Towards the Localisation of Lesions in Diabetic Retinopathy

TL;DR

This paper applies Grad-CAM to deep learning models for diabetic retinopathy to produce lesion localisation maps, aiding ophthalmologists in early diagnosis by highlighting discriminative image regions.

Contribution

It introduces a comparative analysis of four deep learning models using Grad-CAM for lesion localisation in DR fundus images, highlighting InceptionV3's superior performance.

Findings

InceptionV3 achieved 96.07% accuracy.

InceptionV3 localised lesions more accurately and efficiently.

Grad-CAM effectively visualises discriminative regions for diagnosis.

Abstract

Convolutional Neural Networks (CNNs) have successfully been used to classify diabetic retinopathy (DR) fundus images in recent times. However, deeper representations in CNNs may capture higher-level semantics at the expense of spatial resolution. To make predictions usable for ophthalmologists, we use a post-attention technique called Gradient-weighted Class Activation Mapping (Grad-CAM) on the penultimate layer of deep learning models to produce coarse localisation maps on DR fundus images. This is to help identify discriminative regions in the images, consequently providing evidence for ophthalmologists to make a diagnosis and potentially save lives by early diagnosis. Specifically, this study uses pre-trained weights from four state-of-the-art deep learning models to produce and compare localisation maps of DR fundus images. The models used include VGG16, ResNet50, InceptionV3, and InceptionResNetV2. We find that InceptionV3 achieves the best performance with a test classification accuracy of 96.07%, and localise lesions better and faster than the other models.