Adaptive Class Learning to Screen Diabetic Disorders in Fundus Images of Eye

TL;DR

The paper addresses the need for accurate, early screening of ocular diseases from fundus images. It introduces Class Extension with Limited Data (CELD), a staged learning framework that starts by training on Healthy and Diabetic Retinopathy features and then fine-tunes to classify Healthy, DR, and Glaucoma, complemented by perturbation-based explainability. A key result is an overall accuracy of 91% on public datasets, demonstrating data-efficient expansion to a three-class retinal disease task. The approach offers a practical pathway for scalable retinal disease screening with limited labeled data and provides interpretable insights to support clinical deployment.

Abstract

The prevalence of ocular illnesses is growing globally, presenting a substantial public health challenge. Early detection and timely intervention are crucial for averting visual impairment and enhancing patient prognosis. This research introduces a new framework called Class Extension with Limited Data (CELD) to train a classifier to categorize retinal fundus images. The classifier is initially trained to identify relevant features concerning Healthy and Diabetic Retinopathy (DR) classes and later fine-tuned to adapt to the task of classifying the input images into three classes: Healthy, DR, and Glaucoma. This strategy allows the model to gradually enhance its classification capabilities, which is beneficial in situations where there are only a limited number of labeled datasets available. Perturbation methods are also used to identify the input image characteristics responsible for influencing the models decision-making process. We achieve an overall accuracy of 91% on publicly available datasets.

Figures (1)

• Figure 1: A figure caption is always placed below the illustration. Please note that short captions are centered, while long ones are justified by the macro package automatically.

Theorems & Definitions (2)

• theorem 1
• proof