Automatic detection and prediction of nAMD activity change in retinal OCT using Siamese networks and Wasserstein Distance for ordinality

TL;DR

This paper introduces deep learning models using Siamese networks and Wasserstein Distance to detect and predict nAMD activity changes from retinal OCT scans, aiding timely treatment decisions.

Contribution

It presents a novel application of Wasserstein Distance-based loss for ordinal severity change prediction in nAMD using Siamese networks and Vision Transformers.

Findings

Models ranked high on the MICCAI 2024 MARIO Challenge leaderboard.

The Wasserstein Distance loss effectively captures ordinal relations in severity changes.

Deep learning models can assist in managing nAMD treatment timing.

Abstract

Neovascular age-related macular degeneration (nAMD) is a leading cause of vision loss among older adults, where disease activity detection and progression prediction are critical for nAMD management in terms of timely drug administration and improving patient outcomes. Recent advancements in deep learning offer a promising solution for predicting changes in AMD from optical coherence tomography (OCT) retinal volumes. In this work, we proposed deep learning models for the two tasks of the public MARIO Challenge at MICCAI 2024, designed to detect and forecast changes in nAMD severity with longitudinal retinal OCT. For the first task, we employ a Vision Transformer (ViT) based Siamese Network to detect changes in AMD severity by comparing scan embeddings of a patient from different time points. To train a model to forecast the change after 3 months, we exploit, for the first time, an Earth Mover (Wasserstein) Distance-based loss to harness the ordinal relation within the severity change classes. Both models ranked high on the preliminary leaderboard, demonstrating that their predictive capabilities could facilitate nAMD treatment management.