AUTOMATIC MUSCLE PERIMYSIUM ANNOTATION USING DEEP CONVOLUTIONAL NEURAL NETWORK

Deep Convolutional Hashing for Low-Dimensional Binary Embedding of Histopathological Images

Compact binary representations of histopa-thology images using hashing methods provide efficient approximate nearest neighbor search for direct visual query in large-scale databases. They can be utilized to measure the probability of the abnormality of the query image based on the retrieved similar cases, thereby providing support for medical diagnosis. They also allow for efficient managing of large-scale image databases because of a low storage requirement. However, the effectiveness of binary representations heavily relies on the visual descriptors that represent the semantic information in the histopathological images. Traditional approaches with hand-crafted visual descriptors might fail due to significant variations in image appearance. Recently, deep learning architectures provide promising solutions to address this problem using effective semantic representations. In this paper, we propose a deep convolutional hashing method that can be trained "point-wise" to simultaneously learn both semantic and binary representations of histopathological images. Specifically, we propose a convolutional neural network that introduces a latent binary encoding (LBE) layer for low-dimensional feature embedding to learn binary codes. We design a joint optimization objective function that encourages the network to learn discriminative representations from the label information, and reduce the gap between the real-valued low-dimensional embedded features and desired binary values. The binary encoding for new images can be obtained by forward propagating through the network and quantizing the output of the LBE layer. Experimental results on a large-scale histopathological image dataset demonstrate the effectiveness of the proposed method.

A neural network approach to analyze cross-sections of muscle fibers in pathological images

Morphological characteristics of muscle fibers, such as their cross-sections, are important indicators of the health and function of the musculoskeletal system. However, manual analysis of muscle fiber morphology is a labor-intensive and time-consuming process that is prone to errors. Overall, the procedure involves high inter- and intra-observer variability. Therefore, it is desirable for biologists to have a tool that can produce objective and reproducible analysis for muscle fiber images. In this work, we propose a deep convolutional neural network (DCNN) followed by post-processing for detecting and measuring the cross-sections of muscle fibers. We evaluate three segmentation networks for muscle boundary segmentation: (1) U-net, (2) FusionNet, and (3) a customized FusionNet. The customized FusionNet, which had the highest Dice coefficient on the test set, was used for subsequent morphological analysis of the muscle fibers. The proposed method was tested on microscopic images of the tibialis anterior muscles of a pre-clinical model of muscular dystrophy. The dataset contained four mosaic images, totalling more than 3400 fibers. Because of the severity of muscle injury in this pre-clinical model, its muscle fiber images present a challenge for quantitative analysis for several reasons. First, the muscle fibers had inhomogeneous spatial distribution and very different sizes. Second, the membranes of the muscle fibers had uneven signal intensity due to the loss of a membrane protein. Third, the shapes of intact muscle fibers were very different. All these factors contributed to the difficulty of acquiring good training data in the first place. Despite these difficulties, we achieved an average muscle fiber overlay precision of 0.65 and an average recall of 0.49. In this context, overlaid fibers are defined as fibers that have one or more pixels overlaying in the manual and DCNN cross-section segmentation. For the overlaid fibers, the proposed method achieved excellent segmentation accuracy of 94% ± 10.26%, as measured by the Dice-Sorensen coefficient.

Deep Learning in Microscopy Image Analysis: A Survey

Computerized microscopy image analysis plays an important role in computer aided diagnosis and prognosis. Machine learning techniques have powered many aspects of medical investigation and clinical practice. Recently, deep learning is emerging as a leading machine learning tool in computer vision and has attracted considerable attention in biomedical image analysis. In this paper, we provide a snapshot of this fast-growing field, specifically for microscopy image analysis. We briefly introduce the popular deep neural networks and summarize current deep learning achievements in various tasks, such as detection, segmentation, and classification in microscopy image analysis. In particular, we explain the architectures and the principles of convolutional neural networks, fully convolutional networks, recurrent neural networks, stacked autoencoders, and deep belief networks, and interpret their formulations or modelings for specific tasks on various microscopy images. In addition, we discuss the open challenges and the potential trends of future research in microscopy image analysis using deep learning.

AIIMDs: An Integrated Framework of Automatic Idiopathic Inflammatory Myopathy Diagnosis for Muscle

Idiopathic inflammatory myopathy (IIM) is a common skeletal muscle disease that relates to weakness and inflammation of muscle. Early diagnosis and prognosis of different types of IIMs will guide the effective treatment. Interpretation of digitized images of the cross-section muscle biopsy, which is currently done manually, provides the most reliable diagnostic information. With the increasing volume of images, the management and manual interpretation of the digitized muscle images suffer from low efficiency and high interobserver variabilities. In order to address these problems, we propose the first complete framework of automatic IIM diagnosis system for the management and interpretation of digitized skeletal muscle histopathology images. The proposed framework consists of several key components: (1) Automatic cell segmentation, perimysium annotation, and nuclei detection; (2) histogram-based feature extraction and quantification; (3) content-based image retrieval to search and retrieve similar cases in the database for comparative study; and (4) majority voting-based classification to provide decision support for computer-aided clinical diagnosis. Experiments show that the proposed diagnosis system provides efficient and robust interpretation of the digitized muscle image and computer-aided diagnosis of IIM.