Multi-Stage Classification of Retinal OCT Using Multi-Scale Ensemble Deep Architecture.

Accurate noninvasive diagnosis of retinal disorders is required for appropriate treatment or precision medicine. This work proposes a multi-stage classification network built on a multi-scale (pyramidal) feature ensemble architecture for retinal image classification using optical coherence tomography (OCT) images. First, a scale-adaptive neural network is developed to produce multi-scale inputs for feature extraction and ensemble learning. The larger input sizes yield more global information, while the smaller input sizes focus on local details. Then, a feature-rich pyramidal architecture is designed to extract multi-scale features as inputs using DenseNet as the backbone. The advantage of the hierarchical structure is that it allows the system to extract multi-scale, information-rich features for the accurate classification of retinal disorders. Evaluation on two public OCT datasets containing normal and abnormal retinas (e.g., diabetic macular edema (DME), choroidal neovascularization (CNV), age-related macular degeneration (AMD), and Drusen) and comparison against recent networks demonstrates the advantages of the proposed architecture's ability to produce feature-rich classification with average accuracy of 97.78%, 96.83%, and 94.26% for the first (binary) stage, second (three-class) stage, and all-at-once (four-class) classification, respectively, using cross-validation experiments using the first dataset. In the second dataset, our system showed an overall accuracy, sensitivity, and specificity of 99.69%, 99.71%, and 99.87%, respectively. Overall, the tangible advantages of the proposed network for enhanced feature learning might be used in various medical image classification tasks where scale-invariant features are crucial for precise diagnosis.

Preservation of functional descending input to paralyzed upper extremity muscles in motor complete cervical spinal cord injury.

OBJECTIVE: Spinal cord injury (SCI) is classified as complete or incomplete depending on the extent of sensorimotor preservation below the injury level. However, individuals with complete SCIs can voluntarily activate paralyzed lower limb muscles alone or by engaging non-paralyzed muscles during neurophysiological assessments, indicating presence of residual pathways across the injury. However, similar phenomena have not been explored for the upper extremity (UE) muscles following cervical SCIs. METHODS: Eighteen individuals with motor complete cervical SCI (AIS A or B) and five age-matched non-injured (NI) individuals performed various UE events against manual resistance during functional neurophysiological assessment (FNPA), and electromyographic (EMG) activity was recorded from UE muscles. RESULTS: Our findings demonstrated i) voluntary activation of clinically paralyzed muscles as evident from EMG readouts, ii) increased activity in these muscles during events engaging muscles above the injury level, iii) reduced spectral properties of paralyzed muscles in SCI compared to NI participants. CONCLUSIONS: Functional EMG activity in clinically paralyzed muscles indicate presence of residual pathways across the injury establishing supralesional control over the sublesional neural circuitry. SIGNIFICANCE: The findings may help explain the neurophysiological basis for UE recovery and can be exploited in designing rehabilitation techniques to facilitate UE recovery following cervical SCIs.

Multi-Stage Classification-Based Deep Learning for Gleason System Grading Using Histopathological Images.

In this work, we introduced an automated diagnostic system for Gleason system grading and grade groups (GG) classification using whole slide images (WSIs) of digitized prostate biopsy specimens (PBSs). Our system first classifies the Gleason pattern (GP) from PBSs and then identifies the Gleason score (GS) and GG. We developed a comprehensive DL-based approach to develop a grading pipeline system for the digitized PBSs and consider GP as a classification problem (not segmentation) compared to current research studies (deals with as a segmentation problem). A multilevel binary classification was implemented to enhance the segmentation accuracy for GP. Also, we created three levels of analysis (pyramidal levels) to extract different types of features. Each level has four shallow binary CNN to classify five GP labels. A majority fusion is applied for each pixel that has a total of 39 labeled images to create the final output for GP. The proposed framework is trained, validated, and tested on 3080 WSIs of PBS. The overall diagnostic accuracy for each CNN is evaluated using several metrics: precision (PR), recall (RE), and accuracy, which are documented by the confusion matrices.The results proved our system's potential for classifying all five GP and, thus, GG. The overall accuracy for the GG is evaluated using two metrics, PR and RE. The grade GG results are between 50% to 92% for RE and 50% to 92% for PR. Also, a comparison between our CNN architecture and the standard CNN (ResNet50) highlights our system's advantage. Finally, our deep-learning system achieved an agreement with the consensus grade groups.

A Cascaded Mutliresolution Ensemble Deep Learning Framework for Large Scale Alzheimer's Disease Detection using Brain MRIs.

Alzheimer's is progressive and irreversible type of dementia, which causes degeneration and death of cells and their connections in the brain. AD worsens over time and greatly impacts patients' life and affects their important mental functions, including thinking, the ability to carry on a conversation, and judgment and response to environment. Clinically, there is no single test to effectively diagnose Alzheimer disease. However, computed tomography (CT) and magnetic resonance imaging (MRI) scans can be used to help in AD diagnosis by observing critical changes in the size of different brain areas, typically parietal and temporal lobes areas. In this work, an integrative mulitresolutional ensemble deep learning-based framework is proposed to achieve better predictive performance for the diagnosis of Alzheimer disease. Unlike ResNet, DenseNet and their variants proposed pipeline utilizes PartialNet in a hierarchical design tailored to AD detection using brain MRIs. The advantage of the proposed analysis system is that PartialNet diversified the depth and deep supervision. Additionally, it also incorporates the properties of identity mappings which makes it powerful in better learning due to feature reuse. Besides, the proposed ensemble PartialNet is better in vanishing gradient, diminishing forward-flow with low number of parameters and better training time in comparison to its counter network. The proposed analysis pipeline has been tested and evaluated on benchmark ADNI dataset collected from 379 subjects patients. Quantitative validation of the obtained results documented our framework's capability, outperforming state-of-the-art learning approaches for both multi-and binary-class AD detection.

Kidney segmentation from DCE-MRI converging level set methods, fuzzy clustering and Markov random field modeling.

Early diagnosis of transplanted kidney function requires precise Kidney segmentation from Dynamic Contrast-Enhanced Magnetic Resonance Imaging images as a preliminary step. In this regard, this paper aims to propose an automated and accurate DCE-MRI kidney segmentation method integrating fuzzy c-means (FCM) clustering and Markov random field modeling into a level set formulation. The fuzzy memberships, kidney's shape prior model, and spatial interactions modeled using a second-order MRF guide the LS contour evolution towards the target kidney. Several experiments on real medical data of 45 subjects have shown that the proposed method can achieve high and consistent segmentation accuracy regardless of where the LS contour was initialized. It achieves an accuracy of 0.956 ± 0.019 in Dice similarity coefficient (DSC) and 1.15 ± 1.46 in 95% percentile of Hausdorff distance (HD95). Our quantitative comparisons confirm the superiority of the proposed method over several LS methods with an average improvement of more than 0.63 in terms of HD95. It also offers HD95 improvements of 9.62 and 3.94 over two deep neural networks based on the U-Net model. The accuracy improvements are experimentally found to be more profound on low-contrast images as well as DCE-MRI images with high noise levels.

Predicting the Level of Respiratory Support in COVID-19 Patients Using Machine Learning.

In this paper, a machine learning-based system for the prediction of the required level of respiratory support in COVID-19 patients is proposed. The level of respiratory support is divided into three classes: class 0 which refers to minimal support, class 1 which refers to non-invasive support, and class 2 which refers to invasive support. A two-stage classification system is built. First, the classification between class 0 and others is performed. Then, the classification between class 1 and class 2 is performed. The system is built using a dataset collected retrospectively from 3491 patients admitted to tertiary care hospitals at the University of Louisville Medical Center. The use of the feature selection method based on analysis of variance is demonstrated in the paper. Furthermore, a dimensionality reduction method called principal component analysis is used. XGBoost classifier achieves the best classification accuracy (84%) in the first stage. It also achieved optimal performance in the second stage, with a classification accuracy of 83%.

A deep learning framework for automated classification of histopathological kidney whole-slide images.

Background: Renal cell carcinoma is the most common type of malignant kidney tumor and is responsible for 14,830 deaths per year in the United States. Among the four most common subtypes of renal cell carcinoma, clear cell renal cell carcinoma has the worst prognosis and clear cell papillary renal cell carcinoma appears to have no malignant potential. Distinction between these two subtypes can be difficult due to morphologic overlap on examination of histopathological preparation stained with hematoxylin and eosin. Ancillary techniques, such as immunohistochemistry, can be helpful, but they are not universally available. We propose and evaluate a new deep learning framework for tumor classification tasks to distinguish clear cell renal cell carcinoma from papillary renal cell carcinoma. Methods: Our deep learning framework is composed of three convolutional neural networks. We divided whole-slide kidney images into patches with three different sizes where each network processes a specific patch size. Our framework provides patchwise and pixelwise classification. The histopathological kidney data is composed of 64 image slides that belong to 4 categories: fat, parenchyma, clear cell renal cell carcinoma, and clear cell papillary renal cell carcinoma. The final output of our framework is an image map where each pixel is classified into one class. To maintain consistency, we processed the map with Gauss-Markov random field smoothing. Results: Our framework succeeded in classifying the four classes and showed superior performance compared to well-established state-of-the-art methods (pixel accuracy: 0.89 ResNet18, 0.92 proposed). Conclusions: Deep learning techniques have a significant potential for cancer diagnosis.

Biomimetic cardiac tissue culture model (CTCM) to emulate cardiac physiology and pathophysiology ex vivo.

There is need for a reliable in vitro system that can accurately replicate the cardiac physiological environment for drug testing. The limited availability of human heart tissue culture systems has led to inaccurate interpretations of cardiac-related drug effects. Here, we developed a cardiac tissue culture model (CTCM) that can electro-mechanically stimulate heart slices with physiological stretches in systole and diastole during the cardiac cycle. After 12 days in culture, this approach partially improved the viability of heart slices but did not completely maintain their structural integrity. Therefore, following small molecule screening, we found that the incorporation of 100 nM tri-iodothyronine (T3) and 1 µM dexamethasone (Dex) into our culture media preserved the microscopic structure of the slices for 12 days. When combined with T3/Dex treatment, the CTCM system maintained the transcriptional profile, viability, metabolic activity, and structural integrity for 12 days at the same levels as the fresh heart tissue. Furthermore, overstretching the cardiac tissue induced cardiac hypertrophic signaling in culture, which provides a proof of concept for the ability of the CTCM to emulate cardiac stretch-induced hypertrophic conditions. In conclusion, CTCM can emulate cardiac physiology and pathophysiology in culture for an extended time, thereby enabling reliable drug screening.

Conditional GANs based system for fibrosis detection and quantification in Hematoxylin and Eosin whole slide images.

Assessing the degree of liver fibrosis is fundamental for the management of patients with chronic liver disease, in liver transplants procedures, and in general liver disease research. The fibrosis stage is best assessed by histopathologic evaluation, and Masson's Trichrome stain (MT) is the stain of choice for this task in many laboratories around the world. However, the most used stain in histopathology is Hematoxylin Eosin (HE) which is cheaper, has a faster turn-around time and is the primary stain routinely used for evaluation of liver specimens. In this paper, we propose a novel digital pathology system that accurately detects and quantifies the footprint of fibrous tissue in HE whole slide images (WSI). The proposed system produces virtual MT images from HE using a deep learning model that learns deep texture patterns associated with collagen fibers. The training pipeline is based on conditional generative adversarial networks (cGAN), which can achieve accurate pixel-level transformation. Our comprehensive training pipeline features an automatic WSI registration algorithm, which qualifies the HE/MT training slides for the cGAN model. Using liver specimens collected during liver transplantation procedures, we conducted a range of experiments to evaluate the detected footprint of selected anatomical features. Our evaluation includes both image similarity and semantic segmentation metrics. The proposed system achieved enhanced results in the experiments with significant improvement over the state-of-the-art CycleGAN learning style, and over direct prediction of fibrosis in HE without having the virtual MT step.

The Role of Medical Image Modalities and AI in the Early Detection, Diagnosis and Grading of Retinal Diseases: A Survey.

Traditional dilated ophthalmoscopy can reveal diseases, such as age-related macular degeneration (AMD), diabetic retinopathy (DR), diabetic macular edema (DME), retinal tear, epiretinal membrane, macular hole, retinal detachment, retinitis pigmentosa, retinal vein occlusion (RVO), and retinal artery occlusion (RAO). Among these diseases, AMD and DR are the major causes of progressive vision loss, while the latter is recognized as a world-wide epidemic. Advances in retinal imaging have improved the diagnosis and management of DR and AMD. In this review article, we focus on the variable imaging modalities for accurate diagnosis, early detection, and staging of both AMD and DR. In addition, the role of artificial intelligence (AI) in providing automated detection, diagnosis, and staging of these diseases will be surveyed. Furthermore, current works are summarized and discussed. Finally, projected future trends are outlined. The work done on this survey indicates the effective role of AI in the early detection, diagnosis, and staging of DR and/or AMD. In the future, more AI solutions will be presented that hold promise for clinical applications.

The Role of Different Retinal Imaging Modalities in Predicting Progression of Diabetic Retinopathy: A Survey.

Diabetic retinopathy (DR) is a devastating condition caused by progressive changes in the retinal microvasculature. It is a leading cause of retinal blindness in people with diabetes. Long periods of uncontrolled blood sugar levels result in endothelial damage, leading to macular edema, altered retinal permeability, retinal ischemia, and neovascularization. In order to facilitate rapid screening and diagnosing, as well as grading of DR, different retinal modalities are utilized. Typically, a computer-aided diagnostic system (CAD) uses retinal images to aid the ophthalmologists in the diagnosis process. These CAD systems use a combination of machine learning (ML) models (e.g., deep learning (DL) approaches) to speed up the diagnosis and grading of DR. In this way, this survey provides a comprehensive overview of different imaging modalities used with ML/DL approaches in the DR diagnosis process. The four imaging modalities that we focused on are fluorescein angiography, fundus photographs, optical coherence tomography (OCT), and OCT angiography (OCTA). In addition, we discuss limitations of the literature that utilizes such modalities for DR diagnosis. In addition, we introduce research gaps and provide suggested solutions for the researchers to resolve. Lastly, we provide a thorough discussion about the challenges and future directions of the current state-of-the-art DL/ML approaches. We also elaborate on how integrating different imaging modalities with the clinical information and demographic data will lead to promising results for the scientists when diagnosing and grading DR. As a result of this article's comparative analysis and discussion, it remains necessary to use DL methods over existing ML models to detect DR in multiple modalities.

Automated Diagnosis of Optical Coherence Tomography Angiography (OCTA) Based on Machine Learning Techniques.

Diabetic retinopathy (DR) refers to the ophthalmological complications of diabetes mellitus. It is primarily a disease of the retinal vasculature that can lead to vision loss. Optical coherence tomography angiography (OCTA) demonstrates the ability to detect the changes in the retinal vascular system, which can help in the early detection of DR. In this paper, we describe a novel framework that can detect DR from OCTA based on capturing the appearance and morphological markers of the retinal vascular system. This new framework consists of the following main steps: (1) extracting retinal vascular system from OCTA images based on using joint Markov-Gibbs Random Field (MGRF) model to model the appearance of OCTA images and (2) estimating the distance map inside the extracted vascular system to be used as imaging markers that describe the morphology of the retinal vascular (RV) system. The OCTA images, extracted vascular system, and the RV-estimated distance map is then composed into a three-dimensional matrix to be used as an input to a convolutional neural network (CNN). The main motivation for using this data representation is that it combines the low-level data as well as high-level processed data to allow the CNN to capture significant features to increase its ability to distinguish DR from the normal retina. This has been applied on multi-scale levels to include the original full dimension images as well as sub-images extracted from the original OCTA images. The proposed approach was tested on in-vivo data using about 91 patients, which were qualitatively graded by retinal experts. In addition, it was quantitatively validated using datasets based on three metrics: sensitivity, specificity, and overall accuracy. Results showed the capability of the proposed approach, outperforming the current deep learning as well as features-based detecting DR approaches.

Thoracolumbar epidural stimulation effects on bladder and bowel function in uninjured and chronic transected anesthetized rats.

Pre-clinical studies have shown that spinal cord epidural stimulation (scES) at the level of pelvic and pudendal nerve inputs/outputs (L5-S1) alters storage and/or emptying functions of both the bladder and bowel. The current mapping experiments were conducted to investigate scES efficacy at the level of hypogastric nerve inputs/outputs (T13-L2) in male and female rats under urethane anesthesia. As found with L5-S1 scES, T13-L2 scES at select frequencies and intensities of stimulation produced an increase in inter-contraction interval (ICI) in non-injured female rats but a short-latency void in chronic T9 transected rats, as well as reduced rectal activity in all groups. However, the detrusor pressure during the lengthened ICI (i.e., urinary hold) remained at a low pressure and was not elevated as seen with L5-S1 scES, an effect that's critical for translation to the clinic as high fill pressures can damage the kidneys. Furthermore, T13-L2 scES was shown to stimulate voiding post-transection by increasing bladder activity while also directly inhibiting the external urethral sphincter, a pattern necessary to overcome detrusor-sphincter dyssynergia. Additionally, select scES parameters at T13-L2 also increased distal colon activity in all groups. Together, the current findings suggest that optimization of scES for bladder and bowel will likely require multiple electrode cohorts at different locations that target circuitries coordinating sympathetic, parasympathetic and somatic outputs.

A Novel Computer-Aided Diagnostic System for Early Detection of Diabetic Retinopathy Using 3D-OCT Higher-Order Spatial Appearance Model.

Early diagnosis of diabetic retinopathy (DR) is of critical importance to suppress severe damage to the retina and/or vision loss. In this study, an optical coherence tomography (OCT)-based computer-aided diagnosis (CAD) method is proposed to detect DR early using structural 3D retinal scans. This system uses prior shape knowledge to automatically segment all retinal layers of the 3D-OCT scans using an adaptive, appearance-based method. After the segmentation step, novel texture features are extracted from the segmented layers of the OCT B-scans volume for DR diagnosis. For every layer, Markov-Gibbs random field (MGRF) model is used to extract the 2nd-order reflectivity. In order to represent the extracted image-derived features, we employ cumulative distribution function (CDF) descriptors. For layer-wise classification in 3D volume, using the extracted Gibbs energy feature, an artificial neural network (ANN) is fed the extracted feature for every layer. Finally, the classification outputs for all twelve layers are fused using a majority voting schema for global subject diagnosis. A cohort of 188 3D-OCT subjects are used for system evaluation using different k-fold validation techniques and different validation metrics. Accuracy of 90.56%, 93.11%, and 96.88% are achieved using 4-, 5-, and 10-fold cross-validation, respectively. Additional comparison with deep learning networks, which represent the state-of-the-art, documented the promise of our system's ability to diagnose the DR early.

Artificial Intelligence Based Framework to Quantify the Cardiomyocyte Structural Integrity in Heart Slices.

PURPOSE: Drug induced cardiac toxicity is a disruption of the functionality of cardiomyocytes which is highly correlated to the organization of the subcellular structures. We can analyze cellular structures by utilizing microscopy imaging data. However, conventional image analysis methods might miss structural deteriorations that are difficult to perceive. Here, we propose an image-based deep learning pipeline for the automated quantification of drug induced structural deteriorations using a 3D heart slice culture model. METHODS: In our deep learning pipeline, we quantify the induced structural deterioration from three anticancer drugs (doxorubicin, sunitinib, and herceptin) with known adverse cardiac effects. The proposed deep learning framework is composed of three convolutional neural networks that process three different image sizes. The results of the three networks are combined to produce a classification map that shows the locations of the structural deteriorations in the input cardiac image. RESULTS: The result of our technique is the capability of producing classification maps that accurately detect drug induced structural deterioration on the pixel level. CONCLUSION: This technology could be widely applied to perform unbiased quantification of the structural effect of the cardiotoxins on heart slices.

Texture and shape analysis of diffusion-weighted imaging for thyroid nodules classification using machine learning.

PURPOSE: To assess whether the integration between (a) functional imaging features that will be extracted from diffusion-weighted imaging (DWI); and (b) shape and texture imaging features as well as volumetric features that will be extracted from T2-weighted magnetic resonance imaging (MRI) can noninvasively improve the diagnostic accuracy of thyroid nodules classification. PATIENTS AND METHODS: In a retrospective study of 55 patients with pathologically proven thyroid nodules, T2-weighted and diffusion-weighted MRI scans of the thyroid gland were acquired. Spatial maps of the apparent diffusion coefficient (ADC) were reconstructed in all cases. To quantify the nodules' morphology, we used spherical harmonics as a new parametric shape descriptor to describe the complexity of the thyroid nodules in addition to traditional volumetric descriptors (e.g., tumor volume and cuboidal volume). To capture the inhomogeneity of the texture of the thyroid nodules, we used the histogram-based statistics (e.g., kurtosis, entropy, skewness, etc.) of the T2-weighted signal. To achieve the main goal of this paper, a fusion system using an artificial neural network (NN) is proposed to integrate both the functional imaging features (ADC) with the structural morphology and texture features. This framework has been tested on 55 patients (20 patients with malignant nodules and 35 patients with benign nodules), using leave-one-subject-out (LOSO) for training/testing validation tests. RESULTS: The functionality, morphology, and texture imaging features were estimated for 55 patients. The accuracy of the computer-aided diagnosis (CAD) system steadily improved as we integrate the proposed imaging features. The fusion system combining all biomarkers achieved a sensitivity, specificity, positive predictive value, negative predictive value, F1-score, and accuracy of 92.9 % $92.9\%$ (confidence interval [CI]: 78.9 % -- 99.5 % $78.9\%\text{--}99.5\%$ ), 95.8 % $95.8\%$ (CI: 87.4 % -- 99.7 % $87.4\%\text{--}99.7\%$ ), 93 % $93\%$ (CI: 80.7 % -- 99.5 % $80.7\%\text{--}99.5\%$ ), 96 % $96\%$ (CI: 88.8 % -- 99.7 % $88.8\%\text{--}99.7\%$ ), 92.8 % $92.8\%$ (CI: 83.5 % -- 98.5 % $83.5\%\text{--}98.5\%$ ), and 95.5 % $95.5\%$ (CI: 88.8 % -- 99.2 % $88.8\%\text{--}99.2\%$ ), respectively, using the LOSO cross-validation approach. CONCLUSION: The results demonstrated in this paper show the promise that integrating the functional features with morphology as well as texture features by using the current state-of-the-art machine learning approaches will be extremely useful for identifying thyroid nodules as well as diagnosing their malignancy.

A pyramidal deep learning pipeline for kidney whole-slide histology images classification.

Renal cell carcinoma is the most common type of kidney cancer. There are several subtypes of renal cell carcinoma with distinct clinicopathologic features. Among the subtypes, clear cell renal cell carcinoma is the most common and tends to portend poor prognosis. In contrast, clear cell papillary renal cell carcinoma has an excellent prognosis. These two subtypes are primarily classified based on the histopathologic features. However, a subset of cases can a have a significant degree of histopathologic overlap. In cases with ambiguous histologic features, the correct diagnosis is dependent on the pathologist's experience and usage of immunohistochemistry. We propose a new method to address this diagnostic task based on a deep learning pipeline for automated classification. The model can detect tumor and non-tumoral portions of kidney and classify the tumor as either clear cell renal cell carcinoma or clear cell papillary renal cell carcinoma. Our framework consists of three convolutional neural networks and the whole slide images of kidney which were divided into patches of three different sizes for input into the networks. Our approach can provide patchwise and pixelwise classification. The kidney histology images consist of 64 whole slide images. Our framework results in an image map that classifies the slide image on the pixel-level. Furthermore, we applied generalized Gauss-Markov random field smoothing to maintain consistency in the map. Our approach classified the four classes accurately and surpassed other state-of-the-art methods, such as ResNet (pixel accuracy: 0.89 Resnet18, 0.92 proposed). We conclude that deep learning has the potential to augment the pathologist's capabilities by providing automated classification for histopathological images.