Learned Pre-Processing for Automatic Diabetic Retinopathy Detection on Eye Fundus Images

TL;DR

This paper introduces a learned pre-processing step involving shadow removal and color correction for eye fundus images, significantly enhancing the accuracy of diabetic retinopathy detection.

Contribution

It proposes a novel Shadow Removal Layer that learns task-specific pre-processing, improving detection performance over traditional fixed methods.

Findings

Learning the pre-processing improves detection accuracy.

The method outperforms traditional pre-processing techniques.

The approach adapts to specific image conditions.

Abstract

Diabetic Retinopathy is the leading cause of blindness in the working-age population of the world. The main aim of this paper is to improve the accuracy of Diabetic Retinopathy detection by implementing a shadow removal and color correction step as a preprocessing stage from eye fundus images. For this, we rely on recent findings indicating that application of image dehazing on the inverted intensity domain amounts to illumination compensation. Inspired by this work, we propose a Shadow Removal Layer that allows us to learn the pre-processing function for a particular task. We show that learning the pre-processing function improves the performance of the network on the Diabetic Retinopathy detection task.