Interactive Joint Feature Extraction for Diabetic Retinopathy Classification using Multiple Instance Neural Symbolic Learning

Sri Venkatesa Perumal College of Engineering & Technology

In recent years deep learning (DL) techniques have provided state-of-the-art performance on different medical imaging tasks. However the availability of good quality annotated medical data is very cha lenging due to involved time constraints and the availability of expert annotators e.g. radiologists. In addition DL is data-hungry and their training requires extensive computational resources. Another problem with DL is their black-box nature and lack of transparency on its inner working which inhibits causal understanding and reasoning Automatic classification of diabetic retinopathy from retinal images has been widely studied using deep neural networks with impressive results. However there is a clinical need for estimation of the uncertainty in the classifications a shortcoming of modern neural networks. Recently approximate Bayesian deep learning methods have been proposed for the task but the studies have only considered the binary referable/non-referable diabetic retinopathy classification applied to benchmark datasets. We present novel results by systematica ly investigating a clinical dataset and a clinica ly relevant -class classification scheme in addition to benchmark datasets and the binary classification scheme. Moreover we derive a connection between uncertainty measures and classifier risk from which we develop a new uncertainty measure. We observe that the previously proposed entropy-based uncertainty measure generalizes to the clinical dataset on the binary classification scheme.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.