Explainable Diabetic Retinopathy Detection and Retinal Image Generation

TL;DR

This paper introduces an interpretable deep learning approach for diabetic retinopathy detection that uses neuron activation patterns to explain predictions and a novel GAN to generate realistic retinal images with controllable lesions, aiding diagnosis and data augmentation.

Contribution

It presents a method to interpret DR detection via neuron activation descriptors and a new Patho-GAN for realistic, controllable retinal image synthesis, improving explainability and data augmentation.

Findings

Neuron activation patterns relate to lesions for explanation.

Patho-GAN generates high-quality, controllable retinal images.

Generated images outperform previous methods in quality and speed.

Abstract

Though deep learning has shown successful performance in classifying the label and severity stage of certain diseases, most of them give few explanations on how to make predictions. Inspired by Koch's Postulates, the foundation in evidence-based medicine (EBM) to identify the pathogen, we propose to exploit the interpretability of deep learning application in medical diagnosis. By determining and isolating the neuron activation patterns on which diabetic retinopathy (DR) detector relies to make decisions, we demonstrate the direct relation between the isolated neuron activation and lesions for a pathological explanation. To be specific, we first define novel pathological descriptors using activated neurons of the DR detector to encode both spatial and appearance information of lesions. Then, to visualize the symptom encoded in the descriptor, we propose Patho-GAN, a new network to synthesize medically plausible retinal images. By manipulating these descriptors, we could even arbitrarily control the position, quantity, and categories of generated lesions. We also show that our synthesized images carry the symptoms directly related to diabetic retinopathy diagnosis. Our generated images are both qualitatively and quantitatively superior to the ones by previous methods. Besides, compared to existing methods that take hours to generate an image, our second level speed endows the potential to be an effective solution for data augmentation.