Improving Lesion Segmentation for Diabetic Retinopathy using Adversarial Learning

TL;DR

This paper introduces an end-to-end adversarial learning framework that enhances the accuracy of pixel-level lesion segmentation in diabetic retinopathy fundus images, leveraging a cGAN with a specialized loss function.

Contribution

It proposes integrating HEDNet with a cGAN and a novel loss function to improve DR lesion segmentation accuracy.

Findings

Adversarial loss improves segmentation performance.

The system outperforms baseline methods.

Enhanced detection of DR lesions in fundus images.

Abstract

Diabetic Retinopathy (DR) is a leading cause of blindness in working age adults. DR lesions can be challenging to identify in fundus images, and automatic DR detection systems can offer strong clinical value. Of the publicly available labeled datasets for DR, the Indian Diabetic Retinopathy Image Dataset (IDRiD) presents retinal fundus images with pixel-level annotations of four distinct lesions: microaneurysms, hemorrhages, soft exudates and hard exudates. We utilize the HEDNet edge detector to solve a semantic segmentation task on this dataset, and then propose an end-to-end system for pixel-level segmentation of DR lesions by incorporating HEDNet into a Conditional Generative Adversarial Network (cGAN). We design a loss function that adds adversarial loss to segmentation loss. Our experiments show that the addition of the adversarial loss improves the lesion segmentation performance over the baseline.