Deep learning enables automated scoring of liver fibrosis stages
Current liver fibrosis scoring by computer-assisted image analytics is not fully automated as it requires manual preprocessing (segmentation and feature extraction) typically based on domain knowledge in liver pathology. Deep learning-based algorithms can potentially classify these images without the need for preprocessing through learning from a large dataset of images. We investigated the performance of classification models built using a deep learning-based algorithm pre-trained using
... rained using multiple sources of images to score liver fibrosis and compared them against conventional non-deep learning-based algorithms -artificial neural networks (ANN), multinomial logistic regression (MLR), support vector machines (SVM) and random forests (RF). Automated feature classification and fibrosis scoring were achieved by using a transfer learning-based deep learning network, AlexNet-Convolutional Neural Networks (CNN), with balanced area under receiver operating characteristic (AUROC) values of up to 0.85-0.95 versus ANN (AUROC of up to 0.87-1.00), MLR (AUROC of up to 0.73-1.00), SVM (AUROC of up to 0.69-0.99) and RF (AUROC of up to 0.94-0. 99). Results indicate that a deep learning-based algorithm with transfer learning enables the construction of a fully automated and accurate prediction model for scoring liver fibrosis stages that is comparable to other conventional non-deep learning-based algorithms that are not fully automated. Machine-learning is popular for the prediction, classification and assessment of liver fibrosis/cirrhosis, and response to therapy in hepatitis B and C patients 1-3 . At present, histopathological examination of liver biopsy samples remains as the 'gold standard' for liver fibrosis assessment. Histopathological features such as excessive accumulation of extracellular matrix (particularly collagen) and/or parenchymal extinction 4 are incorporated into descriptive or semi-quantitative scoring systems for disease staging. Examples include the Knodell histological activity index (HAI) 5 , Scheuer 6 , Ishak 1 , Metavir 7 , and Ishak-modified systems 8 . However, these scoring systems are inherently semi-quantitative and subject to intra-and inter-observer variability.