Feature Representation Learning for Robust Retinal Disease Detection from Optical Coherence Tomography Images

TL;DR

This paper introduces a novel feature learning architecture for retinal disease detection from OCT images, enhancing robustness and interpretability, and outperforming existing models in accuracy and transferability.

Contribution

The study proposes a three-head learning model with a new representation learning module to improve disease differentiation and robustness in retinal OCT image analysis.

Findings

Outperforms state-of-the-art models in accuracy

Enhances robustness to out-of-distribution data

Improves interpretability of disease features

Abstract

Ophthalmic images may contain identical-looking pathologies that can cause failure in automated techniques to distinguish different retinal degenerative diseases. Additionally, reliance on large annotated datasets and lack of knowledge distillation can restrict ML-based clinical support systems' deployment in real-world environments. To improve the robustness and transferability of knowledge, an enhanced feature-learning module is required to extract meaningful spatial representations from the retinal subspace. Such a module, if used effectively, can detect unique disease traits and differentiate the severity of such retinal degenerative pathologies. In this work, we propose a robust disease detection architecture with three learning heads, i) A supervised encoder for retinal disease classification, ii) An unsupervised decoder for the reconstruction of disease-specific spatial information, and iii) A novel representation learning module for learning the similarity between encoder-decoder feature and enhancing the accuracy of the model. Our experimental results on two publicly available OCT datasets illustrate that the proposed model outperforms existing state-of-the-art models in terms of accuracy, interpretability, and robustness for out-of-distribution retinal disease detection.