Combining Fine- and Coarse-Grained Classifiers for Diabetic Retinopathy Detection

TL;DR

This paper introduces an ensemble approach combining coarse- and fine-grained classifiers for diabetic retinopathy detection, inspired by ophthalmologists' diagnostic process, showing significant performance improvements on public datasets.

Contribution

The paper proposes a novel ensemble method that integrates coarse- and fine-grained classifiers for improved diabetic retinopathy detection, inspired by clinical examination strategies.

Findings

The ensemble outperforms individual classifiers and most existing methods.

Fine-grained classifiers perform notably better than coarse-grained ones.

The approach achieves high accuracy across binary, ternary, and quaternary classification tasks.

Abstract

Visual artefacts of early diabetic retinopathy in retinal fundus images are usually small in size, inconspicuous, and scattered all over retina. Detecting diabetic retinopathy requires physicians to look at the whole image and fixate on some specific regions to locate potential biomarkers of the disease. Therefore, getting inspiration from ophthalmologist, we propose to combine coarse-grained classifiers that detect discriminating features from the whole images, with a recent breed of fine-grained classifiers that discover and pay particular attention to pathologically significant regions. To evaluate the performance of this proposed ensemble, we used publicly available EyePACS and Messidor datasets. Extensive experimentation for binary, ternary and quaternary classification shows that this ensemble largely outperforms individual image classifiers as well as most of the published works in most training setups for diabetic retinopathy detection. Furthermore, the performance of fine-grained classifiers is found notably superior than coarse-grained image classifiers encouraging the development of task-oriented fine-grained classifiers modelled after specialist ophthalmologists.