OCT-SelfNet: A Self-Supervised Framework with Multi-Modal Datasets for Generalized and Robust Retinal Disease Detection

TL;DR

OCT-SelfNet introduces a self-supervised learning framework that leverages multi-modal OCT datasets and a two-phase training process to improve retinal disease detection accuracy and robustness in clinical settings.

Contribution

This work presents a novel self-supervised framework using a mask autoencoder and multi-modal data, enhancing generalization and robustness in retinal disease detection compared to traditional supervised methods.

Findings

Achieved over 77% AUC-ROC across datasets, outperforming baseline models.

Exceeded 42% AUC-PR, showing significant performance improvement.

Demonstrated robustness in low data and unseen data scenarios.

Abstract

Despite the revolutionary impact of AI and the development of locally trained algorithms, achieving widespread generalized learning from multi-modal data in medical AI remains a significant challenge. This gap hinders the practical deployment of scalable medical AI solutions. Addressing this challenge, our research contributes a self-supervised robust machine learning framework, OCT-SelfNet, for detecting eye diseases using optical coherence tomography (OCT) images. In this work, various data sets from various institutions are combined enabling a more comprehensive range of representation. Our method addresses the issue using a two-phase training approach that combines self-supervised pretraining and supervised fine-tuning with a mask autoencoder based on the SwinV2 backbone by providing a solution for real-world clinical deployment. Extensive experiments on three datasets with different encoder backbones, low data settings, unseen data settings, and the effect of augmentation show that our method outperforms the baseline model, Resnet-50 by consistently attaining AUC-ROC performance surpassing 77% across all tests, whereas the baseline model exceeds 54%. Moreover, in terms of the AUC-PR metric, our proposed method exceeded 42%, showcasing a substantial increase of at least 10% in performance compared to the baseline, which exceeded only 33%. This contributes to our understanding of our approach's potential and emphasizes its usefulness in clinical settings.