Training of Deep Convolutional Neural Networks to Identify Critical Liver Alterations in Histopathology Image Samples

Abstract

Nonalcoholic fatty liver disease (NAFLD) is responsible for a wide range of pathological disorders. It is characterized by the prevalence of steatosis, which results in excessive accumulation of triglyceride in the liver tissue. At high rates, it can lead to a partial or total occlusion of the organ. In contrast, nonalcoholic steatohepatitis (NASH) is a progressive form of NAFLD, with the inclusion of hepatocellular injury and inflammation histological diseases. Since there is no approved pharmacotherapeutic solution for both conditions, physicians and engineers are constantly in search for fast and accurate diagnostic methods. The proposed work introduces a fully automated classification approach, taking into consideration the high discrimination capability of four histological tissue alterations. The proposed work utilizes a deep supervised learning method, with a convolutional neural network (CNN) architecture achieving a classification accuracy of 95%. The classification capability of the new CNN model is compared with a pre-trained AlexNet model, a visual geometry group (VGG)-16 deep architecture and a conventional multilayer perceptron (MLP) artificial neural network. The results show that the constructed model can achieve better classification accuracy than VGG-16 (94%) and MLP (90.3%), while AlexNet emerges as the most efficient classifier (97%).

Analysis and Classification of H&E-Stained Oral Cavity Tumour Gradings Using Convolution Neural Network

“…The ReLU activation function is used to avoid the gradient of the loss with respect to weights generated in each layer as shown in Eq. (8).…”

“…In recent years Deep Convolutional Neural Network plays a major role in medical image analysis and proved to show better performances particularly, in cancer grading and classification problem. In [8] a deep CNN model was proposed to classify normal and abnormal non-alcoholic fatty liver disease. The proposed model was designed with three neural network layers for four class classification using Stochastic Gradient Descent (SGD) optimization technique.…”

Two-Stage CBR Based Healthcare Model to Diagnose Liver Disease

“…ANN has been widely used in medical diagnosis applications as it dynamically adapts itself according to the training dataset. For liver diagnosis, ANNs have been extensively used by Rong-yun Mai et al [13] to diagnose liver cirrhosis with Hepatocellular Carcinoma, by Z. Xu to diagnose stage of cirrhosis [14] , by Alexandros Arjmand et al [15] to find out alterations in liver in Histopathology images , by D Santhosh Reddy et al [16] to diagnose and to categorize the ultrasound liver parenchyma texture into four different classes, by Cheng-Hsiung Wenga [19] to diagnose liver disease , Kawaguchi, T., Tokushige, K., Hyogo, H. et al [8] have done data mining analysis on the diagnosis of nonalcoholic fatty liver disease-related hepatocellular carcinoma and concluded that patients treated with hepatectomy and a serum albumin level ≥3.7 g/dLG in Japan have more survival rate. ZhenjieYao et al [17] have used densely deep learning model to detect liver disease.…”

Morphological Estimation of Cellularity on Neo-Adjuvant Treated Breast Cancer Histological Images

“…CNNs have been used for different Histopathology tasks, like segmentation [ 48 , 49 ], detection of a specific image properties [ 50 ], and image grade classification [ 38 , 51 ]. Breast cancer tumour cellularity has also been addressed by deep techniques.…”