A Global and Patch-wise Contrastive Loss for Accurate Automated Exudate Detection

TL;DR

This paper introduces a novel contrastive learning framework with patch-wise density contrasting and a discriminative edge inspection module to improve the segmentation of hard exudates in diabetic retinopathy detection, addressing shape and boundary challenges.

Contribution

It proposes a new supervised contrastive learning approach with patch-wise and boundary analysis modules for more accurate exudate segmentation.

Findings

Outperforms state-of-the-art methods on IDRiD dataset

Effectively segments small and ambiguous exudates

Demonstrates potential for computer-assisted diagnosis

Abstract

Diabetic retinopathy (DR) is a leading global cause of blindness. Early detection of hard exudates plays a crucial role in identifying DR, which aids in treating diabetes and preventing vision loss. However, the unique characteristics of hard exudates, ranging from their inconsistent shapes to indistinct boundaries, pose significant challenges to existing segmentation techniques. To address these issues, we present a novel supervised contrastive learning framework to optimize hard exudate segmentation. Specifically, we introduce a patch-wise density contrasting scheme to distinguish between areas with varying lesion concentrations, and therefore improve the model's proficiency in segmenting small lesions. To handle the ambiguous boundaries, we develop a discriminative edge inspection module to dynamically analyze the pixels that lie around the boundaries and accurately delineate the exudates. Upon evaluation using the IDRiD dataset and comparison with state-of-the-art frameworks, our method exhibits its effectiveness and shows potential for computer-assisted hard exudate detection. The code to replicate experiments is available at github.com/wetang7/HECL/.