Deep Learning for Retinal Degeneration Assessment: A Comprehensive Analysis of the MARIO AMD Progression Challenge

2025-06-03

AI Summary

This study addresses the challenge of automated, dynamic assessment of age-related macular degeneration (AMD). We propose the first multimodal deep learning framework integrating optical coherence tomography (OCT), infrared imaging, and structured clinical variables (e.g., age, visit frequency) to jointly perform: (1) classification of neovascular activity evolution across sequential OCT scans, and (2) prediction of 3-month progression risk following anti-VEGF therapy. Our method combines 3D-CNNs, Transformers, transfer learning, uncertainty quantification, and robustness optimization across devices and populations. We introduce, at MICCAI, the first AMD progression challenge explicitly designed for domain generalization—validated on an external Algerian cohort—and clinical interpretability. Experiments show lesion evolution classification performance matching expert clinicians (AUC >0.92), whereas longitudinal progression prediction remains substantially inferior, highlighting a critical bottleneck in modeling individualized treatment response.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

The MARIO challenge, held at MICCAI 2024, focused on advancing the automated detection and monitoring of age-related macular degeneration (AMD) through the analysis of optical coherence tomography (OCT) images. Designed to evaluate algorithmic performance in detecting neovascular activity changes within AMD, the challenge incorporated unique multi-modal datasets. The primary dataset, sourced from Brest, France, was used by participating teams to train and test their models. The final ranking was determined based on performance on this dataset. An auxiliary dataset from Algeria was used post-challenge to evaluate population and device shifts from submitted solutions. Two tasks were involved in the MARIO challenge. The first one was the classification of evolution between two consecutive 2D OCT B-scans. The second one was the prediction of future AMD evolution over three months for patients undergoing anti-vascular endothelial growth factor (VEGF) therapy. Thirty-five teams participated, with the top 12 finalists presenting their methods. This paper outlines the challenge's structure, tasks, data characteristics, and winning methodologies, setting a benchmark for AMD monitoring using OCT, infrared imaging, and clinical data (such as the number of visits, age, gender, etc.). The results of this challenge indicate that artificial intelligence (AI) performs as well as a physician in measuring AMD progression (Task 1) but is not yet able of predicting future evolution (Task 2).

Problem

• Automated detection of AMD progression using OCT images
• Evaluation of AI performance in predicting future AMD evolution
• Benchmarking multi-modal datasets for retinal degeneration assessment

Innovation

• Multi-modal OCT and clinical data integration
• AI-based AMD progression classification
• Future AMD evolution prediction challenge