Explainable AI: Comparative Analysis of Normal and Dilated ResNet Models for Fundus Disease Classification

TL;DR

This paper compares normal and dilated ResNet models for retinal disease classification, demonstrating that dilated ResNet improves accuracy and transparency in AI-based diagnosis using the ODIR dataset.

Contribution

It introduces dilated ResNet models for fundus disease classification and evaluates their performance against normal ResNet models, emphasizing explainability and efficiency.

Findings

Dilated ResNet achieves higher F1 scores than normal ResNet.

Dilated ResNet enhances receptive field and classification accuracy.

Models show promising results on the ODIR dataset.

Abstract

This paper presents dilated Residual Network (ResNet) models for disease classification from retinal fundus images. Dilated convolution filters are used to replace normal convolution filters in the higher layers of the ResNet model (dilated ResNet) in order to improve the receptive field compared to the normal ResNet model for disease classification. This study introduces computer-assisted diagnostic tools that employ deep learning, enhanced with explainable AI techniques. These techniques aim to make the tool's decision-making process transparent, thereby enabling medical professionals to understand and trust the AI's diagnostic decision. They are particularly relevant in today's healthcare landscape, where there is a growing demand for transparency in AI applications to ensure their reliability and ethical use. The dilated ResNet is used as a replacement for the normal ResNet to enhance the classification accuracy of retinal eye diseases and reduce the required computing time. The dataset used in this work is the Ocular Disease Intelligent Recognition (ODIR) dataset which is a structured ophthalmic database with eight classes covering most of the common retinal eye diseases. The evaluation metrics used in this work include precision, recall, accuracy, and F1 score. In this work, a comparative study has been made between normal ResNet models and dilated ResNet models on five variants namely ResNet-18, ResNet-34, ResNet-50, ResNet-101, and ResNet-152. The dilated ResNet model shows promising results as compared to normal ResNet with an average F1 score of 0.71, 0.70, 0.69, 0.67, and 0.70 respectively for the above respective variants in ODIR multiclass disease classification.