studies. Automated image classification will allow making these analyses promptly. In this study, a method for automatic radiolarian image classification is proposed on Scanning Electron Microscope (SEM) images of radiolarians to ease species identification of fossilized radiolarians. The proposed method uses both hand-crafted features like invariant moments, wavelet moments, Gabor features, basic morphological features and deep features obtained from a pre-trained Convolutional Neural Network (CNN). Feature selection is applied over deep features to reduce high dimensionality. Classification outcomes are analyzed to compare hand-crafted features, deep features, and their combinations. Results show that the deep features obtained from a pre-trained CNN are more discriminative comparing to hand-crafted ones. Additionally, feature selection utilizes to the computational cost of classification algorithms and have no negative effect on classification accuracy.

Deep tissue single cell MSC ablation using a fiber laser source to evaluate therapeutic potential in osteogenesis imperfecta

NASA Astrophysics Data System (ADS)

Tehrani, Kayvan F.; Pendleton, Emily G.; Lin, Charles P.; Mortensen, Luke J.

2016-04-01

Osteogenesis imperfecta (OI) is a currently uncurable disease where a mutation in collagen type I yields brittle bones. One potential therapy is transplantation of mesenchymal stem cells (MSCs), but controlling and enhancing transplanted cell survival has proven challenging. Therefore, we use a 2- photon imaging system to study individual transplanted cells in the living bone marrow. We ablated cells deep in the bone marrow and observed minimal collateral damage to surrounding tissue. Future work will evaluate the local impact of transplanted MSCs on bone deposition in vivo.

Time-reversed ultrasonically encoded (TRUE) focusing for deep-tissue optogenetic modulation

NASA Astrophysics Data System (ADS)

Brake, Joshua; Ruan, Haowen; Robinson, J. Elliott; Liu, Yan; Gradinaru, Viviana; Yang, Changhuei

2018-02-01

The problem of optical scattering was long thought to fundamentally limit the depth at which light could be focused through turbid media such as fog or biological tissue. However, recent work in the field of wavefront shaping has demonstrated that by properly shaping the input light field, light can be noninvasively focused to desired locations deep inside scattering media. This has led to the development of several new techniques which have the potential to enhance the capabilities of existing optical tools in biomedicine. Unfortunately, extending these methods to living tissue has a number of challenges related to the requirements for noninvasive guidestar operation, speed, and focusing fidelity. Of existing wavefront shaping methods, time-reversed ultrasonically encoded (TRUE) focusing is well suited for applications in living tissue since it uses ultrasound as a guidestar which enables noninvasive operation and provides compatibility with optical phase conjugation for high-speed operation. In this paper, we will discuss the results of our recent work to apply TRUE focusing for optogenetic modulation, which enables enhanced optogenetic stimulation deep in tissue with a 4-fold spatial resolution improvement in 800-micron thick acute brain slices compared to conventional focusing, and summarize future directions to further extend the impact of wavefront shaping technologies in biomedicine.

A novel biomedical image indexing and retrieval system via deep preference learning.

PubMed

Pang, Shuchao; Orgun, Mehmet A; Yu, Zhezhou

2018-05-01

The traditional biomedical image retrieval methods as well as content-based image retrieval (CBIR) methods originally designed for non-biomedical images either only consider using pixel and low-level features to describe an image or use deep features to describe images but still leave a lot of room for improving both accuracy and efficiency. In this work, we propose a new approach, which exploits deep learning technology to extract the high-level and compact features from biomedical images. The deep feature extraction process leverages multiple hidden layers to capture substantial feature structures of high-resolution images and represent them at different levels of abstraction, leading to an improved performance for indexing and retrieval of biomedical images. We exploit the current popular and multi-layered deep neural networks, namely, stacked denoising autoencoders (SDAE) and convolutional neural networks (CNN) to represent the discriminative features of biomedical images by transferring the feature representations and parameters of pre-trained deep neural networks from another domain. Moreover, in order to index all the images for finding the similarly referenced images, we also introduce preference learning technology to train and learn a kind of a preference model for the query image, which can output the similarity ranking list of images from a biomedical image database. To the best of our knowledge, this paper introduces preference learning technology for the first time into biomedical image retrieval. We evaluate the performance of two powerful algorithms based on our proposed system and compare them with those of popular biomedical image indexing approaches and existing regular image retrieval methods with detailed experiments over several well-known public biomedical image databases. Based on different criteria for the evaluation of retrieval performance, experimental results demonstrate that our proposed algorithms outperform the state

Deblurring adaptive optics retinal images using deep convolutional neural networks.

PubMed

Fei, Xiao; Zhao, Junlei; Zhao, Haoxin; Yun, Dai; Zhang, Yudong

2017-12-01

The adaptive optics (AO) can be used to compensate for ocular aberrations to achieve near diffraction limited high-resolution retinal images. However, many factors such as the limited aberration measurement and correction accuracy with AO, intraocular scatter, imaging noise and so on will degrade the quality of retinal images. Image post processing is an indispensable and economical method to make up for the limitation of AO retinal imaging procedure. In this paper, we proposed a deep learning method to restore the degraded retinal images for the first time. The method directly learned an end-to-end mapping between the blurred and restored retinal images. The mapping was represented as a deep convolutional neural network that was trained to output high-quality images directly from blurry inputs without any preprocessing. This network was validated on synthetically generated retinal images as well as real AO retinal images. The assessment of the restored retinal images demonstrated that the image quality had been significantly improved.

Deblurring adaptive optics retinal images using deep convolutional neural networks

PubMed Central

Fei, Xiao; Zhao, Junlei; Zhao, Haoxin; Yun, Dai; Zhang, Yudong

2017-01-01

The adaptive optics (AO) can be used to compensate for ocular aberrations to achieve near diffraction limited high-resolution retinal images. However, many factors such as the limited aberration measurement and correction accuracy with AO, intraocular scatter, imaging noise and so on will degrade the quality of retinal images. Image post processing is an indispensable and economical method to make up for the limitation of AO retinal imaging procedure. In this paper, we proposed a deep learning method to restore the degraded retinal images for the first time. The method directly learned an end-to-end mapping between the blurred and restored retinal images. The mapping was represented as a deep convolutional neural network that was trained to output high-quality images directly from blurry inputs without any preprocessing. This network was validated on synthetically generated retinal images as well as real AO retinal images. The assessment of the restored retinal images demonstrated that the image quality had been significantly improved. PMID:29296496

Imaging of non-osteochondral tissues in osteoarthritis.

PubMed

Guermazi, A; Roemer, F W; Crema, M D; Englund, M; Hayashi, D

2014-10-01

The aim of this review is to describe imaging techniques for evaluation of non-osteochondral structures such as the synovium, menisci in the knee, labrum in the hip, ligaments and muscles and to review the literature from recent clinical and epidemiological studies of OA. This is a non-systematic narrative review of published literature on imaging of non-osteochondral tissues in OA. PubMed and MEDLINE search for articles published up to 2014, using the keywords osteoarthritis, synovitis, meniscus, labrum, ligaments, plica, muscles, magnetic resonance imaging (MRI), ultrasound, computed tomography (CT), scintigraphy, and positron emission tomography (PET). Published literature showed imaging of non-osteochondral tissues in OA relies primarily on MRI and ultrasound. The use of semiquantitative and quantitative imaging biomarkers of non-osteochondral tissues in clinical and epidemiological OA studies is reported. We highlight studies that have compared both imaging methodologies directly, and those that have established a relationship between imaging biomarkers and clinical outcomes. We provide recommendations as to which imaging protocols should be used to assess disease-specific changes regarding synovium, meniscus in the knee, labrum in the hip, and ligaments, and highlight potential pitfalls in their usage. MRI and ultrasound are currently the most useful imaging modalities for evaluation of non-osteochondral tissues in OA. MRI evaluation of any tissue needs to be performed using appropriate MR pulse sequences. Ultrasound may be particularly useful for evaluation of small joints of the hand. Nuclear medicine and CT play a limited role in imaging of non-osteochondral tissues in OA. Copyright © 2014 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Deep Learning for Lowtextured Image Matching

NASA Astrophysics Data System (ADS)

Kniaz, V. V.; Fedorenko, V. V.; Fomin, N. A.

2018-05-01

Low-textured objects pose challenges for an automatic 3D model reconstruction. Such objects are common in archeological applications of photogrammetry. Most of the common feature point descriptors fail to match local patches in featureless regions of an object. Hence, automatic documentation of the archeological process using Structure from Motion (SfM) methods is challenging. Nevertheless, such documentation is possible with the aid of a human operator. Deep learning-based descriptors have outperformed most of common feature point descriptors recently. This paper is focused on the development of a new Wide Image Zone Adaptive Robust feature Descriptor (WIZARD) based on the deep learning. We use a convolutional auto-encoder to compress discriminative features of a local path into a descriptor code. We build a codebook to perform point matching on multiple images. The matching is performed using the nearest neighbor search and a modified voting algorithm. We present a new "Multi-view Amphora" (Amphora) dataset for evaluation of point matching algorithms. The dataset includes images of an Ancient Greek vase found at Taman Peninsula in Southern Russia. The dataset provides color images, a ground truth 3D model, and a ground truth optical flow. We evaluated the WIZARD descriptor on the "Amphora" dataset to show that it outperforms the SIFT and SURF descriptors on the complex patch pairs.

Effects of neuromuscular joint facilitation on bridging exercises with respect to deep muscle changes.

PubMed

Zhou, Bin; Huang, QiuChen; Zheng, Tao; Huo, Ming; Maruyama, Hitoshi

2015-05-01

[Purpose] This study examined the effects of neuromuscular joint facilitation on bridging exercises by assessing the cross-sectional area of the multifidus muscle and thickness of the musculus transversus abdominis. [Subjects] Twelve healthy men. [Methods] Four exercises were evaluated: (a) supine resting, (b) bridging resistance exercise involving posterior pelvic tilting, (c) bridging resistance exercise involving anterior pelvic tilting, and (d) bridging resistance exercise involving neuromuscular joint facilitation. The cross-sectional area of the multifidus muscle and thickness of the musculus transversus abdominis were measured during each exercise. [Results] The cross-sectional area of the multifidus muscle and thickness of the musculus transversus abdominis were significantly greater in the neuromuscular joint facilitation group than the others. [Conclusion] Neuromuscular joint facilitation intervention improves the function of deep muscles such as the multifidus muscle and musculus transversus abdominis. Therefore, it can be recommended for application in clinical treatments such as that for back pain.

Incidence of deep vein thrombosis in patients undergoing breast reconstruction with autologous tissue transfer.

PubMed

Konoeda, Hisato; Yamaki, Takashi; Hamahata, Atsumori; Ochi, Masakazu; Osada, Atsuyoshi; Hasegawa, Yuki; Kirita, Miho; Sakurai, Hiroyuki

2017-05-01

Background Breast reconstruction is associated with multiple risk factors for venous thromboembolism. However, the incidence of deep vein thrombosis in patients undergoing breast reconstruction is uncertain. Objective The aim of this study was to prospectively evaluate the incidence of deep vein thrombosis in patients undergoing breast reconstruction using autologous tissue transfer and to identify potential risk factors for deep vein thrombosis. Methods Thirty-five patients undergoing breast reconstruction were enrolled. We measured patients' preoperative characteristics including age, body mass index (kg/m 2 ), and risk factors for deep vein thrombosis. The preoperative diameter of each venous segment in the deep veins was measured using duplex ultrasound. All patients received intermittent pneumatic pump and elastic compression stockings for postoperative thromboprophylaxis. Results Among the 35 patients evaluated, 11 (31.4%) were found to have deep vein thrombosis postoperatively, and one patient was found to have pulmonary embolism postoperatively. All instances of deep vein thrombosis developed in the calf and were asymptomatic. Ten of 11 patients underwent free flap transfer, and the remaining one patient received a latissimus dorsi pedicled flap. Deep vein thrombosis incidence did not significantly differ between patients with a free flap or pedicled flap (P = 0.13). Documented risk factors for deep vein thrombosis demonstrated no significant differences between patients with and without deep vein thrombosis. The diameter of the common femoral vein was significantly larger in patients who developed postoperative deep vein thrombosis than in those who did not ( P < 0.05). Conclusions The morbidity of deep vein thrombosis in patients who underwent breast reconstruction using autologous tissue transfer was relatively high. Since only the diameter of the common femoral vein was predictive of developing postoperative deep vein thrombosis, postoperative

Geocoronal Imaging from the Deep Space Gateway

NASA Astrophysics Data System (ADS)

Waldrop, L.; Immel, T.; Clarke, J.; Fillingim, M.; Rider, K.; Qin, J.; Bhattacharyya, D.; Doe, R.

2018-02-01

UV imaging of geocoronal emission at high spatial and temporal resolution from deep space would provide crucial new constraints on global exospheric structure and dynamics, significantly advancing models of space weather and atmospheric escape.

Computed tomography of deep fat masses in multiple symmetrical lipomatosis

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

Enzi, G.; Biondetti, P.R.; Fiore, D.

1982-07-01

Deep fat masses were evaluated by computed tomography (CT) in 15 patients with multiple symmetrical lipomatosis. In 4 patients, peritracheal accumulations of fat were observed. In 3 of them, tracheal compression by lipomatous tissue was demonstrated: 2 were asymptomatic and the third severe respiratory insufficiency secondary to blockage of the air was by the vocal cords as the result of recurrent nerve palsy. In 6 patients, lipomatous tissue occupied the potential space between the spina scapulae and the trapezius, supraspinatus, and infraspinatus muscles. In 2, calcification of lipomatous masses was observed. There was no relationship between extension of subcutaneous fatmore » and accumulation at deep sites. CT facilitates early detection of peritracheal lipomatous tissue and is helpful in follow-up when deep fat accumulation is responsible for space-occupying lesions requiring surgery.« less

Intelligent Detection of Structure from Remote Sensing Images Based on Deep Learning Method

NASA Astrophysics Data System (ADS)

Xin, L.

2018-04-01

Utilizing high-resolution remote sensing images for earth observation has become the common method of land use monitoring. It requires great human participation when dealing with traditional image interpretation, which is inefficient and difficult to guarantee the accuracy. At present, the artificial intelligent method such as deep learning has a large number of advantages in the aspect of image recognition. By means of a large amount of remote sensing image samples and deep neural network models, we can rapidly decipher the objects of interest such as buildings, etc. Whether in terms of efficiency or accuracy, deep learning method is more preponderant. This paper explains the research of deep learning method by a great mount of remote sensing image samples and verifies the feasibility of building extraction via experiments.

Penetration depth of photons in biological tissues from hyperspectral imaging in shortwave infrared in transmission and reflection geometries

PubMed Central

Zhang, Hairong; Salo, Daniel; Kim, David M.; Komarov, Sergey; Tai, Yuan-Chuan; Berezin, Mikhail Y.

2016-01-01

Abstract. Measurement of photon penetration in biological tissues is a central theme in optical imaging. A great number of endogenous tissue factors such as absorption, scattering, and anisotropy affect the path of photons in tissue, making it difficult to predict the penetration depth at different wavelengths. Traditional studies evaluating photon penetration at different wavelengths are focused on tissue spectroscopy that does not take into account the heterogeneity within the sample. This is especially critical in shortwave infrared where the individual vibration-based absorption properties of the tissue molecules are affected by nearby tissue components. We have explored the depth penetration in biological tissues from 900 to 1650 nm using Monte–Carlo simulation and a hyperspectral imaging system with Michelson spatial contrast as a metric of light penetration. Chromatic aberration-free hyperspectral images in transmission and reflection geometries were collected with a spectral resolution of 5.27 nm and a total acquisition time of 3 min. Relatively short recording time minimized artifacts from sample drying. Results from both transmission and reflection geometries consistently revealed that the highest spatial contrast in the wavelength range for deep tissue lies within 1300 to 1375 nm; however, in heavily pigmented tissue such as the liver, the range 1550 to 1600 nm is also prominent. PMID:27930773

Deep Marginalized Sparse Denoising Auto-Encoder for Image Denoising

NASA Astrophysics Data System (ADS)

Ma, Hongqiang; Ma, Shiping; Xu, Yuelei; Zhu, Mingming

2018-01-01

Stacked Sparse Denoising Auto-Encoder (SSDA) has been successfully applied to image denoising. As a deep network, the SSDA network with powerful data feature learning ability is superior to the traditional image denoising algorithms. However, the algorithm has high computational complexity and slow convergence rate in the training. To address this limitation, we present a method of image denoising based on Deep Marginalized Sparse Denoising Auto-Encoder (DMSDA). The loss function of Sparse Denoising Auto-Encoder is marginalized so that it satisfies both sparseness and marginality. The experimental results show that the proposed algorithm can not only outperform SSDA in the convergence speed and training time, but also has better denoising performance than the current excellent denoising algorithms, including both the subjective and objective evaluation of image denoising.

Ship detection leveraging deep neural networks in WorldView-2 images

NASA Astrophysics Data System (ADS)

Yamamoto, T.; Kazama, Y.

2017-10-01

Interpretation of high-resolution satellite images has been so difficult that skilled interpreters must have checked the satellite images manually because of the following issues. One is the requirement of the high detection accuracy rate. The other is the variety of the target, taking ships for example, there are many kinds of ships, such as boat, cruise ship, cargo ship, aircraft carrier, and so on. Furthermore, there are similar appearance objects throughout the image; therefore, it is often difficult even for the skilled interpreters to distinguish what object the pixels really compose. In this paper, we explore the feasibility of object extraction leveraging deep learning with high-resolution satellite images, especially focusing on ship detection. We calculated the detection accuracy using the WorldView-2 images. First, we collected the training images labelled as "ship" and "not ship". After preparing the training data, we defined the deep neural network model to judge whether ships are existing or not, and trained them with about 50,000 training images for each label. Subsequently, we scanned the evaluation image with different resolution windows and extracted the "ship" images. Experimental result shows the effectiveness of the deep learning based object detection.

Deep learning analyzes Helicobacter pylori infection by upper gastrointestinal endoscopy images.

PubMed

Itoh, Takumi; Kawahira, Hiroshi; Nakashima, Hirotaka; Yata, Noriko

2018-02-01

Helicobacter pylori (HP)-associated chronic gastritis can cause mucosal atrophy and intestinal metaplasia, both of which increase the risk of gastric cancer. The accurate diagnosis of HP infection during routine medical checks is important. We aimed to develop a convolutional neural network (CNN), which is a machine-learning algorithm similar to deep learning, capable of recognizing specific features of gastric endoscopy images. The goal behind developing such a system was to detect HP infection early, thus preventing gastric cancer.  For the development of the CNN, we used 179 upper gastrointestinal endoscopy images obtained from 139 patients (65 were HP-positive: ≥ 10 U/mL and 74 were HP-negative: < 3 U/mL on HP IgG antibody assessment). Of the 179 images, 149 were used as training images, and the remaining 30 (15 from HP-negative patients and 15 from HP-positive patients) were set aside to be used as test images. The 149 training images were subjected to data augmentation, which yielded 596 images. We used the CNN to create a learning tool that would recognize HP infection and assessed the decision accuracy of the CNN with the 30 test images by calculating the sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve (AUC).  The sensitivity and specificity of the CNN for the detection of HP infection were 86.7 % and 86.7 %, respectively, and the AUC was 0.956.  CNN-aided diagnosis of HP infection seems feasible and is expected to facilitate and improve diagnosis during health check-ups.

High-throughput sequencing and analysis of the gill tissue transcriptome from the deep-sea hydrothermal vent mussel Bathymodiolus azoricus

PubMed Central

2010-01-01

have established the first tissue transcriptional analysis of a deep-sea hydrothermal vent animal and generated a searchable catalog of genes that provides a direct method of identifying and retrieving vast numbers of novel coding sequences which can be applied in gene expression profiling experiments from a non-conventional model organism. This provides the most comprehensive sequence resource for identifying novel genes currently available for a deep-sea vent organism, in particular, genes putatively involved in immune and inflammatory reactions in vent mussels. The characterization of the B. azoricus transcriptome will facilitate research into biological processes underlying physiological adaptations to hydrothermal vent environments and will provide a basis for expanding our understanding of genes putatively involved in adaptations processes during post-capture long term acclimatization experiments, at "sea-level" conditions, using B. azoricus as a model organism. PMID:20937131

A fiber-optic-based imaging system for nondestructive assessment of cell-seeded tissue-engineered scaffolds.

PubMed

Hofmann, Matthias C; Whited, Bryce M; Criswell, Tracy; Rylander, Marissa Nichole; Rylander, Christopher G; Soker, Shay; Wang, Ge; Xu, Yong

2012-09-01

A major limitation in tissue engineering is the lack of nondestructive methods that assess the development of tissue scaffolds undergoing preconditioning in bioreactors. Due to significant optical scattering in most scaffolding materials, current microscope-based imaging methods cannot "see" through thick and optically opaque tissue constructs. To address this deficiency, we developed a fiber-optic-based imaging method that is capable of nondestructive imaging of fluorescently labeled cells through a thick and optically opaque scaffold, contained in a bioreactor. This imaging modality is based on the local excitation of fluorescent cells, the acquisition of fluorescence through the scaffold, and fluorescence mapping based on the position of the excitation light. To evaluate the capability and accuracy of the imaging system, human endothelial cells (ECs), stably expressing green fluorescent protein (GFP), were imaged through a fibrous scaffold. Without sacrificing the scaffolds, we nondestructively visualized the distribution of GFP-labeled cells through a ~500 μm thick scaffold with cell-level resolution and distinct localization. These results were similar to control images obtained using an optical microscope with direct line-of-sight access. Through a detailed quantitative analysis, we demonstrated that this method achieved a resolution on the order of 20-30 μm, with 10% or less deviation from standard optical microscopy. Furthermore, we demonstrated that the penetration depth of the imaging method exceeded that of confocal laser scanning microscopy by more than a factor of 2. Our imaging method also possesses a working distance (up to 8 cm) much longer than that of a standard confocal microscopy system, which can significantly facilitate bioreactor integration. This method will enable the nondestructive monitoring of ECs seeded on the lumen of a tissue-engineered vascular graft during preconditioning in vitro, as well as for other tissue-engineered constructs

Improved Contrast-Enhanced Ultrasound Imaging With Multiplane-Wave Imaging.

PubMed

Gong, Ping; Song, Pengfei; Chen, Shigao

2018-02-01

Contrast-enhanced ultrasound (CEUS) imaging has great potential for use in new ultrasound clinical applications such as myocardial perfusion imaging and abdominal lesion characterization. In CEUS imaging, contrast agents (i.e., microbubbles) are used to improve contrast between blood and tissue because of their high nonlinearity under low ultrasound pressure. However, the quality of CEUS imaging sometimes suffers from a low signal-to-noise ratio (SNR) in deeper imaging regions when a low mechanical index (MI) is used to avoid microbubble disruption, especially for imaging at off-resonance transmit frequencies. In this paper, we propose a new strategy of combining CEUS sequences with the recently proposed multiplane-wave (MW) compounding method to improve the SNR of CEUS in deeper imaging regions without increasing MI or sacrificing frame rate. The MW-CEUS method emits multiple Hadamard-coded CEUS pulses in each transmission event (i.e., pulse-echo event). The received echo signals first undergo fundamental bandpass filtering (i.e., the filter is centered on the transmit frequency) to eliminate the microbubble's second-harmonic signals because they cannot be encoded by pulse inversion. The filtered signals are then Hadamard decoded and realigned in fast time to recover the signals as they would have been obtained using classic CEUS pulses, followed by designed recombination to cancel the linear tissue responses. The MW-CEUS method significantly improved contrast-to-tissue ratio and SNR of CEUS imaging by transmitting longer coded pulses. The image resolution was also preserved. The microbubble disruption ratio and motion artifacts in MW-CEUS were similar to those of classic CEUS imaging. In addition, the MW-CEUS sequence can be adapted to other transmission coding formats. These properties of MW-CEUS can potentially facilitate CEUS imaging for many clinical applications, especially assessing deep abdominal organs or the heart.

Deep learning methods to guide CT image reconstruction and reduce metal artifacts

NASA Astrophysics Data System (ADS)

Gjesteby, Lars; Yang, Qingsong; Xi, Yan; Zhou, Ye; Zhang, Junping; Wang, Ge

2017-03-01

The rapidly-rising field of machine learning, including deep learning, has inspired applications across many disciplines. In medical imaging, deep learning has been primarily used for image processing and analysis. In this paper, we integrate a convolutional neural network (CNN) into the computed tomography (CT) image reconstruction process. Our first task is to monitor the quality of CT images during iterative reconstruction and decide when to stop the process according to an intelligent numerical observer instead of using a traditional stopping rule, such as a fixed error threshold or a maximum number of iterations. After training on ground truth images, the CNN was successful in guiding an iterative reconstruction process to yield high-quality images. Our second task is to improve a sinogram to correct for artifacts caused by metal objects. A large number of interpolation and normalization-based schemes were introduced for metal artifact reduction (MAR) over the past four decades. The NMAR algorithm is considered a state-of-the-art method, although residual errors often remain in the reconstructed images, especially in cases of multiple metal objects. Here we merge NMAR with deep learning in the projection domain to achieve additional correction in critical image regions. Our results indicate that deep learning can be a viable tool to address CT reconstruction challenges.

Coherence-Gated Sensorless Adaptive Optics Multiphoton Retinal Imaging

PubMed Central

Cua, Michelle; Wahl, Daniel J.; Zhao, Yuan; Lee, Sujin; Bonora, Stefano; Zawadzki, Robert J.; Jian, Yifan; Sarunic, Marinko V.

2016-01-01

Multiphoton microscopy enables imaging deep into scattering tissues. The efficient generation of non-linear optical effects is related to both the pulse duration (typically on the order of femtoseconds) and the size of the focused spot. Aberrations introduced by refractive index inhomogeneity in the sample distort the wavefront and enlarge the focal spot, which reduces the multiphoton signal. Traditional approaches to adaptive optics wavefront correction are not effective in thick or multi-layered scattering media. In this report, we present sensorless adaptive optics (SAO) using low-coherence interferometric detection of the excitation light for depth-resolved aberration correction of two-photon excited fluorescence (TPEF) in biological tissue. We demonstrate coherence-gated SAO TPEF using a transmissive multi-actuator adaptive lens for in vivo imaging in a mouse retina. This configuration has significant potential for reducing the laser power required for adaptive optics multiphoton imaging, and for facilitating integration with existing systems. PMID:27599635

Multimodal imaging of cutaneous wound tissue

NASA Astrophysics Data System (ADS)

Zhang, Shiwu; Gnyawali, Surya; Huang, Jiwei; Ren, Wenqi; Gordillo, Gayle; Sen, Chandan K.; Xu, Ronald

2015-01-01

Quantitative assessment of wound tissue ischemia, perfusion, and inflammation provides critical information for appropriate detection, staging, and treatment of chronic wounds. However, few methods are available for simultaneous assessment of these tissue parameters in a noninvasive and quantitative fashion. We integrated hyperspectral, laser speckle, and thermographic imaging modalities in a single-experimental setup for multimodal assessment of tissue oxygenation, perfusion, and inflammation characteristics. Algorithms were developed for appropriate coregistration between wound images acquired by different imaging modalities at different times. The multimodal wound imaging system was validated in an occlusion experiment, where oxygenation and perfusion maps of a healthy subject's upper extremity were continuously monitored during a postocclusive reactive hyperemia procedure and compared with standard measurements. The system was also tested in a clinical trial where a wound of three millimeters in diameter was introduced on a healthy subject's lower extremity and the healing process was continuously monitored. Our in vivo experiments demonstrated the clinical feasibility of multimodal cutaneous wound imaging.

Deep Spitzer/IRAC Imaging of the Subaru Deep Field

NASA Astrophysics Data System (ADS)

Jiang, Linhua; Egami, Eiichi; Cohen, Seth; Fan, Xiaohui; Ly, Chun; Mechtley, Matthew; Windhorst, Rogier

2013-10-01

The last decade saw great progress in our understanding of the distant Universe as a number of objects at z > 6 were discovered. The Subaru Deep Field (SDF) project has played an important role on study of high-z galaxies. The SDF is unique: it covers a large area of 850 sq arcmin; it has extremely deep optical images in a series of broad and narrow bands; it has the largest sample of spectroscopically-confirmed galaxies known at z >= 6, including ~100 Lyman alpha emitters (LAEs) and ~50 Lyman break galaxies (LBGs). Here we propose to carry out deep IRAC imaging observations of the central 75% of the SDF. The proposed observations together with those from our previous Spitzer programs will reach a depth of ~10 hours, and enable the first complete census of physical properties and stellar populations of spectroscopically-confirmed galaxies at the end of cosmic reionization. IRAC data is the key to measure stellar masses and constrain stellar populations in high-z galaxies. From SED modeling with secure redshifts, we will characterize the physical properties of these galaxies, and trace their mass assembly and star formation history. In particular, it allows us, for the first time, to study stellar populations in a large sample of z >=6 LAEs. We will also address some critical questions, such as whether LAEs and LBGs represent physically different galaxy populations. All these will help us to understand the earliest galaxy formation and evolution, and better constrain the galaxy contribution to reionization. The IRAC data will also cover 10,000 emission-line selected galaxies at z < 1.5, 50,000 UV and mass selected LBGs at 1.5 < z < 3, and more than 5,000 LBGs at 3 < z < 6. It will have a legacy value for SDF-related programs.

Detection of soft tissue densities from digital breast tomosynthesis: comparison of conventional and deep learning approaches

NASA Astrophysics Data System (ADS)

Fotin, Sergei V.; Yin, Yin; Haldankar, Hrishikesh; Hoffmeister, Jeffrey W.; Periaswamy, Senthil

2016-03-01

Computer-aided detection (CAD) has been used in screening mammography for many years and is likely to be utilized for digital breast tomosynthesis (DBT). Higher detection performance is desirable as it may have an impact on radiologist's decisions and clinical outcomes. Recently the algorithms based on deep convolutional architectures have been shown to achieve state of the art performance in object classification and detection. Similarly, we trained a deep convolutional neural network directly on patches sampled from two-dimensional mammography and reconstructed DBT volumes and compared its performance to a conventional CAD algorithm that is based on computation and classification of hand-engineered features. The detection performance was evaluated on the independent test set of 344 DBT reconstructions (GE SenoClaire 3D, iterative reconstruction algorithm) containing 328 suspicious and 115 malignant soft tissue densities including masses and architectural distortions. Detection sensitivity was measured on a region of interest (ROI) basis at the rate of five detection marks per volume. Moving from conventional to deep learning approach resulted in increase of ROI sensitivity from 0:832 +/- 0:040 to 0:893 +/- 0:033 for suspicious ROIs; and from 0:852 +/- 0:065 to 0:930 +/- 0:046 for malignant ROIs. These results indicate the high utility of deep feature learning in the analysis of DBT data and high potential of the method for broader medical image analysis tasks.

Deep tissue imaging of microfracture and non-displaced fracture of bone using the second and third near-infrared therapeutic windows

NASA Astrophysics Data System (ADS)

Sordillo, Laura A.; Pu, Yang; Sordillo, P. P.; Budansky, Yury; Alfano, Robert R.

2014-03-01

Near-infrared (NIR) light in the wavelengths of 700 nm to 2,000 nm has three NIR optical, or therapeutic, windows, which allow for deeper depth penetration in scattering tissue media. Microfractures secondary to repetitive stress, particularly in the lower extremities, are an important problem for military recruits and athletes. They also frequently occur in the elderly, or in patients taking bisphosphonates or denosumab. Microfractures can be early predictors of a major bone fracture. Using the second and third NIR therapeutic windows, we investigated the results from images of chicken bone and human tibial bone with microfractures and non-displaced fractures with and without overlying tissues of various thicknesses. Images of bone with microfractures and non-displaced fractures with tissue show scattering photons in the third NIR window with wavelengths between 1,650 nm and 1,870 nm are diminished and absorption is increased slightly from and second NIR windows. Results from images of fractured bones show the attenuation length of light through tissue in the third optical window to be larger than in the second therapeutic window. Use of these windows may aid in the detection of bone microfractures, and thus reduce the incidence of major bone fracture in susceptible groups.

Physics of tissue harmonic imaging by ultrasound

NASA Astrophysics Data System (ADS)

Jing, Yuan

Tissue Harmonic Imaging (THI) is an imaging modality that is currently deployed on diagnostic ultrasound scanners. In THI the amplitude of the ultrasonic pulse that is used to probe the tissue is large enough that the pulse undergoes nonlinear distortion as it propagates into the tissue. One result of the distortion is that as the pulse propagates energy is shifted from the fundamental frequency of the source pulse into its higher harmonics. These harmonics will scatter off objects in the tissue and images formed from the scattered higher harmonics are considered to have superior quality to the images formed from the fundamental frequency. Processes that have been suggested as possibly responsible for the improved imaging in THI include: (1) reduced sensitivity to reverberation, (2) reduced sensitivity to aberration, and (3) reduction in side lobes. By using a combination of controlled experiments and numerical simulations, these three reasons have been investigated. A single element transducer and a clinical ultrasound scanner with a phased array transducer were used to image a commercial tissue-mimicking phantom with calibrated targets. The higher image quality achieved with THI was quantified in terms of spatial resolution and "clutter" signals. A three-dimensional model of the forward propagation of nonlinear sound beams in media with arbitrary spatial properties (a generalized KZK equation) was developed. A time-domain code for solving the KZK equation was validated with measurements of the acoustic field generated by the single element transducer and the phased array transducer. The code was used to investigate the impact of aberration using tissue-like media with three-dimensional variations in all acoustic properties. The three-dimensional maps of tissue properties were derived from the datasets available through the Visible Female project. The experiments and simulations demonstrated that second harmonic imaging (1) suffers less clutter associated with

Activatable Fluorescence Probe via Self-Immolative Intramolecular Cyclization for Histone Deacetylase Imaging in Live Cells and Tissues.

PubMed

Liu, Xianjun; Xiang, Meihao; Tong, Zongxuan; Luo, Fengyan; Chen, Wen; Liu, Feng; Wang, Fenglin; Yu, Ru-Qin; Jiang, Jian-Hui

2018-05-01

Histone deacetylases (HDACs) play essential roles in transcription regulation and are valuable theranostic targets. However, there are no activatable fluorescent probes for imaging of HDAC activity in live cells. Here, we develop for the first time a novel activatable two-photon fluorescence probe that enables in situ imaging of HDAC activity in living cells and tissues. The probe is designed by conjugating an acetyl-lysine mimic substrate to a masked aldehyde-containing fluorophore via a cyanoester linker. Upon deacetylation by HDAC, the probe undergoes a rapid self-immolative intramolecular cyclization reaction, producing a cyanohydrin intermediate that is spontaneously rapidly decomposed into the highly fluorescent aldehyde-containing two-photon fluorophore. The probe is shown to exhibit high sensitivity, high specificity, and fast response for HDAC detection in vitro. Imaging studies reveal that the probe is able to directly visualize and monitor HDAC activity in living cells. Moreover, the probe is demonstrated to have the capability of two-photon imaging of HDAC activity in deep tissue slices up to 130 μm. This activatable fluorescent probe affords a useful tool for evaluating HDAC activity and screening HDAC-targeting drugs in both live cell and tissue assays.

Geographical topic learning for social images with a deep neural network

NASA Astrophysics Data System (ADS)

Feng, Jiangfan; Xu, Xin

2017-03-01

The use of geographical tagging in social-media images is becoming a part of image metadata and a great interest for geographical information science. It is well recognized that geographical topic learning is crucial for geographical annotation. Existing methods usually exploit geographical characteristics using image preprocessing, pixel-based classification, and feature recognition. How to effectively exploit the high-level semantic feature and underlying correlation among different types of contents is a crucial task for geographical topic learning. Deep learning (DL) has recently demonstrated robust capabilities for image tagging and has been introduced into geoscience. It extracts high-level features computed from a whole image component, where the cluttered background may dominate spatial features in the deep representation. Therefore, a method of spatial-attentional DL for geographical topic learning is provided and we can regard it as a special case of DL combined with various deep networks and tuning tricks. Results demonstrated that the method is discriminative for different types of geographical topic learning. In addition, it outperforms other sequential processing models in a tagging task for a geographical image dataset.

Phase contrast imaging of cochlear soft tissue.

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

Smith, S.; Hwang, M.; Rau, C.

A noninvasive technique to image soft tissue could expedite diagnosis and disease management in the auditory system. We propose inline phase contrast imaging with hard X-rays as a novel method that overcomes the limitations of conventional absorption radiography for imaging soft tissue. In this study, phase contrast imaging of mouse cochleae was performed at the Argonne National Laboratory Advanced Photon Source. The phase contrast tomographic reconstructions show soft tissue structures of the cochlea, including the inner pillar cells, the inner spiral sulcus, the tectorial membrane, the basilar membrane, and the Reissner's membrane. The results suggest that phase contrast X-ray imagingmore » and tomographic techniques hold promise to noninvasively image cochlear structures at an unprecedented cellular level.« less

VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images.

PubMed

Chen, Hao; Dou, Qi; Yu, Lequan; Qin, Jing; Heng, Pheng-Ann

2018-04-15

Segmentation of key brain tissues from 3D medical images is of great significance for brain disease diagnosis, progression assessment and monitoring of neurologic conditions. While manual segmentation is time-consuming, laborious, and subjective, automated segmentation is quite challenging due to the complicated anatomical environment of brain and the large variations of brain tissues. We propose a novel voxelwise residual network (VoxResNet) with a set of effective training schemes to cope with this challenging problem. The main merit of residual learning is that it can alleviate the degradation problem when training a deep network so that the performance gains achieved by increasing the network depth can be fully leveraged. With this technique, our VoxResNet is built with 25 layers, and hence can generate more representative features to deal with the large variations of brain tissues than its rivals using hand-crafted features or shallower networks. In order to effectively train such a deep network with limited training data for brain segmentation, we seamlessly integrate multi-modality and multi-level contextual information into our network, so that the complementary information of different modalities can be harnessed and features of different scales can be exploited. Furthermore, an auto-context version of the VoxResNet is proposed by combining the low-level image appearance features, implicit shape information, and high-level context together for further improving the segmentation performance. Extensive experiments on the well-known benchmark (i.e., MRBrainS) of brain segmentation from 3D magnetic resonance (MR) images corroborated the efficacy of the proposed VoxResNet. Our method achieved the first place in the challenge out of 37 competitors including several state-of-the-art brain segmentation methods. Our method is inherently general and can be readily applied as a powerful tool to many brain-related studies, where accurate segmentation of brain

Deep multi-scale convolutional neural network for hyperspectral image classification

NASA Astrophysics Data System (ADS)

Zhang, Feng-zhe; Yang, Xia

2018-04-01

In this paper, we proposed a multi-scale convolutional neural network for hyperspectral image classification task. Firstly, compared with conventional convolution, we utilize multi-scale convolutions, which possess larger respective fields, to extract spectral features of hyperspectral image. We design a deep neural network with a multi-scale convolution layer which contains 3 different convolution kernel sizes. Secondly, to avoid overfitting of deep neural network, dropout is utilized, which randomly sleeps neurons, contributing to improve the classification accuracy a bit. In addition, new skills like ReLU in deep learning is utilized in this paper. We conduct experiments on University of Pavia and Salinas datasets, and obtained better classification accuracy compared with other methods.

Pitfalls of CT for deep neck abscess imaging assessment: a retrospective review of 162 cases.

PubMed

Chuang, S Y; Lin, H T; Wen, Y S; Hsu, F J

2013-01-01

To investigate the diagnostic value of contrast-enhanced computed tomography (CT) for the prediction of deep neck abscesses in different deep neck spaces and to evaluate the false-positive results. We retrospectively analysed the clinical charts, CT examinations, surgical findings, bacteriology, pathological examinations and complications of hospitalised patients with a diagnosis of deep neck abscess from 2004 to 2010. The positive predictive values (PPV) for the prediction of abscesses by CT scan in different deep neck spaces were calculated individually on the basis of surgical findings. A total of 162 patients were included in this study. All patients received both intravenous antibiotics and surgical drainage. The parapharyngeal space was the most commonly involved space. The overall PPV for the prediction of deep neck abscess with contrast-enhanced CT was 79.6%. The PPV was 91.3% when more than one deep neck space was involved but only 50.0% in patients with isolated retropharyngeal abscesses. In the false-positive group, cellulitis was the most common final result, followed by cystic degeneration of cervical metastases. Five specimens taken intra-operatively revealed malignancy and four of these were not infected. There are some limitations affecting the differentiation of abscesses and cellulitis, particularly in the retropharyngeal space. A central necrotic cervical metastatic lymph node may sometimes also mimic a simple pyogenic deep neck abscess on both clinical pictures and CT images. Routine biopsy of the tissue must be performed during surgical drainage.

A Deep Convolutional Neural Network for segmenting and classifying epithelial and stromal regions in histopathological images

PubMed Central

Xu, Jun; Luo, Xiaofei; Wang, Guanhao; Gilmore, Hannah; Madabhushi, Anant

2016-01-01

Epithelial (EP) and stromal (ST) are two types of tissues in histological images. Automated segmentation or classification of EP and ST tissues is important when developing computerized system for analyzing the tumor microenvironment. In this paper, a Deep Convolutional Neural Networks (DCNN) based feature learning is presented to automatically segment or classify EP and ST regions from digitized tumor tissue microarrays (TMAs). Current approaches are based on handcraft feature representation, such as color, texture, and Local Binary Patterns (LBP) in classifying two regions. Compared to handcrafted feature based approaches, which involve task dependent representation, DCNN is an end-to-end feature extractor that may be directly learned from the raw pixel intensity value of EP and ST tissues in a data driven fashion. These high-level features contribute to the construction of a supervised classifier for discriminating the two types of tissues. In this work we compare DCNN based models with three handcraft feature extraction based approaches on two different datasets which consist of 157 Hematoxylin and Eosin (H&E) stained images of breast cancer and 1376 immunohistological (IHC) stained images of colorectal cancer, respectively. The DCNN based feature learning approach was shown to have a F1 classification score of 85%, 89%, and 100%, accuracy (ACC) of 84%, 88%, and 100%, and Matthews Correlation Coefficient (MCC) of 86%, 77%, and 100% on two H&E stained (NKI and VGH) and IHC stained data, respectively. Our DNN based approach was shown to outperform three handcraft feature extraction based approaches in terms of the classification of EP and ST regions. PMID:28154470

A Deep Convolutional Neural Network for segmenting and classifying epithelial and stromal regions in histopathological images.

PubMed

Xu, Jun; Luo, Xiaofei; Wang, Guanhao; Gilmore, Hannah; Madabhushi, Anant

2016-05-26

Epithelial (EP) and stromal (ST) are two types of tissues in histological images. Automated segmentation or classification of EP and ST tissues is important when developing computerized system for analyzing the tumor microenvironment. In this paper, a Deep Convolutional Neural Networks (DCNN) based feature learning is presented to automatically segment or classify EP and ST regions from digitized tumor tissue microarrays (TMAs). Current approaches are based on handcraft feature representation, such as color, texture, and Local Binary Patterns (LBP) in classifying two regions. Compared to handcrafted feature based approaches, which involve task dependent representation, DCNN is an end-to-end feature extractor that may be directly learned from the raw pixel intensity value of EP and ST tissues in a data driven fashion. These high-level features contribute to the construction of a supervised classifier for discriminating the two types of tissues. In this work we compare DCNN based models with three handcraft feature extraction based approaches on two different datasets which consist of 157 Hematoxylin and Eosin (H&E) stained images of breast cancer and 1376 immunohistological (IHC) stained images of colorectal cancer, respectively. The DCNN based feature learning approach was shown to have a F1 classification score of 85%, 89%, and 100%, accuracy (ACC) of 84%, 88%, and 100%, and Matthews Correlation Coefficient (MCC) of 86%, 77%, and 100% on two H&E stained (NKI and VGH) and IHC stained data, respectively. Our DNN based approach was shown to outperform three handcraft feature extraction based approaches in terms of the classification of EP and ST regions.

Deep learning for tumor classification in imaging mass spectrometry.

PubMed

Behrmann, Jens; Etmann, Christian; Boskamp, Tobias; Casadonte, Rita; Kriegsmann, Jörg; Maaß, Peter

2018-04-01

Tumor classification using imaging mass spectrometry (IMS) data has a high potential for future applications in pathology. Due to the complexity and size of the data, automated feature extraction and classification steps are required to fully process the data. Since mass spectra exhibit certain structural similarities to image data, deep learning may offer a promising strategy for classification of IMS data as it has been successfully applied to image classification. Methodologically, we propose an adapted architecture based on deep convolutional networks to handle the characteristics of mass spectrometry data, as well as a strategy to interpret the learned model in the spectral domain based on a sensitivity analysis. The proposed methods are evaluated on two algorithmically challenging tumor classification tasks and compared to a baseline approach. Competitiveness of the proposed methods is shown on both tasks by studying the performance via cross-validation. Moreover, the learned models are analyzed by the proposed sensitivity analysis revealing biologically plausible effects as well as confounding factors of the considered tasks. Thus, this study may serve as a starting point for further development of deep learning approaches in IMS classification tasks. https://gitlab.informatik.uni-bremen.de/digipath/Deep_Learning_for_Tumor_Classification_in_IMS. jbehrmann@uni-bremen.de or christianetmann@uni-bremen.de. Supplementary data are available at Bioinformatics online.

Accelerating Spaceborne SAR Imaging Using Multiple CPU/GPU Deep Collaborative Computing.

PubMed

Zhang, Fan; Li, Guojun; Li, Wei; Hu, Wei; Hu, Yuxin

2016-04-07

With the development of synthetic aperture radar (SAR) technologies in recent years, the huge amount of remote sensing data brings challenges for real-time imaging processing. Therefore, high performance computing (HPC) methods have been presented to accelerate SAR imaging, especially the GPU based methods. In the classical GPU based imaging algorithm, GPU is employed to accelerate image processing by massive parallel computing, and CPU is only used to perform the auxiliary work such as data input/output (IO). However, the computing capability of CPU is ignored and underestimated. In this work, a new deep collaborative SAR imaging method based on multiple CPU/GPU is proposed to achieve real-time SAR imaging. Through the proposed tasks partitioning and scheduling strategy, the whole image can be generated with deep collaborative multiple CPU/GPU computing. In the part of CPU parallel imaging, the advanced vector extension (AVX) method is firstly introduced into the multi-core CPU parallel method for higher efficiency. As for the GPU parallel imaging, not only the bottlenecks of memory limitation and frequent data transferring are broken, but also kinds of optimized strategies are applied, such as streaming, parallel pipeline and so on. Experimental results demonstrate that the deep CPU/GPU collaborative imaging method enhances the efficiency of SAR imaging on single-core CPU by 270 times and realizes the real-time imaging in that the imaging rate outperforms the raw data generation rate.

Accelerating Spaceborne SAR Imaging Using Multiple CPU/GPU Deep Collaborative Computing

PubMed Central

Zhang, Fan; Li, Guojun; Li, Wei; Hu, Wei; Hu, Yuxin

2016-01-01

With the development of synthetic aperture radar (SAR) technologies in recent years, the huge amount of remote sensing data brings challenges for real-time imaging processing. Therefore, high performance computing (HPC) methods have been presented to accelerate SAR imaging, especially the GPU based methods. In the classical GPU based imaging algorithm, GPU is employed to accelerate image processing by massive parallel computing, and CPU is only used to perform the auxiliary work such as data input/output (IO). However, the computing capability of CPU is ignored and underestimated. In this work, a new deep collaborative SAR imaging method based on multiple CPU/GPU is proposed to achieve real-time SAR imaging. Through the proposed tasks partitioning and scheduling strategy, the whole image can be generated with deep collaborative multiple CPU/GPU computing. In the part of CPU parallel imaging, the advanced vector extension (AVX) method is firstly introduced into the multi-core CPU parallel method for higher efficiency. As for the GPU parallel imaging, not only the bottlenecks of memory limitation and frequent data transferring are broken, but also kinds of optimized strategies are applied, such as streaming, parallel pipeline and so on. Experimental results demonstrate that the deep CPU/GPU collaborative imaging method enhances the efficiency of SAR imaging on single-core CPU by 270 times and realizes the real-time imaging in that the imaging rate outperforms the raw data generation rate. PMID:27070606

Comparison of mechanisms involved in image enhancement of Tissue Harmonic Imaging

NASA Astrophysics Data System (ADS)

Cleveland, Robin O.; Jing, Yuan

2006-05-01

Processes that have been suggested as responsible for the improved imaging in Tissue Harmonic Imaging (THI) include: 1) reduced sensitivity to reverberation, 2) reduced sensitivity to aberration, and 3) reduction in the amplitude of diffraction side lobes. A three-dimensional model of the forward propagation of nonlinear sound beams in media with arbitrary spatial properties (a generalized KZK equation) was developed and solved using a time-domain code. The numerical simulations were validated through experiments with tissue mimicking phantoms. The impact of aberration from tissue-like media was determined through simulations using three-dimensional maps of tissue properties derived from datasets available through the Visible Female Project. The experiments and simulations demonstrated that second harmonic imaging suffers less clutter from reverberation and side-lobes but is not immune to aberration effects. The results indicate that side lobe suppression is the most significant reason for the improvement of second harmonic imaging.

Imaging the hard/soft tissue interface.

PubMed

Bannerman, Alistair; Paxton, Jennifer Z; Grover, Liam M

2014-03-01

Interfaces between different tissues play an essential role in the biomechanics of native tissues and their recapitulation is now recognized as critical to function. As a consequence, imaging the hard/soft tissue interface has become increasingly important in the area of tissue engineering. Particularly as several biotechnology based products have made it onto the market or are close to human trials and an understanding of their function and development is essential. A range of imaging modalities have been developed that allow a wealth of information on the morphological and physical properties of samples to be obtained non-destructively in vivo or via destructive means. This review summarizes the use of a selection of imaging modalities on interfaces to date considering the strengths and weaknesses of each. We will also consider techniques which have not yet been utilized to their full potential or are likely to play a role in future work in the area.

Earth-from-Luna Limb Imager (ELLI) for Deep Space Gateway

NASA Astrophysics Data System (ADS)

Gorkavyi, N.; DeLand, M.

2018-02-01

The new type of limb imager with a high-frequency imaging proposed for Deep Space Gateway. Each day this CubeSat' scale imager will generate the global 3D model of the aerosol component of the Earth's atmosphere and Polar Mesospheric Clouds.

Application of Deep Learning in Automated Analysis of Molecular Images in Cancer: A Survey

PubMed Central

Xue, Yong; Chen, Shihui; Liu, Yong

2017-01-01

Molecular imaging enables the visualization and quantitative analysis of the alterations of biological procedures at molecular and/or cellular level, which is of great significance for early detection of cancer. In recent years, deep leaning has been widely used in medical imaging analysis, as it overcomes the limitations of visual assessment and traditional machine learning techniques by extracting hierarchical features with powerful representation capability. Research on cancer molecular images using deep learning techniques is also increasing dynamically. Hence, in this paper, we review the applications of deep learning in molecular imaging in terms of tumor lesion segmentation, tumor classification, and survival prediction. We also outline some future directions in which researchers may develop more powerful deep learning models for better performance in the applications in cancer molecular imaging. PMID:29114182

ST Spot Detector: a web-based application for automatic spot and tissue detection for spatial Transcriptomics image datasets.

PubMed

Wong, Kim; Navarro, José Fernández; Bergenstråhle, Ludvig; Ståhl, Patrik L; Lundeberg, Joakim

2018-06-01

Spatial Transcriptomics (ST) is a method which combines high resolution tissue imaging with high troughput transcriptome sequencing data. This data must be aligned with the images for correct visualization, a process that involves several manual steps. Here we present ST Spot Detector, a web tool that automates and facilitates this alignment through a user friendly interface. jose.fernandez.navarro@scilifelab.se. Supplementary data are available at Bioinformatics online.

Correlative Imaging of Fluorescent Proteins in Resin-Embedded Plant Material1

PubMed Central

Bell, Karen; Mitchell, Steve; Paultre, Danae; Posch, Markus; Oparka, Karl

2013-01-01

Fluorescent proteins (FPs) were developed for live-cell imaging and have revolutionized cell biology. However, not all plant tissues are accessible to live imaging using confocal microscopy, necessitating alternative approaches for protein localization. An example is the phloem, a tissue embedded deep within plant organs and sensitive to damage. To facilitate accurate localization of FPs within recalcitrant tissues, we developed a simple method for retaining FPs after resin embedding. This method is based on low-temperature fixation and dehydration, followed by embedding in London Resin White, and avoids the need for cryosections. We show that a palette of FPs can be localized in plant tissues while retaining good structural cell preservation, and that the polymerized block face can be counterstained with cell wall probes. Using this method we have been able to image green fluorescent protein-labeled plasmodesmata to a depth of more than 40 μm beneath the resin surface. Using correlative light and electron microscopy of the phloem, we were able to locate the same FP-labeled sieve elements in semithin and ultrathin sections. Sections were amenable to antibody labeling, and allowed a combination of confocal and superresolution imaging (three-dimensional-structured illumination microscopy) on the same cells. These correlative imaging methods should find several uses in plant cell biology. PMID:23457228

Bioluminescence-Activated Deep-Tissue Photodynamic Therapy of Cancer

PubMed Central

Kim, Yi Rang; Kim, Seonghoon; Choi, Jin Woo; Choi, Sung Yong; Lee, Sang-Hee; Kim, Homin; Hahn, Sei Kwang; Koh, Gou Young; Yun, Seok Hyun

2015-01-01

Optical energy can trigger a variety of photochemical processes useful for therapies. Owing to the shallow penetration of light in tissues, however, the clinical applications of light-activated therapies have been limited. Bioluminescence resonant energy transfer (BRET) may provide a new way of inducing photochemical activation. Here, we show that efficient bioluminescence energy-induced photodynamic therapy (PDT) of macroscopic tumors and metastases in deep tissue. For monolayer cell culture in vitro incubated with Chlorin e6, BRET energy of about 1 nJ per cell generated as strong cytotoxicity as red laser light irradiation at 2.2 mW/cm2 for 180 s. Regional delivery of bioluminescence agents via draining lymphatic vessels killed tumor cells spread to the sentinel and secondary lymph nodes, reduced distant metastases in the lung and improved animal survival. Our results show the promising potential of novel bioluminescence-activated PDT. PMID:26000054

Optical metabolic imaging of live tissue cultures

NASA Astrophysics Data System (ADS)

Walsh, Alex J.; Cook, Rebecca S.; Arteaga, Carlos L.; Skala, Melissa C.

2013-02-01

The fluorescence properties, both intensity and fluorescence lifetime, of NADH and FAD, two coenzymes of metabolism, are sensitive, high resolution measures of cellular metabolism. However, often in vivo measurements of tissue are not feasible. In this study, we investigate the stability over time of two-photon auto-fluorescence imaging of NADH and FAD in live-cultured tissues. Our results demonstrate that cultured tissues remain viable for at least several days post excision. Furthermore, the optical redox ratio, NADH fluorescence lifetime, and FAD fluorescence lifetime do not significantly change in the cultured tissues over time. With these findings, we demonstrate the potential of sustained tissue culture techniques for optical metabolic imaging.

Monitoring of tissue modification with optical coherence tomography

NASA Astrophysics Data System (ADS)

Zhang, Wei; Luo, Qingming; Yao, Lei; Cheng, Haiying; Zeng, Shaoqun

2002-04-01

An experimental monitoring of tissue modification of in vitro and in vivo rabbit dura mater with administration of osmotical agents, 40% glucose solution and glycerol, using optical coherence tomography was presented. The preliminary results of experimental study of influence of osmotical liquids (glucose solutions, glycerol) of rabbit dura mater were reported. The significant decreasing of the light from surface and increasing of the light from the deep of dura mater under action of osmotical solutions and the increasing of OCT imaging depth were demonstrated. Experiments showed that administration of osmolytes to dura mater allowed for effective and temporary control of its optical characteristics, which made dura mater more transparent, increased the ability of light penetrating the tissue, and consequently improved the optical imaging depth. It is a significant study, which can improve penetration of optical imaging of cerebral function and acquire more information of the deep brain tissue.

Deformation and reperfusion damages and their accumulation in subcutaneous tissues during loading and unloading: a theoretical modeling of deep tissue injuries.

PubMed

Mak, Arthur F T; Yu, Yanyan; Kwan, Linda P C; Sun, Lei; Tam, Eric W C

2011-11-21

Deep tissue injuries (DTI) involve damages in the subcutaneous tissues under intact skin incurred by prolonged excessive epidermal loadings. This paper presents a new theoretical model for the development of DTI, broadly based on the experimental evidence in the literatures. The model covers the loading damages implicitly inclusive of both the direct mechanical and ischemic injuries, and the additional reperfusion damages and the competing healing processes during the unloading phase. Given the damage accumulated at the end of the loading period, the relative strength of the reperfusion and the healing capacity of the involved tissues system, the model provides a description of the subsequent damage evolution during unloading. The model is used to study parametrically the scenario when reperfusion damage dominates over healing upon unloading and the opposite scenario when the loading and subsequent reperfusion damages remain small relative to the healing capacity of the tissues system. The theoretical model provides an integrated understanding of how tissue damage may further build-up paradoxically even with unloading, how long it would take for the loading and reperfusion damages in the tissues to become fully recovered, and how such loading and reperfusion damages, if not given sufficient time for recovery, may accumulate over multiple loading and unloading cycles, leading to clinical deep tissues ulceration. Copyright © 2011 Elsevier Ltd. All rights reserved.

Three-dimensional fuse deposition modeling of tissue-simulating phantom for biomedical optical imaging

NASA Astrophysics Data System (ADS)

Dong, Erbao; Zhao, Zuhua; Wang, Minjie; Xie, Yanjun; Li, Shidi; Shao, Pengfei; Cheng, Liuquan; Xu, Ronald X.

2015-12-01

Biomedical optical devices are widely used for clinical detection of various tissue anomalies. However, optical measurements have limited accuracy and traceability, partially owing to the lack of effective calibration methods that simulate the actual tissue conditions. To facilitate standardized calibration and performance evaluation of medical optical devices, we develop a three-dimensional fuse deposition modeling (FDM) technique for freeform fabrication of tissue-simulating phantoms. The FDM system uses transparent gel wax as the base material, titanium dioxide (TiO2) powder as the scattering ingredient, and graphite powder as the absorption ingredient. The ingredients are preheated, mixed, and deposited at the designated ratios layer-by-layer to simulate tissue structural and optical heterogeneities. By printing the sections of human brain model based on magnetic resonance images, we demonstrate the capability for simulating tissue structural heterogeneities. By measuring optical properties of multilayered phantoms and comparing with numerical simulation, we demonstrate the feasibility for simulating tissue optical properties. By creating a rat head phantom with embedded vasculature, we demonstrate the potential for mimicking physiologic processes of a living system.

Potential for Imaging Engineered Tissues with X-Ray Phase Contrast

PubMed Central

Appel, Alyssa; Anastasio, Mark A.

2011-01-01

As the field of tissue engineering advances, it is crucial to develop imaging methods capable of providing detailed three-dimensional information on tissue structure. X-ray imaging techniques based on phase-contrast (PC) have great potential for a number of biomedical applications due to their ability to provide information about soft tissue structure without exogenous contrast agents. X-ray PC techniques retain the excellent spatial resolution, tissue penetration, and calcified tissue contrast of conventional X-ray techniques while providing drastically improved imaging of soft tissue and biomaterials. This suggests that X-ray PC techniques are very promising for evaluation of engineered tissues. In this review, four different implementations of X-ray PC imaging are described and applications to tissues of relevance to tissue engineering reviewed. In addition, recent applications of X-ray PC to the evaluation of biomaterial scaffolds and engineered tissues are presented and areas for further development and application of these techniques are discussed. Imaging techniques based on X-ray PC have significant potential for improving our ability to image and characterize engineered tissues, and their continued development and optimization could have significant impact on the field of tissue engineering. PMID:21682604

Evaluation of multimodality imaging using image fusion with ultrasound tissue elasticity imaging in an experimental animal model.

PubMed

Paprottka, P M; Zengel, P; Cyran, C C; Ingrisch, M; Nikolaou, K; Reiser, M F; Clevert, D A

2014-01-01

To evaluate the ultrasound tissue elasticity imaging by comparison to multimodality imaging using image fusion with Magnetic Resonance Imaging (MRI) and conventional grey scale imaging with additional elasticity-ultrasound in an experimental small-animal-squamous-cell carcinoma-model for the assessment of tissue morphology. Human hypopharynx carcinoma cells were subcutaneously injected into the left flank of 12 female athymic nude rats. After 10 days (SD ± 2) of subcutaneous tumor growth, sonographic grey scale including elasticity imaging and MRI measurements were performed using a high-end ultrasound system and a 3T MR. For image fusion the contrast-enhanced MRI DICOM data set was uploaded in the ultrasonic device which has a magnetic field generator, a linear array transducer (6-15 MHz) and a dedicated software package (GE Logic E9), that can detect transducers by means of a positioning system. Conventional grey scale and elasticity imaging were integrated in the image fusion examination. After successful registration and image fusion the registered MR-images were simultaneously shown with the respective ultrasound sectional plane. Data evaluation was performed using the digitally stored video sequence data sets by two experienced radiologist using a modified Tsukuba Elasticity score. The colors "red and green" are assigned for an area of soft tissue, "blue" indicates hard tissue. In all cases a successful image fusion and plan registration with MRI and ultrasound imaging including grey scale and elasticity imaging was possible. The mean tumor volume based on caliper measurements in 3 dimensions was ~323 mm3. 4/12 rats were evaluated with Score I, 5/12 rates were evaluated with Score II, 3/12 rates were evaluated with Score III. There was a close correlation in the fused MRI with existing small necrosis in the tumor. None of the scored II or III lesions was visible by conventional grey scale. The comparison of ultrasound tissue elasticity imaging enables a

Deep Learning for Image-Based Cassava Disease Detection.

PubMed

Ramcharan, Amanda; Baranowski, Kelsee; McCloskey, Peter; Ahmed, Babuali; Legg, James; Hughes, David P

2017-01-01

Cassava is the third largest source of carbohydrates for human food in the world but is vulnerable to virus diseases, which threaten to destabilize food security in sub-Saharan Africa. Novel methods of cassava disease detection are needed to support improved control which will prevent this crisis. Image recognition offers both a cost effective and scalable technology for disease detection. New deep learning models offer an avenue for this technology to be easily deployed on mobile devices. Using a dataset of cassava disease images taken in the field in Tanzania, we applied transfer learning to train a deep convolutional neural network to identify three diseases and two types of pest damage (or lack thereof). The best trained model accuracies were 98% for brown leaf spot (BLS), 96% for red mite damage (RMD), 95% for green mite damage (GMD), 98% for cassava brown streak disease (CBSD), and 96% for cassava mosaic disease (CMD). The best model achieved an overall accuracy of 93% for data not used in the training process. Our results show that the transfer learning approach for image recognition of field images offers a fast, affordable, and easily deployable strategy for digital plant disease detection.

Sensitive red protein calcium indicators for imaging neural activity

PubMed Central

Dana, Hod; Mohar, Boaz; Sun, Yi; Narayan, Sujatha; Gordus, Andrew; Hasseman, Jeremy P; Tsegaye, Getahun; Holt, Graham T; Hu, Amy; Walpita, Deepika; Patel, Ronak; Macklin, John J; Bargmann, Cornelia I; Ahrens, Misha B; Schreiter, Eric R; Jayaraman, Vivek; Looger, Loren L; Svoboda, Karel; Kim, Douglas S

2016-01-01

Genetically encoded calcium indicators (GECIs) allow measurement of activity in large populations of neurons and in small neuronal compartments, over times of milliseconds to months. Although GFP-based GECIs are widely used for in vivo neurophysiology, GECIs with red-shifted excitation and emission spectra have advantages for in vivo imaging because of reduced scattering and absorption in tissue, and a consequent reduction in phototoxicity. However, current red GECIs are inferior to the state-of-the-art GFP-based GCaMP6 indicators for detecting and quantifying neural activity. Here we present improved red GECIs based on mRuby (jRCaMP1a, b) and mApple (jRGECO1a), with sensitivity comparable to GCaMP6. We characterized the performance of the new red GECIs in cultured neurons and in mouse, Drosophila, zebrafish and C. elegans in vivo. Red GECIs facilitate deep-tissue imaging, dual-color imaging together with GFP-based reporters, and the use of optogenetics in combination with calcium imaging. DOI: http://dx.doi.org/10.7554/eLife.12727.001 PMID:27011354

Deep Learning in Gastrointestinal Endoscopy.

PubMed

Patel, Vivek; Armstrong, David; Ganguli, Malika; Roopra, Sandeep; Kantipudi, Neha; Albashir, Siwar; Kamath, Markad V

2016-01-01

Gastrointestinal (GI) endoscopy is used to inspect the lumen or interior of the GI tract for several purposes, including, (1) making a clinical diagnosis, in real time, based on the visual appearances; (2) taking targeted tissue samples for subsequent histopathological examination; and (3) in some cases, performing therapeutic interventions targeted at specific lesions. GI endoscopy is therefore predicated on the assumption that the operator-the endoscopist-is able to identify and characterize abnormalities or lesions accurately and reproducibly. However, as in other areas of clinical medicine, such as histopathology and radiology, many studies have documented marked interobserver and intraobserver variability in lesion recognition. Thus, there is a clear need and opportunity for techniques or methodologies that will enhance the quality of lesion recognition and diagnosis and improve the outcomes of GI endoscopy. Deep learning models provide a basis to make better clinical decisions in medical image analysis. Biomedical image segmentation, classification, and registration can be improved with deep learning. Recent evidence suggests that the application of deep learning methods to medical image analysis can contribute significantly to computer-aided diagnosis. Deep learning models are usually considered to be more flexible and provide reliable solutions for image analysis problems compared to conventional computer vision models. The use of fast computers offers the possibility of real-time support that is important for endoscopic diagnosis, which has to be made in real time. Advanced graphics processing units and cloud computing have also favored the use of machine learning, and more particularly, deep learning for patient care. This paper reviews the rapidly evolving literature on the feasibility of applying deep learning algorithms to endoscopic imaging.

Deep learning for medical image segmentation - using the IBM TrueNorth neurosynaptic system

NASA Astrophysics Data System (ADS)

Moran, Steven; Gaonkar, Bilwaj; Whitehead, William; Wolk, Aidan; Macyszyn, Luke; Iyer, Subramanian S.

2018-03-01

Deep convolutional neural networks have found success in semantic image segmentation tasks in computer vision and medical imaging. These algorithms are executed on conventional von Neumann processor architectures or GPUs. This is suboptimal. Neuromorphic processors that replicate the structure of the brain are better-suited to train and execute deep learning models for image segmentation by relying on massively-parallel processing. However, given that they closely emulate the human brain, on-chip hardware and digital memory limitations also constrain them. Adapting deep learning models to execute image segmentation tasks on such chips, requires specialized training and validation. In this work, we demonstrate for the first-time, spinal image segmentation performed using a deep learning network implemented on neuromorphic hardware of the IBM TrueNorth Neurosynaptic System and validate the performance of our network by comparing it to human-generated segmentations of spinal vertebrae and disks. To achieve this on neuromorphic hardware, the training model constrains the coefficients of individual neurons to {-1,0,1} using the Energy Efficient Deep Neuromorphic (EEDN)1 networks training algorithm. Given the 1 million neurons and 256 million synapses, the scale and size of the neural network implemented by the IBM TrueNorth allows us to execute the requisite mapping between segmented images and non-uniform intensity MR images >20 times faster than on a GPU-accelerated network and using <0.1 W. This speed and efficiency implies that a trained neuromorphic chip can be deployed in intra-operative environments where real-time medical image segmentation is necessary.

Imaging challenges in biomaterials and tissue engineering

PubMed Central

Appel, Alyssa A.; Anastasio, Mark A.; Larson, Jeffery C.; Brey, Eric M.

2013-01-01

Biomaterials are employed in the fields of tissue engineering and regenerative medicine (TERM) in order to enhance the regeneration or replacement of tissue function and/or structure. The unique environments resulting from the presence of biomaterials, cells, and tissues result in distinct challenges in regards to monitoring and assessing the results of these interventions. Imaging technologies for three-dimensional (3D) analysis have been identified as a strategic priority in TERM research. Traditionally, histological and immunohistochemical techniques have been used to evaluate engineered tissues. However, these methods do not allow for an accurate volume assessment, are invasive, and do not provide information on functional status. Imaging techniques are needed that enable non-destructive, longitudinal, quantitative, and three-dimensional analysis of TERM strategies. This review focuses on evaluating the application of available imaging modalities for assessment of biomaterials and tissue in TERM applications. Included is a discussion of limitations of these techniques and identification of areas for further development. PMID:23768903

Particle velocity measurements with macroscopic fluorescence imaging in lymph tissue mimicking microfluidic phantoms

NASA Astrophysics Data System (ADS)

Hennessy, Ricky; Koo, Chiwan; Ton, Phuc; Han, Arum; Righetti, Raffaella; Maitland, Kristen C.

2011-03-01

Ultrasound poroelastography can quantify structural and mechanical properties of tissues such as stiffness, compressibility, and fluid flow rate. This novel ultrasound technique is being explored to detect tissue changes associated with lymphatic disease. We have constructed a macroscopic fluorescence imaging system to validate ultrasonic fluid flow measurements and to provide high resolution imaging of microfluidic phantoms. The optical imaging system is composed of a white light source, excitation and emission filters, and a camera with a zoom lens. The field of view can be adjusted from 100 mm x 75 mm to 10 mm x 7.5 mm. The microfluidic device is made of polydimethylsiloxane (PDMS) and has 9 channels, each 40 μm deep with widths ranging from 30 μm to 200 μm. A syringe pump was used to propel water containing 15 μm diameter fluorescent microspheres through the microchannels, with flow rates ranging from 0.5 μl/min to 10 μl/min. Video was captured at a rate of 25 frames/sec. The velocity of the microspheres in the microchannels was calculated using an algorithm that tracked the movement of the fluorescent microspheres. The imaging system was able to measure particle velocities ranging from 0.2 mm/sec to 10 mm/sec. The range of flow velocities of interest in lymph vessels is between 1 mm/sec to 10 mm/sec; therefore our imaging system is sufficient to measure particle velocity in phantoms modeling lymphatic flow.

Deep image mining for diabetic retinopathy screening.

PubMed

Quellec, Gwenolé; Charrière, Katia; Boudi, Yassine; Cochener, Béatrice; Lamard, Mathieu

2017-07-01

Deep learning is quickly becoming the leading methodology for medical image analysis. Given a large medical archive, where each image is associated with a diagnosis, efficient pathology detectors or classifiers can be trained with virtually no expert knowledge about the target pathologies. However, deep learning algorithms, including the popular ConvNets, are black boxes: little is known about the local patterns analyzed by ConvNets to make a decision at the image level. A solution is proposed in this paper to create heatmaps showing which pixels in images play a role in the image-level predictions. In other words, a ConvNet trained for image-level classification can be used to detect lesions as well. A generalization of the backpropagation method is proposed in order to train ConvNets that produce high-quality heatmaps. The proposed solution is applied to diabetic retinopathy (DR) screening in a dataset of almost 90,000 fundus photographs from the 2015 Kaggle Diabetic Retinopathy competition and a private dataset of almost 110,000 photographs (e-ophtha). For the task of detecting referable DR, very good detection performance was achieved: A z =0.954 in Kaggle's dataset and A z =0.949 in e-ophtha. Performance was also evaluated at the image level and at the lesion level in the DiaretDB1 dataset, where four types of lesions are manually segmented: microaneurysms, hemorrhages, exudates and cotton-wool spots. For the task of detecting images containing these four lesion types, the proposed detector, which was trained to detect referable DR, outperforms recent algorithms trained to detect those lesions specifically, with pixel-level supervision. At the lesion level, the proposed detector outperforms heatmap generation algorithms for ConvNets. This detector is part of the Messidor® system for mobile eye pathology screening. Because it does not rely on expert knowledge or manual segmentation for detecting relevant patterns, the proposed solution is a promising image

Scalable High Performance Image Registration Framework by Unsupervised Deep Feature Representations Learning

PubMed Central

Wu, Guorong; Kim, Minjeong; Wang, Qian; Munsell, Brent C.

2015-01-01

Feature selection is a critical step in deformable image registration. In particular, selecting the most discriminative features that accurately and concisely describe complex morphological patterns in image patches improves correspondence detection, which in turn improves image registration accuracy. Furthermore, since more and more imaging modalities are being invented to better identify morphological changes in medical imaging data,, the development of deformable image registration method that scales well to new image modalities or new image applications with little to no human intervention would have a significant impact on the medical image analysis community. To address these concerns, a learning-based image registration framework is proposed that uses deep learning to discover compact and highly discriminative features upon observed imaging data. Specifically, the proposed feature selection method uses a convolutional stacked auto-encoder to identify intrinsic deep feature representations in image patches. Since deep learning is an unsupervised learning method, no ground truth label knowledge is required. This makes the proposed feature selection method more flexible to new imaging modalities since feature representations can be directly learned from the observed imaging data in a very short amount of time. Using the LONI and ADNI imaging datasets, image registration performance was compared to two existing state-of-the-art deformable image registration methods that use handcrafted features. To demonstrate the scalability of the proposed image registration framework image registration experiments were conducted on 7.0-tesla brain MR images. In all experiments, the results showed the new image registration framework consistently demonstrated more accurate registration results when compared to state-of-the-art. PMID:26552069

Scalable High-Performance Image Registration Framework by Unsupervised Deep Feature Representations Learning.

PubMed

Wu, Guorong; Kim, Minjeong; Wang, Qian; Munsell, Brent C; Shen, Dinggang

2016-07-01

Feature selection is a critical step in deformable image registration. In particular, selecting the most discriminative features that accurately and concisely describe complex morphological patterns in image patches improves correspondence detection, which in turn improves image registration accuracy. Furthermore, since more and more imaging modalities are being invented to better identify morphological changes in medical imaging data, the development of deformable image registration method that scales well to new image modalities or new image applications with little to no human intervention would have a significant impact on the medical image analysis community. To address these concerns, a learning-based image registration framework is proposed that uses deep learning to discover compact and highly discriminative features upon observed imaging data. Specifically, the proposed feature selection method uses a convolutional stacked autoencoder to identify intrinsic deep feature representations in image patches. Since deep learning is an unsupervised learning method, no ground truth label knowledge is required. This makes the proposed feature selection method more flexible to new imaging modalities since feature representations can be directly learned from the observed imaging data in a very short amount of time. Using the LONI and ADNI imaging datasets, image registration performance was compared to two existing state-of-the-art deformable image registration methods that use handcrafted features. To demonstrate the scalability of the proposed image registration framework, image registration experiments were conducted on 7.0-T brain MR images. In all experiments, the results showed that the new image registration framework consistently demonstrated more accurate registration results when compared to state of the art.

Cascaded deep decision networks for classification of endoscopic images

NASA Astrophysics Data System (ADS)

Murthy, Venkatesh N.; Singh, Vivek; Sun, Shanhui; Bhattacharya, Subhabrata; Chen, Terrence; Comaniciu, Dorin

2017-02-01

Both traditional and wireless capsule endoscopes can generate tens of thousands of images for each patient. It is desirable to have the majority of irrelevant images filtered out by automatic algorithms during an offline review process or to have automatic indication for highly suspicious areas during an online guidance. This also applies to the newly invented endomicroscopy, where online indication of tumor classification plays a significant role. Image classification is a standard pattern recognition problem and is well studied in the literature. However, performance on the challenging endoscopic images still has room for improvement. In this paper, we present a novel Cascaded Deep Decision Network (CDDN) to improve image classification performance over standard Deep neural network based methods. During the learning phase, CDDN automatically builds a network which discards samples that are classified with high confidence scores by a previously trained network and concentrates only on the challenging samples which would be handled by the subsequent expert shallow networks. We validate CDDN using two different types of endoscopic imaging, which includes a polyp classification dataset and a tumor classification dataset. From both datasets we show that CDDN can outperform other methods by about 10%. In addition, CDDN can also be applied to other image classification problems.

Applying Deep Learning in Medical Images: The Case of Bone Age Estimation.

PubMed

Lee, Jang Hyung; Kim, Kwang Gi

2018-01-01

A diagnostic need often arises to estimate bone age from X-ray images of the hand of a subject during the growth period. Together with measured physical height, such information may be used as indicators for the height growth prognosis of the subject. We present a way to apply the deep learning technique to medical image analysis using hand bone age estimation as an example. Age estimation was formulated as a regression problem with hand X-ray images as input and estimated age as output. A set of hand X-ray images was used to form a training set with which a regression model was trained. An image preprocessing procedure is described which reduces image variations across data instances that are unrelated to age-wise variation. The use of Caffe, a deep learning tool is demonstrated. A rather simple deep learning network was adopted and trained for tutorial purpose. A test set distinct from the training set was formed to assess the validity of the approach. The measured mean absolute difference value was 18.9 months, and the concordance correlation coefficient was 0.78. It is shown that the proposed deep learning-based neural network can be used to estimate a subject's age from hand X-ray images, which eliminates the need for tedious atlas look-ups in clinical environments and should improve the time and cost efficiency of the estimation process.

Multiplexed aberration measurement for deep tissue imaging in vivo

PubMed Central

Wang, Chen; Liu, Rui; Milkie, Daniel E.; Sun, Wenzhi; Tan, Zhongchao; Kerlin, Aaron; Chen, Tsai-Wen; Kim, Douglas S.; Ji, Na

2014-01-01

We describe a multiplexed aberration measurement method that modulates the intensity or phase of light rays at multiple pupil segments in parallel to determine their phase gradients. Applicable to fluorescent-protein-labeled structures of arbitrary complexity, it allows us to obtain diffraction-limited resolution in various samples in vivo. For the strongly scattering mouse brain, a single aberration correction improves structural and functional imaging of fine neuronal processes over a large imaging volume. PMID:25128976

Polymer dots enable deep in vivo multiphoton fluorescence imaging of cerebrovascular architecture

NASA Astrophysics Data System (ADS)

Hassan, Ahmed M.; Wu, Xu; Jarrett, Jeremy W.; Xu, Shihan; Miller, David R.; Yu, Jiangbo; Perillo, Evan P.; Liu, Yen-Liang; Chiu, Daniel T.; Yeh, Hsin-Chih; Dunn, Andrew K.

2018-02-01

Deep in vivo imaging of vasculature requires small, bright, and photostable fluorophores suitable for multiphoton microscopy (MPM). Although semiconducting polymer dots (pdots) are an emerging class of highly fluorescent contrast agents with favorable advantages for the next generation of in vivo imaging, their use for deep multiphoton imaging has never before been demonstrated. Here we characterize the multiphoton properties of three pdot variants (CNPPV, PFBT, and PFPV) and demonstrate deep imaging of cortical microvasculature in C57 mice. Specifically, we measure the two- versus three-photon power dependence of these pdots and observe a clear three-photon excitation signature at wavelengths longer than 1300 nm, and a transition from two-photon to three-photon excitation within a 1060 - 1300 nm excitation range. Furthermore, we show that pdots enable in vivo two-photon imaging of cerebrovascular architecture in mice up to 850 μm beneath the pial surface using 800 nm excitation. In contrast with traditional multiphoton probes, we also demonstrate that the broad multiphoton absorption spectrum of pdots permits imaging at longer wavelengths (λex = 1,060 and 1225 nm). These wavelengths approach an ideal biological imaging wavelength near 1,300 nm and confer compatibility with a high-power ytterbium-fiber laser and a high pulse energy optical parametric amplifier, resulting in substantial improvements in signal-to-background ratio (>3.5-fold) and greater cortical imaging depths of 900 μm and 1300 μm. Ultimately, pdots are a versatile tool for MPM due to their extraordinary brightness and broad absorption, which will undoubtedly unlock the ability to interrogate deep structures in vivo.

Phase contrast imaging of buccal mucosa tissues-Feasibility study

NASA Astrophysics Data System (ADS)

Fatima, A.; Tripathi, S.; Shripathi, T.; Kulkarni, V. K.; Banda, N. R.; Agrawal, A. K.; Sarkar, P. S.; Kashyap, Y.; Sinha, A.

2015-06-01

Phase Contrast Imaging (PCI) technique has been used to interpret physical parameters obtained from the image taken on the normal buccal mucosa tissue extracted from cheek of a patient. The advantages of this method over the conventional imaging techniques are discussed. PCI technique uses the X-ray phase shift at the edges differentiated by very minute density differences and the edge enhanced high contrast images reveal details of soft tissues. The contrast in the images produced is related to changes in the X-ray refractive index of the tissues resulting in higher clarity compared with conventional absorption based X-ray imaging. The results show that this type of imaging has better ability to visualize microstructures of biological soft tissues with good contrast, which can lead to the diagnosis of lesions at an early stage of the diseases.

Objective breast tissue image classification using Quantitative Transmission ultrasound tomography

NASA Astrophysics Data System (ADS)

Malik, Bilal; Klock, John; Wiskin, James; Lenox, Mark

2016-12-01

Quantitative Transmission Ultrasound (QT) is a powerful and emerging imaging paradigm which has the potential to perform true three-dimensional image reconstruction of biological tissue. Breast imaging is an important application of QT and allows non-invasive, non-ionizing imaging of whole breasts in vivo. Here, we report the first demonstration of breast tissue image classification in QT imaging. We systematically assess the ability of the QT images’ features to differentiate between normal breast tissue types. The three QT features were used in Support Vector Machines (SVM) classifiers, and classification of breast tissue as either skin, fat, glands, ducts or connective tissue was demonstrated with an overall accuracy of greater than 90%. Finally, the classifier was validated on whole breast image volumes to provide a color-coded breast tissue volume. This study serves as a first step towards a computer-aided detection/diagnosis platform for QT.

Research on image retrieval using deep convolutional neural network combining L1 regularization and PRelu activation function

NASA Astrophysics Data System (ADS)

QingJie, Wei; WenBin, Wang

2017-06-01

In this paper, the image retrieval using deep convolutional neural network combined with regularization and PRelu activation function is studied, and improves image retrieval accuracy. Deep convolutional neural network can not only simulate the process of human brain to receive and transmit information, but also contains a convolution operation, which is very suitable for processing images. Using deep convolutional neural network is better than direct extraction of image visual features for image retrieval. However, the structure of deep convolutional neural network is complex, and it is easy to over-fitting and reduces the accuracy of image retrieval. In this paper, we combine L1 regularization and PRelu activation function to construct a deep convolutional neural network to prevent over-fitting of the network and improve the accuracy of image retrieval

Research on simulated infrared image utility evaluation using deep representation

NASA Astrophysics Data System (ADS)

Zhang, Ruiheng; Mu, Chengpo; Yang, Yu; Xu, Lixin

2018-01-01

Infrared (IR) image simulation is an important data source for various target recognition systems. However, whether simulated IR images could be used as training data for classifiers depends on the features of fidelity and authenticity of simulated IR images. For evaluation of IR image features, a deep-representation-based algorithm is proposed. Being different from conventional methods, which usually adopt a priori knowledge or manually designed feature, the proposed method can extract essential features and quantitatively evaluate the utility of simulated IR images. First, for data preparation, we employ our IR image simulation system to generate large amounts of IR images. Then, we present the evaluation model of simulated IR image, for which an end-to-end IR feature extraction and target detection model based on deep convolutional neural network is designed. At last, the experiments illustrate that our proposed method outperforms other verification algorithms in evaluating simulated IR images. Cross-validation, variable proportion mixed data validation, and simulation process contrast experiments are carried out to evaluate the utility and objectivity of the images generated by our simulation system. The optimum mixing ratio between simulated and real data is 0.2≤γ≤0.3, which is an effective data augmentation method for real IR images.

Two-dimensional and 3-D images of thick tissue using time-constrained times-of-flight and absorbance spectrophotometry

NASA Astrophysics Data System (ADS)

Benaron, David A.; Lennox, M.; Stevenson, David K.

1992-05-01

Reconstructing deep-tissue images in real time using spectrophotometric data from optically diffusing thick tissues has been problematic. Continuous wave applications (e.g., pulse oximetry, regional cerebral saturation) ignore both the multiple paths traveled by the photons through the tissue and the effects of scattering, allowing scalar measurements but only under limited conditions; interferometry works poorly in thick, highly-scattering media; frequency- modulated approaches may not allow full deconvolution of scattering and absorbance; and pulsed-light techniques allow for preservation of information regarding the multiple paths taken by light through the tissue, but reconstruction is both computation intensive and limited by the relative surface area available for detection of photons. We have developed a picosecond times-of-flight and absorbance (TOFA) optical system, time-constrained to measure only photons with a narrow range of path lengths and arriving within a narrow angel of the emitter-detector axis. The delay until arrival of the earliest arriving photons is a function of both the scattering and absorbance of the tissues in a direct line between the emitter and detector, reducing the influence of surrounding tissues. Measurement using a variety of emitter and detector locations produces spatial information which can be analyzed in a standard 2-D grid, or subject to computer reconstruction to produce tomographic images representing 3-D structure. Using such a technique, we have been able to demonstrate the principles of tc-TOFA, detect and localize diffusive and/or absorptive objects suspended in highly scattering media (such as blood admixed with yeast), and perform simple 3-D reconstructions using phantom objects. We are now attempting to obtain images in vivo. Potential future applications include use as a research tool, and as a continuous, noninvasive, nondestructive monitor in diagnostic imaging, fetal monitoring, neurologic and cardiac

Improving face image extraction by using deep learning technique

NASA Astrophysics Data System (ADS)

Xue, Zhiyun; Antani, Sameer; Long, L. R.; Demner-Fushman, Dina; Thoma, George R.

2016-03-01

The National Library of Medicine (NLM) has made a collection of over a 1.2 million research articles containing 3.2 million figure images searchable using the Open-iSM multimodal (text+image) search engine. Many images are visible light photographs, some of which are images containing faces ("face images"). Some of these face images are acquired in unconstrained settings, while others are studio photos. To extract the face regions in the images, we first applied one of the most widely-used face detectors, a pre-trained Viola-Jones detector implemented in Matlab and OpenCV. The Viola-Jones detector was trained for unconstrained face image detection, but the results for the NLM database included many false positives, which resulted in a very low precision. To improve this performance, we applied a deep learning technique, which reduced the number of false positives and as a result, the detection precision was improved significantly. (For example, the classification accuracy for identifying whether the face regions output by this Viola- Jones detector are true positives or not in a test set is about 96%.) By combining these two techniques (Viola-Jones and deep learning) we were able to increase the system precision considerably, while avoiding the need to manually construct a large training set by manual delineation of the face regions.

Propionibacterium acnes in shoulder surgery: true infection, contamination, or commensal of the deep tissue?

PubMed

Hudek, Robert; Sommer, Frank; Kerwat, Martina; Abdelkawi, Ayman F; Loos, Franziska; Gohlke, Frank

2014-12-01

Propionibacterium acnes has been linked to chronic infections in shoulder surgery. Whether the bacterium is a contaminant or commensal of the deep tissue is unclear. We aimed to assess P. acnes in intraoperative samples of different tissue layers in patients undergoing first-time shoulder surgery. In 118 consecutive patients (mean age, 59.2 years; 75 men, 43 women), intraoperative samples were correlated to preoperative subacromial injection, the type of surgical approach, and gender. One skin, one superficial, one deep tissue, and one test sample were cultured for each patient. The cultures were positive for P. acnes in 36.4% (n = 43) of cases. Subacromial injection was not associated with bacterial growth rates (P = .88 for P. acnes; P = .20 for bacteria other than P. acnes; P = .85 for the anterolateral approach; P = .92 for the deltopectoral approach; P = .56 for men; P = .51 for women). Skin samples were positive for P. acnes in 8.5% (n = 10), superficial samples were positive in 7.6% (n = 9), deep samples were positive in 13.6% (n = 16), and both samples (superficial and deep) were positive in 15.3% (n = 18) of cases (P < .0001). P. acnes was detected in the anterolateral approach in 27.1% (n = 32) of cases and in the deltopectoral approach in 9.3% (n = 11) of cases (P = .01; relative risk, 1.93; 95% confidence interval, 1.08-3.43). Thirty-five of the P. acnes-positive patients were men (81.4%), and 8 patients were women (18.6%; P = .001; relative risk, 2.51; 95% confidence interval, 1.28-4.90). P. acnes was detected in more than one third of patients undergoing first-time shoulder surgery. Preoperative subacromial injection was not associated with bacterial growth. P. acnes was observed more frequently in the deep tissues than in the superficial tissues. The relative risk for obtaining a positive P. acnes culture was 2-fold greater for the anterolateral approach than for the deltopectoral approach, and the risk was 2.5-fold greater

Formulation design facilitates magnetic nanoparticle delivery to diseased cells and tissues

PubMed Central

Singh, Dhirender; McMillan, JoEllyn M; Liu, Xin-Ming; Vishwasrao, Hemant M; Kabanov, Alexander V; Sokolsky-Papkov, Marina; Gendelman, Howard E

2015-01-01

Magnetic nanoparticles (MNPs) accumulate at disease sites with the aid of magnetic fields; biodegradable MNPs can be designed to facilitate drug delivery, influence disease diagnostics, facilitate tissue regeneration and permit protein purification. Because of their limited toxicity, MNPs are widely used in theranostics, simultaneously facilitating diagnostics and therapeutics. To realize therapeutic end points, iron oxide nanoparticle cores (5–30 nm) are encapsulated in a biocompatible polymer shell with drug cargos. Although limited, the toxic potential of MNPs parallels magnetite composition, along with shape, size and surface chemistry. Clearance is hastened by the reticuloendothelial system. To surmount translational barriers, the crystal structure, particle surface and magnetic properties of MNPs need to be optimized. With this in mind, we provide a comprehensive evaluation of advancements in MNP synthesis, functionalization and design, with an eye towards bench-to-bedside translation. PMID:24646020

Pre-operative CT angiography and three-dimensional image post processing for deep inferior epigastric perforator flap breast reconstructive surgery.

PubMed

Lam, D L; Mitsumori, L M; Neligan, P C; Warren, B H; Shuman, W P; Dubinsky, T J

2012-12-01

Autologous breast reconstructive surgery with deep inferior epigastric artery (DIEA) perforator flaps has become the mainstay for breast reconstructive surgery. CT angiography and three-dimensional image post processing can depict the number, size, course and location of the DIEA perforating arteries for the pre-operative selection of the best artery to use for the tissue flap. Knowledge of the location and selection of the optimal perforating artery shortens operative times and decreases patient morbidity.

Mesoporous composite nanoparticles for dual-modality ultrasound/magnetic resonance imaging and synergistic chemo-/thermotherapy against deep tumors.

PubMed

Zhang, Nan; Wang, Ronghui; Hao, Junnian; Yang, Yang; Zou, Hongmi; Wang, Zhigang

2017-01-01

High-intensity focused ultrasound (HIFU) is a promising and noninvasive treatment for solid tumors, which has been explored for potential clinical applications. However, the clinical applications of HIFU for large and deep tumors such as hepatocellular carcinoma (HCC) are severely limited by unsatisfactory imaging guidance, long therapeutic times, and damage to normal tissue around the tumor due to the high power applied. In this study, we developed doxorubicin/perfluorohexane-encapsulated hollow mesoporous Prussian blue nanoparticles (HMPBs-DOX/PFH) as theranostic agents, which can effectively guide HIFU therapy and enhance its therapeutic effects in combination with chemotherapy, by decreasing the cavitation threshold. We investigated the effects of this agent on ultrasound and magnetic resonance imaging in vitro and in vivo. In addition, we showed a highly efficient HIFU therapeutic effect against HCC tumors, as well as controlled drug release, owing to the phase-transitional performance of the PFH. We therefore conclude that HMPB-DOX/PFH is a safe and efficient nanoplatform, which holds significant promise for cancer theranostics against deep tumors in clinical settings.

Mesoporous composite nanoparticles for dual-modality ultrasound/magnetic resonance imaging and synergistic chemo-/thermotherapy against deep tumors

PubMed Central

Zhang, Nan; Wang, Ronghui; Hao, Junnian; Yang, Yang; Zou, Hongmi; Wang, Zhigang

2017-01-01

High-intensity focused ultrasound (HIFU) is a promising and noninvasive treatment for solid tumors, which has been explored for potential clinical applications. However, the clinical applications of HIFU for large and deep tumors such as hepatocellular carcinoma (HCC) are severely limited by unsatisfactory imaging guidance, long therapeutic times, and damage to normal tissue around the tumor due to the high power applied. In this study, we developed doxorubicin/perfluorohexane-encapsulated hollow mesoporous Prussian blue nanoparticles (HMPBs-DOX/PFH) as theranostic agents, which can effectively guide HIFU therapy and enhance its therapeutic effects in combination with chemotherapy, by decreasing the cavitation threshold. We investigated the effects of this agent on ultrasound and magnetic resonance imaging in vitro and in vivo. In addition, we showed a highly efficient HIFU therapeutic effect against HCC tumors, as well as controlled drug release, owing to the phase-transitional performance of the PFH. We therefore conclude that HMPB-DOX/PFH is a safe and efficient nanoplatform, which holds significant promise for cancer theranostics against deep tumors in clinical settings. PMID:29042775

"Deep-media culture condition" promoted lumen formation of endothelial cells within engineered three-dimensional tissues in vitro.

PubMed

Sekiya, Sachiko; Shimizu, Tatsuya; Yamato, Masayuki; Okano, Teruo

2011-03-01

In the field of tissue engineering, the induction of microvessels into tissues is an important task because of the need to overcome diffusion limitations of oxygen and nutrients within tissues. Powerful methods to create vessels in engineered tissues are needed for creating real living tissues. In this study, we utilized three-dimensional (3D) highly cell dense tissues fabricated by cell sheet technology. The 3D tissue constructs are close to living-cell dense tissue in vivo. Additionally, creating an endothelial cell (EC) network within tissues promoted neovascularization promptly within the tissue after transplantation in vivo. Compared to the conditions in vivo, however, common in vitro cell culture conditions provide a poor environment for creating lumens within 3D tissue constructs. Therefore, for determining adequate conditions for vascularizing engineered tissue in vitro, our 3D tissue constructs were cultured under a "deep-media culture conditions." Compared to the control conditions, the morphology of ECs showed a visibly strained cytoskeleton, and the density of lumen formation within tissues increased under hydrostatic pressure conditions. Moreover, the increasing expression of vascular endothelial cadherin in the lumens suggested that the vessels were stabilized in the stimulated tissues compared with the control. These findings suggested that deep-media culture conditions improved lumen formation in engineered tissues in vitro.

Tissue feature-based intra-fractional motion tracking for stereoscopic x-ray image guided radiotherapy

NASA Astrophysics Data System (ADS)

Xie, Yaoqin; Xing, Lei; Gu, Jia; Liu, Wu

2013-06-01

Real-time knowledge of tumor position during radiation therapy is essential to overcome the adverse effect of intra-fractional organ motion. The goal of this work is to develop a tumor tracking strategy by effectively utilizing the inherent image features of stereoscopic x-ray images acquired during dose delivery. In stereoscopic x-ray image guided radiation delivery, two orthogonal x-ray images are acquired either simultaneously or sequentially. The essence of markerless tumor tracking is the reliable identification of inherent points with distinct tissue features on each projection image and their association between two images. The identification of the feature points on a planar x-ray image is realized by searching for points with high intensity gradient. The feature points are associated by using the scale invariance features transform descriptor. The performance of the proposed technique is evaluated by using images of a motion phantom and four archived clinical cases acquired using either a CyberKnife equipped with a stereoscopic x-ray imaging system, or a LINAC equipped with an onboard kV imager and an electronic portal imaging device. In the phantom study, the results obtained using the proposed method agree with the measurements to within 2 mm in all three directions. In the clinical study, the mean error is 0.48 ± 0.46 mm for four patient data with 144 sequential images. In this work, a tissue feature-based tracking method for stereoscopic x-ray image guided radiation therapy is developed. The technique avoids the invasive procedure of fiducial implantation and may greatly facilitate the clinical workflow.

Tissue microarrays and quantitative tissue-based image analysis as a tool for oncology biomarker and diagnostic development.

PubMed

Dolled-Filhart, Marisa P; Gustavson, Mark D

2012-11-01

Translational oncology has been improved by using tissue microarrays (TMAs), which facilitate biomarker analysis of large cohorts on a single slide. This has allowed for rapid analysis and validation of potential biomarkers for prognostic and predictive value, as well as for evaluation of biomarker prevalence. Coupled with quantitative analysis of immunohistochemical (IHC) staining, objective and standardized biomarker data from tumor samples can further advance companion diagnostic approaches for the identification of drug-responsive or resistant patient subpopulations. This review covers the advantages, disadvantages and applications of TMAs for biomarker research. Research literature and reviews of TMAs and quantitative image analysis methodology have been surveyed for this review (with an AQUA® analysis focus). Applications such as multi-marker diagnostic development and pathway-based biomarker subpopulation analyses are described. Tissue microarrays are a useful tool for biomarker analyses including prevalence surveys, disease progression assessment and addressing potential prognostic or predictive value. By combining quantitative image analysis with TMAs, analyses will be more objective and reproducible, allowing for more robust IHC-based diagnostic test development. Quantitative multi-biomarker IHC diagnostic tests that can predict drug response will allow for greater success of clinical trials for targeted therapies and provide more personalized clinical decision making.

Multimodal tissue perfusion imaging using multi-spectral and thermographic imaging systems applied on clinical data

NASA Astrophysics Data System (ADS)

Klaessens, John H. G. M.; Nelisse, Martin; Verdaasdonk, Rudolf M.; Noordmans, Herke Jan

2013-03-01

Clinical interventions can cause changes in tissue perfusion, oxygenation or temperature. Real-time imaging of these phenomena could be useful for surgical strategy or understanding of physiological regulation mechanisms. Two noncontact imaging techniques were applied for imaging of large tissue areas: LED based multispectral imaging (MSI, 17 different wavelengths 370 nm-880 nm) and thermal imaging (7.5 to 13.5 μm). Oxygenation concentration changes were calculated using different analyzing methods. The advantages of these methods are presented for stationary and dynamic applications. Concentration calculations of chromophores in tissue require right choices of wavelengths The effects of different wavelength choices for hemoglobin concentration calculations were studied in laboratory conditions and consequently applied in clinical studies. Corrections for interferences during the clinical registrations (ambient light fluctuations, tissue movements) were performed. The wavelength dependency of the algorithms were studied and wavelength sets with the best results will be presented. The multispectral and thermal imaging systems were applied during clinical intervention studies: reperfusion of tissue flap transplantation (ENT), effectiveness of local anesthetic block and during open brain surgery in patients with epileptic seizures. The LED multispectral imaging system successfully imaged the perfusion and oxygenation changes during clinical interventions. The thermal images show local heat distributions over tissue areas as a result of changes in tissue perfusion. Multispectral imaging and thermal imaging provide complementary information and are promising techniques for real-time diagnostics of physiological processes in medicine.

Classify epithelium-stroma in histopathological images based on deep transferable network.

PubMed

Yu, X; Zheng, H; Liu, C; Huang, Y; Ding, X

2018-04-20

Recently, the deep learning methods have received more attention in histopathological image analysis. However, the traditional deep learning methods assume that training data and test data have the same distributions, which causes certain limitations in real-world histopathological applications. However, it is costly to recollect a large amount of labeled histology data to train a new neural network for each specified image acquisition procedure even for similar tasks. In this paper, an unsupervised domain adaptation is introduced into a typical deep convolutional neural network (CNN) model to mitigate the repeating of the labels. The unsupervised domain adaptation is implemented by adding two regularisation terms, namely the feature-based adaptation and entropy minimisation, to the object function of a widely used CNN model called the AlexNet. Three independent public epithelium-stroma datasets were used to verify the proposed method. The experimental results have demonstrated that in the epithelium-stroma classification, the proposed method can achieve better performance than the commonly used deep learning methods and some existing deep domain adaptation methods. Therefore, the proposed method can be considered as a better option for the real-world applications of histopathological image analysis because there is no requirement for recollection of large-scale labeled data for every specified domain. © 2018 The Authors Journal of Microscopy © 2018 Royal Microscopical Society.

Two-Stage Approach to Image Classification by Deep Neural Networks

NASA Astrophysics Data System (ADS)

Ososkov, Gennady; Goncharov, Pavel

2018-02-01

The paper demonstrates the advantages of the deep learning networks over the ordinary neural networks on their comparative applications to image classifying. An autoassociative neural network is used as a standalone autoencoder for prior extraction of the most informative features of the input data for neural networks to be compared further as classifiers. The main efforts to deal with deep learning networks are spent for a quite painstaking work of optimizing the structures of those networks and their components, as activation functions, weights, as well as the procedures of minimizing their loss function to improve their performances and speed up their learning time. It is also shown that the deep autoencoders develop the remarkable ability for denoising images after being specially trained. Convolutional Neural Networks are also used to solve a quite actual problem of protein genetics on the example of the durum wheat classification. Results of our comparative study demonstrate the undoubted advantage of the deep networks, as well as the denoising power of the autoencoders. In our work we use both GPU and cloud services to speed up the calculations.

Integrated Raman spectroscopy and trimodal wide-field imaging techniques for real-time in vivo tissue Raman measurements at endoscopy.

PubMed

Huang, Zhiwei; Teh, Seng Khoon; Zheng, Wei; Mo, Jianhua; Lin, Kan; Shao, Xiaozhuo; Ho, Khek Yu; Teh, Ming; Yeoh, Khay Guan

2009-03-15

We report an integrated Raman spectroscopy and trimodal (white-light reflectance, autofluorescence, and narrow-band) imaging techniques for real-time in vivo tissue Raman measurements at endoscopy. A special 1.8 mm endoscopic Raman probe with filtering modules is developed, permitting effective elimination of interference of fluorescence background and silica Raman in fibers while maximizing tissue Raman collections. We demonstrate that high-quality in vivo Raman spectra of upper gastrointestinal tract can be acquired within 1 s or subseconds under the guidance of wide-field endoscopic imaging modalities, greatly facilitating the adoption of Raman spectroscopy into clinical research and practice during routine endoscopic inspections.

Probing neural tissue with airy light-sheet microscopy: investigation of imaging performance at depth within turbid media

NASA Astrophysics Data System (ADS)

Nylk, Jonathan; McCluskey, Kaley; Aggarwal, Sanya; Tello, Javier A.; Dholakia, Kishan

2017-02-01

Light-sheet microscopy (LSM) has received great interest for fluorescent imaging applications in biomedicine as it facilitates three-dimensional visualisation of large sample volumes with high spatiotemporal resolution whilst minimising irradiation of, and photo-damage to the specimen. Despite these advantages, LSM can only visualize superficial layers of turbid tissues, such as mammalian neural tissue. Propagation-invariant light modes have played a key role in the development of high-resolution LSM techniques as they overcome the natural divergence of a Gaussian beam, enabling uniform and thin light-sheets over large distances. Most notably, Bessel and Airy beam-based light-sheet imaging modalities have been demonstrated. In the single-photon excitation regime and in lightly scattering specimens, Airy-LSM has given competitive performance with advanced Bessel-LSM techniques. Airy and Bessel beams share the property of self-healing, the ability of the beam to regenerate its transverse beam profile after propagation around an obstacle. Bessel-LSM techniques have been shown to increase the penetration-depth of the illumination into turbid specimens but this effect has been understudied in biologically relevant tissues, particularly for Airy beams. It is expected that Airy-LSM will give a similar enhancement over Gaussian-LSM. In this paper, we report on the comparison of Airy-LSM and Gaussian-LSM imaging modalities within cleared and non-cleared mouse brain tissue. In particular, we examine image quality versus tissue depth by quantitative spatial Fourier analysis of neural structures in virally transduced fluorescent tissue sections, showing a three-fold enhancement at 50 μm depth into non-cleared tissue with Airy-LSM. Complimentary analysis is performed by resolution measurements in bead-injected tissue sections.

Assessing microscope image focus quality with deep learning.

PubMed

Yang, Samuel J; Berndl, Marc; Michael Ando, D; Barch, Mariya; Narayanaswamy, Arunachalam; Christiansen, Eric; Hoyer, Stephan; Roat, Chris; Hung, Jane; Rueden, Curtis T; Shankar, Asim; Finkbeiner, Steven; Nelson, Philip

2018-03-15

Large image datasets acquired on automated microscopes typically have some fraction of low quality, out-of-focus images, despite the use of hardware autofocus systems. Identification of these images using automated image analysis with high accuracy is important for obtaining a clean, unbiased image dataset. Complicating this task is the fact that image focus quality is only well-defined in foreground regions of images, and as a result, most previous approaches only enable a computation of the relative difference in quality between two or more images, rather than an absolute measure of quality. We present a deep neural network model capable of predicting an absolute measure of image focus on a single image in isolation, without any user-specified parameters. The model operates at the image-patch level, and also outputs a measure of prediction certainty, enabling interpretable predictions. The model was trained on only 384 in-focus Hoechst (nuclei) stain images of U2OS cells, which were synthetically defocused to one of 11 absolute defocus levels during training. The trained model can generalize on previously unseen real Hoechst stain images, identifying the absolute image focus to within one defocus level (approximately 3 pixel blur diameter difference) with 95% accuracy. On a simpler binary in/out-of-focus classification task, the trained model outperforms previous approaches on both Hoechst and Phalloidin (actin) stain images (F-scores of 0.89 and 0.86, respectively over 0.84 and 0.83), despite only having been presented Hoechst stain images during training. Lastly, we observe qualitatively that the model generalizes to two additional stains, Hoechst and Tubulin, of an unseen cell type (Human MCF-7) acquired on a different instrument. Our deep neural network enables classification of out-of-focus microscope images with both higher accuracy and greater precision than previous approaches via interpretable patch-level focus and certainty predictions. The use of

An analysis of image storage systems for scalable training of deep neural networks

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

Lim, Seung-Hwan; Young, Steven R; Patton, Robert M

This study presents a principled empirical evaluation of image storage systems for training deep neural networks. We employ the Caffe deep learning framework to train neural network models for three different data sets, MNIST, CIFAR-10, and ImageNet. While training the models, we evaluate five different options to retrieve training image data: (1) PNG-formatted image files on local file system; (2) pushing pixel arrays from image files into a single HDF5 file on local file system; (3) in-memory arrays to hold the pixel arrays in Python and C++; (4) loading the training data into LevelDB, a log-structured merge tree based key-valuemore » storage; and (5) loading the training data into LMDB, a B+tree based key-value storage. The experimental results quantitatively highlight the disadvantage of using normal image files on local file systems to train deep neural networks and demonstrate reliable performance with key-value storage based storage systems. When training a model on the ImageNet dataset, the image file option was more than 17 times slower than the key-value storage option. Along with measurements on training time, this study provides in-depth analysis on the cause of performance advantages/disadvantages of each back-end to train deep neural networks. We envision the provided measurements and analysis will shed light on the optimal way to architect systems for training neural networks in a scalable manner.« less

A novel end-to-end classifier using domain transferred deep convolutional neural networks for biomedical images.

PubMed

Pang, Shuchao; Yu, Zhezhou; Orgun, Mehmet A

2017-03-01

Highly accurate classification of biomedical images is an essential task in the clinical diagnosis of numerous medical diseases identified from those images. Traditional image classification methods combined with hand-crafted image feature descriptors and various classifiers are not able to effectively improve the accuracy rate and meet the high requirements of classification of biomedical images. The same also holds true for artificial neural network models directly trained with limited biomedical images used as training data or directly used as a black box to extract the deep features based on another distant dataset. In this study, we propose a highly reliable and accurate end-to-end classifier for all kinds of biomedical images via deep learning and transfer learning. We first apply domain transferred deep convolutional neural network for building a deep model; and then develop an overall deep learning architecture based on the raw pixels of original biomedical images using supervised training. In our model, we do not need the manual design of the feature space, seek an effective feature vector classifier or segment specific detection object and image patches, which are the main technological difficulties in the adoption of traditional image classification methods. Moreover, we do not need to be concerned with whether there are large training sets of annotated biomedical images, affordable parallel computing resources featuring GPUs or long times to wait for training a perfect deep model, which are the main problems to train deep neural networks for biomedical image classification as observed in recent works. With the utilization of a simple data augmentation method and fast convergence speed, our algorithm can achieve the best accuracy rate and outstanding classification ability for biomedical images. We have evaluated our classifier on several well-known public biomedical datasets and compared it with several state-of-the-art approaches. We propose a robust

Image-based metrology of porous tissue engineering scaffolds

NASA Astrophysics Data System (ADS)

Rajagopalan, Srinivasan; Robb, Richard A.

2006-03-01

Tissue engineering is an interdisciplinary effort aimed at the repair and regeneration of biological tissues through the application and control of cells, porous scaffolds and growth factors. The regeneration of specific tissues guided by tissue analogous substrates is dependent on diverse scaffold architectural indices that can be derived quantitatively from the microCT and microMR images of the scaffolds. However, the randomness of pore-solid distributions in conventional stochastic scaffolds presents unique computational challenges. As a result, image-based characterization of scaffolds has been predominantly qualitative. In this paper, we discuss quantitative image-based techniques that can be used to compute the metrological indices of porous tissue engineering scaffolds. While bulk averaged quantities such as porosity and surface are derived directly from the optimal pore-solid delineations, the spatially distributed geometric indices are derived from the medial axis representations of the pore network. The computational framework proposed (to the best of our knowledge for the first time in tissue engineering) in this paper might have profound implications towards unraveling the symbiotic structure-function relationship of porous tissue engineering scaffolds.

Stokes polarimetry imaging of dog prostate tissue

NASA Astrophysics Data System (ADS)

Kim, Jihoon; Johnston, William K., III; Walsh, Joseph T., Jr.

2010-02-01

Prostate cancer is the second leading cause of death in the United States in 2009. Radical prostatectomy (complete removal of the prostate) is the most common treatment for prostate cancer, however, differentiating prostate tissue from adjacent bladder, nerves, and muscle is difficult. Improved visualization could improve oncologic outcomes and decrease damage to adjacent nerves and muscle important for preservation of potency and continence. A novel Stokes polarimetry imaging (SPI) system was developed and evaluated using a dog prostate specimen in order to examine the feasibility of the system to differentiate prostate from bladder. The degree of linear polarization (DOLP) image maps from linearly polarized light illumination at different visible wavelengths (475, 510, and 650 nm) were constructed. The SPI system used the polarization property of the prostate tissue. The DOLP images allowed advanced differentiation by distinguishing glandular tissue of prostate from the muscular-stromal tissue in the bladder. The DOLP image at 650 nm effectively differentiated prostate and bladder by strong DOLP in bladder. SPI system has the potential to improve surgical outcomes in open or robotic-assisted laparoscopic removal of the prostate. Further in vivo testing is warranted.

Sensitivity and accuracy of hybrid fluorescence-mediated tomography in deep tissue regions.

PubMed

Rosenhain, Stefanie; Al Rawashdeh, Wa'el; Kiessling, Fabian; Gremse, Felix

2017-09-01

Fluorescence-mediated tomography (FMT) enables noninvasive assessment of the three-dimensional distribution of near-infrared fluorescence in mice. The combination with micro-computed tomography (µCT) provides anatomical data, enabling improved fluorescence reconstruction and image analysis. The aim of our study was to assess sensitivity and accuracy of µCT-FMT under realistic in vivo conditions in deeply-seated regions. Accordingly, we acquired fluorescence reflectance images (FRI) and µCT-FMT scans of mice which were prepared with rectal insertions with different amounts of fluorescent dye. Default and high-sensitivity scans were acquired and background signal was analyzed for three FMT channels (670 nm, 745 nm, and 790 nm). Analysis was performed for the original and an improved FMT reconstruction using the µCT data. While FRI and the original FMT reconstruction could detect 100 pmol, the improved FMT reconstruction could detect 10 pmol and significantly improved signal localization. By using a finer sampling grid and increasing the exposure time, the sensitivity could be further improved to detect 0.5 pmol. Background signal was highest in the 670 nm channel and most prominent in the gastro-intestinal tract and in organs with high relative amounts of blood. In conclusion, we show that µCT-FMT allows sensitive and accurate assessment of fluorescence in deep tissue regions. © 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

Referred pain and cutaneous responses from deep tissue electrical pain stimulation in the groin.

PubMed

Aasvang, E K; Werner, M U; Kehlet, H

2015-08-01

Persistent postherniotomy pain is located around the scar and external inguinal ring and is often described as deep rather than cutaneous, with frequent complaints of pain in adjacent areas. Whether this pain is due to local pathology or referred/projected pain is unknown, hindering mechanism-based treatment. Deep tissue electrical pain stimulation by needle electrodes in the right groin (rectus muscle, ilioinguinal/iliohypogastric nerve and perispermatic cord) was combined with assessment of referred/projected pain and the cutaneous heat pain threshold (HPT) at three prespecified areas (both groins and the lower right arm) in 19 healthy subjects. The assessment was repeated 10 days later to assess the reproducibility of individual responses. Deep electrical stimulation elicited pain at the stimulation site in all subjects, and in 15 subjects, pain from areas outside the stimulation area was reported, with 90-100% having the same response on both days, depending on the location. Deep pain stimulation significantly increased the cutaneous HPT (P<0.014). Individual HPT responses before and during deep electrical pain stimulation were significantly correlated (ρ>0.474, P≤0.040) at the two test days for the majority of test areas. Our results corroborate a systematic relationship between deep pain and changes in cutaneous nociception. The individual referred/projected pain patterns and cutaneous responses are variable, but reproducible, supporting individual differences in anatomy and sensory processing. Future studies investigating the responses to deep tissue electrical stimulation in persistent postherniotomy pain patients may advance our understanding of underlying pathophysiological mechanisms and strategies for treatment and prevention. ClinicalTrials.gov (NCT01701427). © The Author 2015. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Deep Multimodal Distance Metric Learning Using Click Constraints for Image Ranking.

PubMed

Yu, Jun; Yang, Xiaokang; Gao, Fei; Tao, Dacheng

2017-12-01

How do we retrieve images accurately? Also, how do we rank a group of images precisely and efficiently for specific queries? These problems are critical for researchers and engineers to generate a novel image searching engine. First, it is important to obtain an appropriate description that effectively represent the images. In this paper, multimodal features are considered for describing images. The images unique properties are reflected by visual features, which are correlated to each other. However, semantic gaps always exist between images visual features and semantics. Therefore, we utilize click feature to reduce the semantic gap. The second key issue is learning an appropriate distance metric to combine these multimodal features. This paper develops a novel deep multimodal distance metric learning (Deep-MDML) method. A structured ranking model is adopted to utilize both visual and click features in distance metric learning (DML). Specifically, images and their related ranking results are first collected to form the training set. Multimodal features, including click and visual features, are collected with these images. Next, a group of autoencoders is applied to obtain initially a distance metric in different visual spaces, and an MDML method is used to assign optimal weights for different modalities. Next, we conduct alternating optimization to train the ranking model, which is used for the ranking of new queries with click features. Compared with existing image ranking methods, the proposed method adopts a new ranking model to use multimodal features, including click features and visual features in DML. We operated experiments to analyze the proposed Deep-MDML in two benchmark data sets, and the results validate the effects of the method.

Classification of microscopic images of breast tissue

NASA Astrophysics Data System (ADS)

Ballerini, Lucia; Franzen, Lennart

2004-05-01

Breast cancer is the most common form of cancer among women. The diagnosis is usually performed by the pathologist, that subjectively evaluates tissue samples. The aim of our research is to develop techniques for the automatic classification of cancerous tissue, by analyzing histological samples of intact tissue taken with a biopsy. In our study, we considered 200 images presenting four different conditions: normal tissue, fibroadenosis, ductal cancer and lobular cancer. Methods to extract features have been investigated and described. One method is based on granulometries, which are size-shape descriptors widely used in mathematical morphology. Applications of granulometries lead to distribution functions whose moments are used as features. A second method is based on fractal geometry, that seems very suitable to quantify biological structures. The fractal dimension of binary images has been computed using the euclidean distance mapping. Image classification has then been performed using the extracted features as input of a back-propagation neural network. A new method that combines genetic algorithms and morphological filters has been also investigated. In this case, the classification is based on a correlation measure. Very encouraging results have been obtained with pilot experiments using a small subset of images as training set. Experimental results indicate the effectiveness of the proposed methods. Cancerous tissue was correctly classified in 92.5% of the cases.

Contrast-enhanced photoacoustic imaging with an optical wavelength of 1064 nm

NASA Astrophysics Data System (ADS)

Kim, Jeesu; Park, Sara; Park, Gyeong Bae; Choi, Wonseok; Jeong, Unyong; Kim, Chulhong

2018-02-01

Photoacoustic (PA) imaging is a biomedical imaging method that can provide both structural and functional information of living tissues beyond the optical diffusion limit by combining the concepts of conventional optical and ultrasound imaging methods. Although endogenous chromophores can be utilized to acquire PA images of biological tissues, exogenous contrast agents that absorb near-infrared (NIR) lights have been extensively explored to improve the contrast and penetration depth of PA images. Here, we demonstrate Bi2Se3 nanoplates, that strongly absorbs NIR lights, as a contrast agent for PA imaging. In particularly, the Bi2Se3 nanoplates produce relatively strong PA signals with an optical wavelength of 1064 nm, which has several advantages for deep tissue imaging including: (1) relatively low absorption by other intrinsic chromophores, (2) cost-effective light source using Nd:YAG laser, and (3) higher available energy than other NIR lights according to American National Standards Institute (ANSI) safety limit. We have investigated deep tissue imaging capability of the Bi2Se3 nanoplates by acquiring in vitro PA images of microtubes under chicken breast tissues. We have also acquired in vivo PA images of bladders, gastrointestinal tracts, and sentinel lymph nodes in mice after injection of the Bi2Se3 nanoplates to verify their applicability to a variety of biomedical research. The results show the promising potential of the Bi2Se3 nanoplates as a PA contrast agent for deep tissue imaging with an optical wavelength of 1064 nm.

Adipose Tissue Quantification by Imaging Methods: A Proposed Classification

PubMed Central

Shen, Wei; Wang, ZiMian; Punyanita, Mark; Lei, Jianbo; Sinav, Ahmet; Kral, John G.; Imielinska, Celina; Ross, Robert; Heymsfield, Steven B.

2007-01-01

Recent advances in imaging techniques and understanding of differences in the molecular biology of adipose tissue has rendered classical anatomy obsolete, requiring a new classification of the topography of adipose tissue. Adipose tissue is one of the largest body compartments, yet a classification that defines specific adipose tissue depots based on their anatomic location and related functions is lacking. The absence of an accepted taxonomy poses problems for investigators studying adipose tissue topography and its functional correlates. The aim of this review was to critically examine the literature on imaging of whole body and regional adipose tissue and to create the first systematic classification of adipose tissue topography. Adipose tissue terminology was examined in over 100 original publications. Our analysis revealed inconsistencies in the use of specific definitions, especially for the compartment termed “visceral” adipose tissue. This analysis leads us to propose an updated classification of total body and regional adipose tissue, providing a well-defined basis for correlating imaging studies of specific adipose tissue depots with molecular processes. PMID:12529479

Lesion Detection in CT Images Using Deep Learning Semantic Segmentation Technique

NASA Astrophysics Data System (ADS)

Kalinovsky, A.; Liauchuk, V.; Tarasau, A.

2017-05-01

In this paper, the problem of automatic detection of tuberculosis lesion on 3D lung CT images is considered as a benchmark for testing out algorithms based on a modern concept of Deep Learning. For training and testing of the algorithms a domestic dataset of 338 3D CT scans of tuberculosis patients with manually labelled lesions was used. The algorithms which are based on using Deep Convolutional Networks were implemented and applied in three different ways including slice-wise lesion detection in 2D images using semantic segmentation, slice-wise lesion detection in 2D images using sliding window technique as well as straightforward detection of lesions via semantic segmentation in whole 3D CT scans. The algorithms demonstrate superior performance compared to algorithms based on conventional image analysis methods.

[Advantages and Application Prospects of Deep Learning in Image Recognition and Bone Age Assessment].

PubMed

Hu, T H; Wan, L; Liu, T A; Wang, M W; Chen, T; Wang, Y H

2017-12-01

Deep learning and neural network models have been new research directions and hot issues in the fields of machine learning and artificial intelligence in recent years. Deep learning has made a breakthrough in the applications of image and speech recognitions, and also has been extensively used in the fields of face recognition and information retrieval because of its special superiority. Bone X-ray images express different variations in black-white-gray gradations, which have image features of black and white contrasts and level differences. Based on these advantages of deep learning in image recognition, we combine it with the research of bone age assessment to provide basic datum for constructing a forensic automatic system of bone age assessment. This paper reviews the basic concept and network architectures of deep learning, and describes its recent research progress on image recognition in different research fields at home and abroad, and explores its advantages and application prospects in bone age assessment. Copyright© by the Editorial Department of Journal of Forensic Medicine.

Classification of time-series images using deep convolutional neural networks

NASA Astrophysics Data System (ADS)

Hatami, Nima; Gavet, Yann; Debayle, Johan

2018-04-01

Convolutional Neural Networks (CNN) has achieved a great success in image recognition task by automatically learning a hierarchical feature representation from raw data. While the majority of Time-Series Classification (TSC) literature is focused on 1D signals, this paper uses Recurrence Plots (RP) to transform time-series into 2D texture images and then take advantage of the deep CNN classifier. Image representation of time-series introduces different feature types that are not available for 1D signals, and therefore TSC can be treated as texture image recognition task. CNN model also allows learning different levels of representations together with a classifier, jointly and automatically. Therefore, using RP and CNN in a unified framework is expected to boost the recognition rate of TSC. Experimental results on the UCR time-series classification archive demonstrate competitive accuracy of the proposed approach, compared not only to the existing deep architectures, but also to the state-of-the art TSC algorithms.

Pre-operative intra-articular deep tissue sampling with novel retrograde forceps improves the diagnostics in periprosthetic joint infection.

PubMed

Wimmer, Matthias D; Ploeger, Milena M; Friedrich, Max J; Hügle, Thomas; Gravius, Sascha; Randau, Thomas M

2017-07-01

Histopathological tissue analysis is a key parameter within the diagnostic algorithm for suspected periprosthetic joint infections (PJIs), conventionally acquired in open surgery. In 2014, Hügle and co-workers introduced novel retrograde forceps for retrograde synovial biopsy with simultaneous fluid aspiration of the knee joint. We hypothesised that tissue samples acquired by retrograde synovial biopsy are equal to intra-operatively acquired deep representative tissue samples regarding bacterial detection and differentiation of periprosthetic infectious membranes. Thirty patients (male n = 15, 50%; female n = 15, 50%) with 30 suspected PJIs in painful total hip arthroplasties (THAs) were included in this prospective, controlled, non-blinded trial. The results were compared with intra-operatively obtained representative deep tissue samples. In summary, 27 out of 30 patients were diagnosed correctly as infected (17/17) or non-infected (10/13). The sensitivity to predict a PJI using the Retroforce® sampling forceps in addition to standard diagnostics was 85%, the specificity 100%. Retrograde synovial biopsy is a new and rapid diagnostic procedure under local anaesthesia in patients with painful THAs with similar histological results compared to deep tissue sampling.

Focusing light through biological tissue and tissue-mimicking phantoms up to 9.6 cm in thickness with digital optical phase conjugation

NASA Astrophysics Data System (ADS)

Shen, Yuecheng; Liu, Yan; Ma, Cheng; Wang, Lihong V.

2016-08-01

Optical phase conjugation (OPC)-based wavefront shaping techniques focus light through or within scattering media, which is critically important for deep-tissue optical imaging, manipulation, and therapy. However, to date, the sample thickness in OPC experiments has been limited to only a few millimeters. Here, by using a laser with a long coherence length and an optimized digital OPC system that can safely deliver more light power, we focused 532-nm light through tissue-mimicking phantoms up to 9.6 cm thick, as well as through ex vivo chicken breast tissue up to 2.5 cm thick. Our results demonstrate that OPC can be achieved even when photons have experienced on average 1000 scattering events. The demonstrated penetration of nearly 10 cm (˜100 transport mean free paths) has never been achieved before by any optical focusing technique, and it shows the promise of OPC for deep-tissue noninvasive optical imaging, manipulation, and therapy.

Deep learning methods for CT image-domain metal artifact reduction

NASA Astrophysics Data System (ADS)

Gjesteby, Lars; Yang, Qingsong; Xi, Yan; Shan, Hongming; Claus, Bernhard; Jin, Yannan; De Man, Bruno; Wang, Ge

2017-09-01

Artifacts resulting from metal objects have been a persistent problem in CT images over the last four decades. A common approach to overcome their effects is to replace corrupt projection data with values synthesized from an interpolation scheme or by reprojection of a prior image. State-of-the-art correction methods, such as the interpolation- and normalization-based algorithm NMAR, often do not produce clinically satisfactory results. Residual image artifacts remain in challenging cases and even new artifacts can be introduced by the interpolation scheme. Metal artifacts continue to be a major impediment, particularly in radiation and proton therapy planning as well as orthopedic imaging. A new solution to the long-standing metal artifact reduction (MAR) problem is deep learning, which has been successfully applied to medical image processing and analysis tasks. In this study, we combine a convolutional neural network (CNN) with the state-of-the-art NMAR algorithm to reduce metal streaks in critical image regions. Training data was synthesized from CT simulation scans of a phantom derived from real patient images. The CNN is able to map metal-corrupted images to artifact-free monoenergetic images to achieve additional correction on top of NMAR for improved image quality. Our results indicate that deep learning is a novel tool to address CT reconstruction challenges, and may enable more accurate tumor volume estimation for radiation therapy planning.

Image-guided urologic surgery: intraoperative optical imaging and tissue interrogation (Conference Presentation)

NASA Astrophysics Data System (ADS)

Liao, Joseph C.

2017-02-01

Emerging optical imaging technologies can be integrated in the operating room environment during minimally invasive and open urologic surgery, including oncologic surgery of the bladder, prostate, and kidney. These technologies include macroscopic fluorescence imaging that provides contrast enhancement between normal and diseased tissue and microscopic imaging that provides tissue characterization. Optical imaging technologies that have reached the clinical arena in urologic surgery are reviewed, including photodynamic diagnosis, near infrared fluorescence imaging, optical coherence tomography, and confocal laser endomicroscopy. Molecular imaging represents an exciting future arena in conjugating cancer-specific contrast agents to fluorophores to improve the specificity of disease detection. Ongoing efforts are underway to translate optimal targeting agents and imaging modalities, with the goal to improve cancer-specific and functional outcomes.

Functional imaging of small tissue volumes with diffuse optical tomography

NASA Astrophysics Data System (ADS)

Klose, Alexander D.; Hielscher, Andreas H.

2006-03-01

Imaging of dynamic changes in blood parameters, functional brain imaging, and tumor imaging are the most advanced application areas of diffuse optical tomography (DOT). When dealing with the image reconstruction problem one is faced with the fact that near-infrared photons, unlike X-rays, are highly scattered when they traverse biological tissue. Image reconstruction schemes are required that model the light propagation inside biological tissue and predict measurements on the tissue surface. By iteratively changing the tissue-parameters until the predictions agree with the real measurements, a spatial distribution of optical properties inside the tissue is found. The optical properties can be related to the tissue oxygenation, inflammation, or to the fluorophore concentration of a biochemical marker. If the model of light propagation is inaccurate, the reconstruction process will lead to an inaccurate result as well. Here, we focus on difficulties that are encountered when DOT is employed for functional imaging of small tissue volumes, for example, in cancer studies involving small animals, or human finger joints for early diagnosis of rheumatoid arthritis. Most of the currently employed image reconstruction methods rely on the diffusion theory that is an approximation to the equation of radiative transfer. But, in the cases of small tissue volumes and tissues that contain low scattering regions diffusion theory has been shown to be of limited applicability Therefore, we employ a light propagation model that is based on the equation of radiative transfer, which promises to overcome the limitations.

Classifying magnetic resonance image modalities with convolutional neural networks

NASA Astrophysics Data System (ADS)

Remedios, Samuel; Pham, Dzung L.; Butman, John A.; Roy, Snehashis

2018-02-01

Magnetic Resonance (MR) imaging allows the acquisition of images with different contrast properties depending on the acquisition protocol and the magnetic properties of tissues. Many MR brain image processing techniques, such as tissue segmentation, require multiple MR contrasts as inputs, and each contrast is treated differently. Thus it is advantageous to automate the identification of image contrasts for various purposes, such as facilitating image processing pipelines, and managing and maintaining large databases via content-based image retrieval (CBIR). Most automated CBIR techniques focus on a two-step process: extracting features from data and classifying the image based on these features. We present a novel 3D deep convolutional neural network (CNN)- based method for MR image contrast classification. The proposed CNN automatically identifies the MR contrast of an input brain image volume. Specifically, we explored three classification problems: (1) identify T1-weighted (T1-w), T2-weighted (T2-w), and fluid-attenuated inversion recovery (FLAIR) contrasts, (2) identify pre vs postcontrast T1, (3) identify pre vs post-contrast FLAIR. A total of 3418 image volumes acquired from multiple sites and multiple scanners were used. To evaluate each task, the proposed model was trained on 2137 images and tested on the remaining 1281 images. Results showed that image volumes were correctly classified with 97.57% accuracy.

Introducing Convective Cloud Microphysics to a Deep Convection Parameterization Facilitating Aerosol Indirect Effects

NASA Astrophysics Data System (ADS)

Alapaty, K.; Zhang, G. J.; Song, X.; Kain, J. S.; Herwehe, J. A.

2012-12-01

Short lived pollutants such as aerosols play an important role in modulating not only the radiative balance but also cloud microphysical properties and precipitation rates. In the past, to understand the interactions of aerosols with clouds, several cloud-resolving modeling studies were conducted. These studies indicated that in the presence of anthropogenic aerosols, single-phase deep convection precipitation is reduced or suppressed. On the other hand, anthropogenic aerosol pollution led to enhanced precipitation for mixed-phase deep convective clouds. To date, there have not been many efforts to incorporate such aerosol indirect effects (AIE) in mesoscale models or global models that use parameterization schemes for deep convection. Thus, the objective of this work is to implement a diagnostic cloud microphysical scheme directly into a deep convection parameterization facilitating aerosol indirect effects in the WRF-CMAQ integrated modeling systems. Major research issues addressed in this study are: What is the sensitivity of a deep convection scheme to cloud microphysical processes represented by a bulk double-moment scheme? How close are the simulated cloud water paths as compared to observations? Does increased aerosol pollution lead to increased precipitation for mixed-phase clouds? These research questions are addressed by performing several WRF simulations using the Kain-Fritsch convection parameterization and a diagnostic cloud microphysical scheme. In the first set of simulations (control simulations) the WRF model is used to simulate two scenarios of deep convection over the continental U.S. during two summer periods at 36 km grid resolution. In the second set, these simulations are repeated after incorporating a diagnostic cloud microphysical scheme to study the impacts of inclusion of cloud microphysical processes. Finally, in the third set, aerosol concentrations simulated by the CMAQ modeling system are supplied to the embedded cloud microphysical

Combined spectroscopic imaging and chemometric approach for automatically partitioning tissue types in human prostate tissue biopsies

NASA Astrophysics Data System (ADS)

Haka, Abigail S.; Kidder, Linda H.; Lewis, E. Neil

2001-07-01

We have applied Fourier transform infrared (FTIR) spectroscopic imaging, coupling a mercury cadmium telluride (MCT) focal plane array detector (FPA) and a Michelson step scan interferometer, to the investigation of various states of malignant human prostate tissue. The MCT FPA used consists of 64x64 pixels, each 61 micrometers 2, and has a spectral range of 2-10.5 microns. Each imaging data set was collected at 16-1 resolution, resulting in 512 image planes and a total of 4096 interferograms. In this article we describe a method for separating different tissue types contained within FTIR spectroscopic imaging data sets of human prostate tissue biopsies. We present images, generated by the Fuzzy C-Means clustering algorithm, which demonstrate the successful partitioning of distinct tissue type domains. Additionally, analysis of differences in the centroid spectra corresponding to different tissue types provides an insight into their biochemical composition. Lastly, we demonstrate the ability to partition tissue type regions in a different data set using centroid spectra calculated from the original data set. This has implications for the use of the Fuzzy C-Means algorithm as an automated technique for the separation and examination of tissue domains in biopsy samples.

Global Magnetospheric Imaging from the Deep Space Gateway in Lunar Orbit

NASA Astrophysics Data System (ADS)

Chua, D. H.; Socker, D. G.; Englert, C. R.; Carter, M. T.; Plunkett, S. P.; Korendyke, C. M.; Meier, R. R.

2018-02-01

We propose to use the Deep Space Gateway as an observing platform for a magnetospheric imager that will capture the first direct global images of the interface between the incident solar wind and the Earth's magnetosphere.

Confocal Imaging of the Embryonic Heart: How Deep?

NASA Astrophysics Data System (ADS)

Miller, Christine E.; Thompson, Robert P.; Bigelow, Michael R.; Gittinger, George; Trusk, Thomas C.; Sedmera, David

2005-06-01

Confocal microscopy allows for optical sectioning of tissues, thus obviating the need for physical sectioning and subsequent registration to obtain a three-dimensional representation of tissue architecture. However, practicalities such as tissue opacity, light penetration, and detector sensitivity have usually limited the available depth of imaging to 200 [mu]m. With the emergence of newer, more powerful systems, we attempted to push these limits to those dictated by the working distance of the objective. We used whole-mount immunohistochemical staining followed by clearing with benzyl alcohol-benzyl benzoate (BABB) to visualize three-dimensional myocardial architecture. Confocal imaging of entire chick embryonic hearts up to a depth of 1.5 mm with voxel dimensions of 3 [mu]m was achieved with a 10× dry objective. For the purpose of screening for congenital heart defects, we used endocardial painting with fluorescently labeled poly-L-lysine and imaged BABB-cleared hearts with a 5× objective up to a depth of 2 mm. Two-photon imaging of whole-mount specimens stained with Hoechst nuclear dye produced clear images all the way through stage 29 hearts without significant signal attenuation. Thus, currently available systems allow confocal imaging of fixed samples to previously unattainable depths, the current limiting factors being objective working distance, antibody penetration, specimen autofluorescence, and incomplete clearing.

Large-Scale Image Analytics Using Deep Learning

NASA Astrophysics Data System (ADS)

Ganguly, S.; Nemani, R. R.; Basu, S.; Mukhopadhyay, S.; Michaelis, A.; Votava, P.

2014-12-01

High resolution land cover classification maps are needed to increase the accuracy of current Land ecosystem and climate model outputs. Limited studies are in place that demonstrates the state-of-the-art in deriving very high resolution (VHR) land cover products. In addition, most methods heavily rely on commercial softwares that are difficult to scale given the region of study (e.g. continents to globe). Complexities in present approaches relate to (a) scalability of the algorithm, (b) large image data processing (compute and memory intensive), (c) computational cost, (d) massively parallel architecture, and (e) machine learning automation. In addition, VHR satellite datasets are of the order of terabytes and features extracted from these datasets are of the order of petabytes. In our present study, we have acquired the National Agricultural Imaging Program (NAIP) dataset for the Continental United States at a spatial resolution of 1-m. This data comes as image tiles (a total of quarter million image scenes with ~60 million pixels) and has a total size of ~100 terabytes for a single acquisition. Features extracted from the entire dataset would amount to ~8-10 petabytes. In our proposed approach, we have implemented a novel semi-automated machine learning algorithm rooted on the principles of "deep learning" to delineate the percentage of tree cover. In order to perform image analytics in such a granular system, it is mandatory to devise an intelligent archiving and query system for image retrieval, file structuring, metadata processing and filtering of all available image scenes. Using the Open NASA Earth Exchange (NEX) initiative, which is a partnership with Amazon Web Services (AWS), we have developed an end-to-end architecture for designing the database and the deep belief network (following the distbelief computing model) to solve a grand challenge of scaling this process across quarter million NAIP tiles that cover the entire Continental United States. The

Deep Convolutional Neural Networks Enable Discrimination of Heterogeneous Digital Pathology Images.

PubMed

Khosravi, Pegah; Kazemi, Ehsan; Imielinski, Marcin; Elemento, Olivier; Hajirasouliha, Iman

2018-01-01

Pathological evaluation of tumor tissue is pivotal for diagnosis in cancer patients and automated image analysis approaches have great potential to increase precision of diagnosis and help reduce human error. In this study, we utilize several computational methods based on convolutional neural networks (CNN) and build a stand-alone pipeline to effectively classify different histopathology images across different types of cancer. In particular, we demonstrate the utility of our pipeline to discriminate between two subtypes of lung cancer, four biomarkers of bladder cancer, and five biomarkers of breast cancer. In addition, we apply our pipeline to discriminate among four immunohistochemistry (IHC) staining scores of bladder and breast cancers. Our classification pipeline includes a basic CNN architecture, Google's Inceptions with three training strategies, and an ensemble of two state-of-the-art algorithms, Inception and ResNet. Training strategies include training the last layer of Google's Inceptions, training the network from scratch, and fine-tunning the parameters for our data using two pre-trained version of Google's Inception architectures, Inception-V1 and Inception-V3. We demonstrate the power of deep learning approaches for identifying cancer subtypes, and the robustness of Google's Inceptions even in presence of extensive tumor heterogeneity. On average, our pipeline achieved accuracies of 100%, 92%, 95%, and 69% for discrimination of various cancer tissues, subtypes, biomarkers, and scores, respectively. Our pipeline and related documentation is freely available at https://github.com/ih-_lab/CNN_Smoothie. Copyright © 2017 The Authors. Published by Elsevier B.V. All rights reserved.

High-Resolution Photoacoustic Imaging of Ocular Tissues

PubMed Central

Silverman, Ronald H.; Kong, Fanting; Chen, Y.C.; Lloyd, Harriet O.; Kim, Hyung Ham; Cannata, Jonathan M.; Shung, K. Kirk; Coleman, D Jackson

2010-01-01

Optical coherence tomography (OCT) and ultrasound (US) are methods widely used for diagnostic imaging of the eye. These techniques detect discontinuities in optical refractive index and acoustic impedance respectively. Because these both relate to variations in tissue density or composition, OCT and US images share a qualitatively similar appearance. In photoacoustic imaging (PAI), short light pulses are directed at tissues, pressure is generated due to a rapid energy deposition in the tissue volume, and thermoelastic expansion results in generation of broadband US. PAI thus depicts optical absorption, which is independent of the tissue characteristics imaged by OCT or US. Our aim was to demonstrate the application of PAI in ocular tissues and to do so with lateral resolution comparable to OCT. We developed two PAI assemblies, both of which used single-element US transducers and lasers sharing a common focus. The first assembly had optical and 35-MHz US axes offset by a 30° angle. The second assembly consisted of a 20-MHz ring transducer with a coaxial optics. The laser emitted 5-ns pulses at either 532-nm or 1064-nm, with spot sizes at the focus of 35-μm for the angled probe and 20-μm for the coaxial probe. We compared lateral resolution by scanning 12.5-μm diameter wire targets with pulse/echo US and PAI at each wavelength. We then imaged the anterior segment in whole ex vivo pig eyes and the choroid and ciliary body region in sectioned eyes. PAI data obtained at 1064 nm in the near infrared had higher penetration but reduced signal amplitude compared to that obtained using the 532-nm green wavelength. Images were obtained of the iris, choroid and ciliary processes. The zonules and anterior cornea and lens surfaces were seen at 532 nm. Because the laser spot size was significantly smaller than the US beamwidth at the focus, PAI images had superior resolution than those obtained using conventional US. PMID:20420969

High-resolution photoacoustic imaging of ocular tissues.

PubMed

Silverman, Ronald H; Kong, Fanting; Chen, Y C; Lloyd, Harriet O; Kim, Hyung Ham; Cannata, Jonathan M; Shung, K Kirk; Coleman, D Jackson

2010-05-01

Optical coherence tomography (OCT) and ultrasound (US) are methods widely used for diagnostic imaging of the eye. These techniques detect discontinuities in optical refractive index and acoustic impedance, respectively. Because these both relate to variations in tissue density or composition, OCT and US images share a qualitatively similar appearance. In photoacoustic imaging (PAI), short light pulses are directed at tissues, pressure is generated due to a rapid energy deposition in the tissue volume and thermoelastic expansion results in generation of broadband US. PAI thus depicts optical absorption, which is independent of the tissue characteristics imaged by OCT or US. Our aim was to demonstrate the application of PAI in ocular tissues and to do so with lateral resolution comparable to OCT. We developed two PAI assemblies, both of which used single-element US transducers and lasers sharing a common focus. The first assembly had optical and 35-MHz US axes offset by a 30 degrees angle. The second assembly consisted of a 20-MHz ring transducer with a coaxial optics. The laser emitted 5-ns pulses at either 532 nm or 1064 nm, with spot sizes at the focus of 35 microm for the angled probe and 20 microm for the coaxial probe. We compared lateral resolution by scanning 12.5 microm diameter wire targets with pulse/echo US and PAI at each wavelength. We then imaged the anterior segment in whole ex vivo pig eyes and the choroid and ciliary body region in sectioned eyes. PAI data obtained at 1064 nm in the near infrared had higher penetration but reduced signal amplitude compared to that obtained using the 532 nm green wavelength. Images were obtained of the iris, choroid and ciliary processes. The zonules and anterior cornea and lens surfaces were seen at 532 nm. Because the laser spot size was significantly smaller than the US beamwidth at the focus, PAI images had superior resolution than those obtained using conventional US. Copyright 2010 World Federation for

Fine-grained leukocyte classification with deep residual learning for microscopic images.

PubMed

Qin, Feiwei; Gao, Nannan; Peng, Yong; Wu, Zizhao; Shen, Shuying; Grudtsin, Artur

2018-08-01

Leukocyte classification and cytometry have wide applications in medical domain, previous researches usually exploit machine learning techniques to classify leukocytes automatically. However, constrained by the past development of machine learning techniques, for example, extracting distinctive features from raw microscopic images are difficult, the widely used SVM classifier only has relative few parameters to tune, these methods cannot efficiently handle fine-grained classification cases when the white blood cells have up to 40 categories. Based on deep learning theory, a systematic study is conducted on finer leukocyte classification in this paper. A deep residual neural network based leukocyte classifier is constructed at first, which can imitate the domain expert's cell recognition process, and extract salient features robustly and automatically. Then the deep neural network classifier's topology is adjusted according to the prior knowledge of white blood cell test. After that the microscopic image dataset with almost one hundred thousand labeled leukocytes belonging to 40 categories is built, and combined training strategies are adopted to make the designed classifier has good generalization ability. The proposed deep residual neural network based classifier was tested on microscopic image dataset with 40 leukocyte categories. It achieves top-1 accuracy of 77.80%, top-5 accuracy of 98.75% during the training procedure. The average accuracy on the test set is nearly 76.84%. This paper presents a fine-grained leukocyte classification method for microscopic images, based on deep residual learning theory and medical domain knowledge. Experimental results validate the feasibility and effectiveness of our approach. Extended experiments support that the fine-grained leukocyte classifier could be used in real medical applications, assist doctors in diagnosing diseases, reduce human power significantly. Copyright © 2018 Elsevier B.V. All rights reserved.

Mapping tissue oxygen in vivo by photoacoustic lifetime imaging

NASA Astrophysics Data System (ADS)

Shao, Qi; Morgounova, Ekaterina; Choi, Jeung-Hwan; Jiang, Chunlan; Bischof, John; Ashkenazi, Shai

2013-03-01

Oxygen plays a key role in the energy metabolism of living organisms. Any imbalance in the oxygen levels will affect the metabolic homeostasis and lead to pathophysiological diseases. Hypoxia, a status of low tissue oxygen, is a key factor in tumor biology as it is highly prominent in tumor tissues. However, clinical tools for assessing tissue oxygenation are limited. The gold standard is polarographic needle electrode which is invasive and not capable of mapping (imaging) the oxygen content in tissue. We applied the method of photoacoustic lifetime imaging (PALI) of oxygen-sensitive dye to small animal tissue hypoxia research. PALI is new technology for direct, non-invasive imaging of oxygen. The technique is based on mapping the oxygen-dependent transient optical absorption of Methylene Blue (MB) by pump-probe photoacoustic imaging. Our studies show the feasibility of imaging of dissolved oxygen distribution in phantoms. In vivo experiments demonstrate that the hypoxia region is consistent with the site of subcutaneously xenografted prostate tumor in mice with adequate spatial resolution and penetration depth.

Interactive classification and content-based retrieval of tissue images

NASA Astrophysics Data System (ADS)

Aksoy, Selim; Marchisio, Giovanni B.; Tusk, Carsten; Koperski, Krzysztof

2002-11-01

We describe a system for interactive classification and retrieval of microscopic tissue images. Our system models tissues in pixel, region and image levels. Pixel level features are generated using unsupervised clustering of color and texture values. Region level features include shape information and statistics of pixel level feature values. Image level features include statistics and spatial relationships of regions. To reduce the gap between low-level features and high-level expert knowledge, we define the concept of prototype regions. The system learns the prototype regions in an image collection using model-based clustering and density estimation. Different tissue types are modeled using spatial relationships of these regions. Spatial relationships are represented by fuzzy membership functions. The system automatically selects significant relationships from training data and builds models which can also be updated using user relevance feedback. A Bayesian framework is used to classify tissues based on these models. Preliminary experiments show that the spatial relationship models we developed provide a flexible and powerful framework for classification and retrieval of tissue images.

3D reconstruction of synapses with deep learning based on EM Images

NASA Astrophysics Data System (ADS)

Xiao, Chi; Rao, Qiang; Zhang, Dandan; Chen, Xi; Han, Hua; Xie, Qiwei

2017-03-01

Recently, due to the rapid development of electron microscope (EM) with its high resolution, stacks delivered by EM can be used to analyze a variety of components that are critical to understand brain function. Since synaptic study is essential in neurobiology and can be analyzed by EM stacks, the automated routines for reconstruction of synapses based on EM Images can become a very useful tool for analyzing large volumes of brain tissue and providing the ability to understand the mechanism of brain. In this article, we propose a novel automated method to realize 3D reconstruction of synapses for Automated Tapecollecting Ultra Microtome Scanning Electron Microscopy (ATUM-SEM) with deep learning. Being different from other reconstruction algorithms, which employ classifier to segment synaptic clefts directly. We utilize deep learning method and segmentation algorithm to obtain synaptic clefts as well as promote the accuracy of reconstruction. The proposed method contains five parts: (1) using modified Moving Least Square (MLS) deformation algorithm and Scale Invariant Feature Transform (SIFT) features to register adjacent sections, (2) adopting Faster Region Convolutional Neural Networks (Faster R-CNN) algorithm to detect synapses, (3) utilizing screening method which takes context cues of synapses into consideration to reduce the false positive rate, (4) combining a practical morphology algorithm with a suitable fitting function to segment synaptic clefts and optimize the shape of them, (5) applying the plugin in FIJI to show the final 3D visualization of synapses. Experimental results on ATUM-SEM images demonstrate the effectiveness of our proposed method.

Polarimetric signature imaging of anisotropic bio-medical tissues

NASA Astrophysics Data System (ADS)

Wu, Stewart H.; Yang, De-Ming; Chiou, Arthur; Nee, Soe-Mie F.; Nee, Tsu-Wei

2010-02-01

Polarimetric imaging of Stokes vector (I, Q, U, V) can provide 4 independent signatures showing the linear and circular polarizations of biological tissues and cells. Using a recently developed Stokes digital imaging system, we measured the Stokes vector images of tissue samples from sections of rat livers containing normal portions and hematomas. The derived Mueller matrix elements can quantitatively provide multi-signature data of the bio-sample. This polarimetric optical technology is a new option of biosensing technology to inspect the structures of tissue samples, particularly for discriminating tumor and non-tumor biopsy. This technology is useful for critical disease discrimination and medical diagnostics applications.

Back-scattered electron imaging of skeletal tissues.

PubMed

Boyde, A; Jones, S J

The use of solid-state back-scattered electron (BSE) detectors in the scanning electron microscopic study of skeletal tissues has been investigated. To minimize the topographic element in the image, flat samples and a ring detector configuration with the sample at normal incidence to the beam and the detector are used. Very flat samples are prepared by diamond micromilling or diamond polishing plastic-embedded tissue. Density discrimination in the image is so good that different density phases within mineralized bone can be imaged. For unembedded spongy bone, cut surfaces can be discriminated from natural surfaces by a topographic contrast mechanism. BSE imaging also presents advantages for unembedded samples with rough topography, such as anorganic preparations of the mineralization zone in cartilage, which give rise to severe charging problems with conventional secondary electron imaging.

Online quantitative analysis of multispectral images of human body tissues

NASA Astrophysics Data System (ADS)

Lisenko, S. A.

2013-08-01

A method is developed for online monitoring of structural and morphological parameters of biological tissues (haemoglobin concentration, degree of blood oxygenation, average diameter of capillaries and the parameter characterising the average size of tissue scatterers), which involves multispectral tissue imaging, image normalisation to one of its spectral layers and determination of unknown parameters based on their stable regression relation with the spectral characteristics of the normalised image. Regression is obtained by simulating numerically the diffuse reflectance spectrum of the tissue by the Monte Carlo method at a wide variation of model parameters. The correctness of the model calculations is confirmed by the good agreement with the experimental data. The error of the method is estimated under conditions of general variability of structural and morphological parameters of the tissue. The method developed is compared with the traditional methods of interpretation of multispectral images of biological tissues, based on the solution of the inverse problem for each pixel of the image in the approximation of different analytical models.

Characterization of human breast cancer tissues by infrared imaging.

PubMed

Verdonck, M; Denayer, A; Delvaux, B; Garaud, S; De Wind, R; Desmedt, C; Sotiriou, C; Willard-Gallo, K; Goormaghtigh, E

2016-01-21

Fourier Transform InfraRed (FTIR) spectroscopy coupled to microscopy (IR imaging) has shown unique advantages in detecting morphological and molecular pathologic alterations in biological tissues. The aim of this study was to evaluate the potential of IR imaging as a diagnostic tool to identify characteristics of breast epithelial cells and the stroma. In this study a total of 19 breast tissue samples were obtained from 13 patients. For 6 of the patients, we also obtained Non-Adjacent Non-Tumor tissue samples. Infrared images were recorded on the main cell/tissue types identified in all breast tissue samples. Unsupervised Principal Component Analyses and supervised Partial Least Square Discriminant Analyses (PLS-DA) were used to discriminate spectra. Leave-one-out cross-validation was used to evaluate the performance of PLS-DA models. Our results show that IR imaging coupled with PLS-DA can efficiently identify the main cell types present in FFPE breast tissue sections, i.e. epithelial cells, lymphocytes, connective tissue, vascular tissue and erythrocytes. A second PLS-DA model could distinguish normal and tumor breast epithelial cells in the breast tissue sections. A patient-specific model reached particularly high sensitivity, specificity and MCC rates. Finally, we showed that the stroma located close or at distance from the tumor exhibits distinct spectral characteristics. In conclusion FTIR imaging combined with computational algorithms could be an accurate, rapid and objective tool to identify/quantify breast epithelial cells and differentiate tumor from normal breast tissue as well as normal from tumor-associated stroma, paving the way to the establishment of a potential complementary tool to ensure safe tumor margins.

Deep ultraviolet resonant Raman imaging of a cell

NASA Astrophysics Data System (ADS)

Kumamoto, Yasuaki; Taguchi, Atsushi; Smith, Nicholas Isaac; Kawata, Satoshi

2012-07-01

We report the first demonstration of deep ultraviolet (DUV) Raman imaging of a cell. Nucleotide distributions in a HeLa cell were observed without any labeling at 257 nm excitation with resonant bands attributable to guanine and adenine. Obtained images represent DNA localization at nucleoli in the nucleus and RNA distribution in the cytoplasm. The presented technique extends the potential of Raman microscopy as a tool to selectively probe nucleic acids in a cell with high sensitivity due to resonance.

Ultrasonically Encoded Photoacoustic Flowgraphy in Biological Tissue

NASA Astrophysics Data System (ADS)

Wang, Lidai; Xia, Jun; Yao, Junjie; Maslov, Konstantin I.; Wang, Lihong V.

2013-11-01

Blood flow speed is an important functional parameter. Doppler ultrasound flowmetry lacks sufficient sensitivity to slow blood flow (several to tens of millimeters per second) in deep tissue. To address this challenge, we developed ultrasonically encoded photoacoustic flowgraphy combining ultrasonic thermal tagging with photoacoustic imaging. Focused ultrasound generates a confined heat source in acoustically absorptive fluid. Thermal waves propagate with the flow and are directly visualized in pseudo color using photoacoustic computed tomography. The Doppler shift is employed to calculate the flow speed. This method requires only acoustic and optical absorption, and thus is applicable to continuous fluid. A blood flow speed as low as 0.24mm·s-1 was successfully measured. Deep blood flow imaging was experimentally demonstrated under 5-mm-thick chicken breast tissue.

Transfer, Imaging, and Analysis Plate for Facile Handling of 384 Hanging Drop 3D Tissue Spheroids

PubMed Central

Cavnar, Stephen P.; Salomonsson, Emma; Luker, Kathryn E.; Luker, Gary D.; Takayama, Shuichi

2014-01-01

Three-dimensional culture systems bridge the experimental gap between in vivo and in vitro physiology. However, nonstandardized formation and limited downstream adaptability of 3D cultures have hindered mainstream adoption of these systems for biological applications, especially for low- and moderate-throughput assays commonly used in biomedical research. Here we build on our recent development of a 384-well hanging drop plate for spheroid culture to design a complementary spheroid transfer and imaging (TRIM) plate. The low-aspect ratio wells of the TRIM plate facilitated highfidelity, user-independent, contact-based collection of hanging drop spheroids. Using the TRIM plate, we demonstrated several downstream analyses, including bulk tissue collection for flow cytometry, high-resolution low working-distance immersion imaging, and timely reagent delivery for enzymatic studies. Low working-distance multiphoton imaging revealed a cell type–dependent, macroscopic spheroid structure. Unlike ovarian cancer spheroids, which formed loose, disk-shaped spheroids, human mammary fibroblasts formed tight, spherical, and nutrient-limited spheroids. Beyond the applications we describe here, we expect the hanging drop spheroid plate and complementary TRIM plate to facilitate analyses of spheroids across the spectrum of throughput, particularly for bulk collection of spheroids and high-content imaging. PMID:24051516

Transfer, imaging, and analysis plate for facile handling of 384 hanging drop 3D tissue spheroids.

PubMed

Cavnar, Stephen P; Salomonsson, Emma; Luker, Kathryn E; Luker, Gary D; Takayama, Shuichi

2014-04-01

Three-dimensional culture systems bridge the experimental gap between in vivo and in vitro physiology. However, nonstandardized formation and limited downstream adaptability of 3D cultures have hindered mainstream adoption of these systems for biological applications, especially for low- and moderate-throughput assays commonly used in biomedical research. Here we build on our recent development of a 384-well hanging drop plate for spheroid culture to design a complementary spheroid transfer and imaging (TRIM) plate. The low-aspect ratio wells of the TRIM plate facilitated high-fidelity, user-independent, contact-based collection of hanging drop spheroids. Using the TRIM plate, we demonstrated several downstream analyses, including bulk tissue collection for flow cytometry, high-resolution low working-distance immersion imaging, and timely reagent delivery for enzymatic studies. Low working-distance multiphoton imaging revealed a cell type-dependent, macroscopic spheroid structure. Unlike ovarian cancer spheroids, which formed loose, disk-shaped spheroids, human mammary fibroblasts formed tight, spherical, and nutrient-limited spheroids. Beyond the applications we describe here, we expect the hanging drop spheroid plate and complementary TRIM plate to facilitate analyses of spheroids across the spectrum of throughput, particularly for bulk collection of spheroids and high-content imaging.

X-ray Phase Contrast Imaging of Calcified Tissue and Biomaterial Structure in Bioreactor Engineered Tissues

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

Appel, Alyssa A.; Larson, Jeffery C.; Garson, III, Alfred B.

2014-11-04

Tissues engineered in bioreactor systems have been used clinically to replace damaged tissues and organs. In addition, these systems are under continued development for many tissue engineering applications. The ability to quantitatively assess material structure and tissue formation is critical for evaluating bioreactor efficacy and for preimplantation assessment of tissue quality. These techniques allow for the nondestructive and longitudinal monitoring of large engineered tissues within the bioreactor systems and will be essential for the translation of these strategies to viable clinical therapies. X-ray Phase Contrast (XPC) imaging techniques have shown tremendous promise for a number of biomedical applications owing tomore » their ability to provide image contrast based on multiple X-ray properties, including absorption, refraction, and scatter. In this research, mesenchymal stem cell-seeded alginate hydrogels were prepared and cultured under osteogenic conditions in a perfusion bioreactor. The constructs were imaged at various time points using XPC microcomputed tomography (µCT). Imaging was performed with systems using both synchrotron- and tube-based X-ray sources. XPC µCT allowed for simultaneous three-dimensional (3D) quantification of hydrogel size and mineralization, as well as spatial information on hydrogel structure and mineralization. Samples were processed for histological evaluation and XPC showed similar features to histology and quantitative analysis consistent with the histomorphometry. Furthermore, these results provide evidence of the significant potential of techniques based on XPC for noninvasive 3D imaging engineered tissues grown in bioreactors.« less

Distinguishing tracheal and esophageal tissues with hyperspectral imaging and fiber-optic sensing

NASA Astrophysics Data System (ADS)

Nawn, Corinne D.; Souhan, Brian E.; Carter, Robert, III; Kneapler, Caitlin; Fell, Nicholas; Ye, Jing Yong

2016-11-01

During emergency medical situations, where the patient has an obstructed airway or necessitates respiratory support, endotracheal intubation (ETI) is the medical technique of placing a tube into the trachea in order to facilitate adequate ventilation of the lungs. Complications during ETI, such as repeated attempts, failed intubation, or accidental intubation of the esophagus, can lead to severe consequences or ultimately death. Consequently, a need exists for a feedback mechanism to aid providers in performing successful ETI. Our study examined the spectral reflectance properties of the tracheal and esophageal tissue to determine whether a unique spectral profile exists for either tissue for the purpose of detection. The study began by using a hyperspectral camera to image excised pig tissue samples exposed to white and UV light in order to capture the spectral reflectance properties with high fidelity. After identifying a unique spectral characteristic of the trachea that significantly differed from esophageal tissue, a follow-up investigation used a fiber optic probe to confirm the detectability and consistency of the different reflectance characteristics in a pig model. Our results characterize the unique and consistent spectral reflectance characteristic of tracheal tissue, thereby providing foundational support for exploiting spectral properties to detect the trachea during medical procedures.

Terahertz time-lapse imaging of hydration in physiological tissues

NASA Astrophysics Data System (ADS)

Bennett, David B.; Taylor, Zachary D.; Bajwa, Neha; Tewari, Priyamvada; Maccabi, Ashkan; Sung, Shijun; Singh, Rahul S.; Culjat, Martin O.; Grundfest, Warren S.; Brown, Elliott R.

2011-02-01

This study describes terahertz (THz) imaging of hydration changes in physiological tissues with high water concentration sensitivity. A fast-scanning, pulsed THz imaging system (centered at 525 GHz; 125 GHz bandwidth) was utilized to acquire a 35 mm x 35 mm field-of-view with 0.5 mm x 0.5 mm pixels in less than two minutes. THz time-lapsed images were taken on three sample systems: (1) a simple binary system of water evaporating from a polypropylene towel, (2) the accumulation of fluid at the site of a sulfuric acid burn on ex vivo porcine skin, and (3) the evaporative dehydration of an ex vivo porcine cornea. The diffusion-regulating behavior of corneal tissue is elucidated, and the correlation of THz reflectivity with tissue hydration is measured using THz spectroscopy on four ex vivo corneas. We conclude that THz imaging can discern small differences in the distribution of water in physiological tissues and is a good candidate for burn and corneal imaging.

Cytopathological image analysis using deep-learning networks in microfluidic microscopy.

PubMed

Gopakumar, G; Hari Babu, K; Mishra, Deepak; Gorthi, Sai Siva; Sai Subrahmanyam, Gorthi R K

2017-01-01

Cytopathologic testing is one of the most critical steps in the diagnosis of diseases, including cancer. However, the task is laborious and demands skill. Associated high cost and low throughput drew considerable interest in automating the testing process. Several neural network architectures were designed to provide human expertise to machines. In this paper, we explore and propose the feasibility of using deep-learning networks for cytopathologic analysis by performing the classification of three important unlabeled, unstained leukemia cell lines (K562, MOLT, and HL60). The cell images used in the classification are captured using a low-cost, high-throughput cell imaging technique: microfluidics-based imaging flow cytometry. We demonstrate that without any conventional fine segmentation followed by explicit feature extraction, the proposed deep-learning algorithms effectively classify the coarsely localized cell lines. We show that the designed deep belief network as well as the deeply pretrained convolutional neural network outperform the conventionally used decision systems and are important in the medical domain, where the availability of labeled data is limited for training. We hope that our work enables the development of a clinically significant high-throughput microfluidic microscopy-based tool for disease screening/triaging, especially in resource-limited settings.

Magnetoencephalographic imaging of deep corticostriatal network activity during a rewards paradigm.

PubMed

Kanal, Eliezer Y; Sun, Mingui; Ozkurt, Tolga E; Jia, Wenyan; Sclabassi, Robert

2009-01-01

The human rewards network is a complex system spanning both cortical and subcortical regions. While much is known about the functions of the various components of the network, research on the behavior of the network as a whole has been stymied due to an inability to detect signals at a high enough temporal resolution from both superficial and deep network components simultaneously. In this paper, we describe the application of magnetoencephalographic imaging (MEG) combined with advanced signal processing techniques to this problem. Using data collected while subjects performed a rewards-related gambling paradigm demonstrated to activate the rewards network, we were able to identify neural signals which correspond to deep network activity. We also show that this signal was not observable prior to filtration. These results suggest that MEG imaging may be a viable tool for the detection of deep neural activity.

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs.

PubMed

Chen, Liang-Chieh; Papandreou, George; Kokkinos, Iasonas; Murphy, Kevin; Yuille, Alan L

2018-04-01

In this work we address the task of semantic image segmentation with Deep Learning and make three main contributions that are experimentally shown to have substantial practical merit. First, we highlight convolution with upsampled filters, or 'atrous convolution', as a powerful tool in dense prediction tasks. Atrous convolution allows us to explicitly control the resolution at which feature responses are computed within Deep Convolutional Neural Networks. It also allows us to effectively enlarge the field of view of filters to incorporate larger context without increasing the number of parameters or the amount of computation. Second, we propose atrous spatial pyramid pooling (ASPP) to robustly segment objects at multiple scales. ASPP probes an incoming convolutional feature layer with filters at multiple sampling rates and effective fields-of-views, thus capturing objects as well as image context at multiple scales. Third, we improve the localization of object boundaries by combining methods from DCNNs and probabilistic graphical models. The commonly deployed combination of max-pooling and downsampling in DCNNs achieves invariance but has a toll on localization accuracy. We overcome this by combining the responses at the final DCNN layer with a fully connected Conditional Random Field (CRF), which is shown both qualitatively and quantitatively to improve localization performance. Our proposed "DeepLab" system sets the new state-of-art at the PASCAL VOC-2012 semantic image segmentation task, reaching 79.7 percent mIOU in the test set, and advances the results on three other datasets: PASCAL-Context, PASCAL-Person-Part, and Cityscapes. All of our code is made publicly available online.

Automated and Adaptable Quantification of Cellular Alignment from Microscopic Images for Tissue Engineering Applications

PubMed Central