Predicting Diabetic Macular Edema Treatment Responses Using OCT: Dataset and Methods of APTOS Competition

2025-05-09

AI Summary

This study addresses the high inter-patient heterogeneity in response to anti-VEGF therapy among diabetic macular edema (DME) patients and the absence of pre-treatment predictive tools. We propose the first AI-driven paradigm for predicting DME treatment response directly from optical coherence tomography (OCT) images. To support this, we construct and publicly release the first large-scale, multi-task annotated OCT dataset for anti-VEGF response in DME—comprising 2,000 patients—with labels across four clinically relevant endpoints: treatment response, visual acuity change, edema resolution, and retreatment necessity. We further develop a multi-task deep learning model integrating transfer learning and ensemble optimization. Evaluated on an international challenge, our winning solution achieves an AUC of 80.06%, demonstrating the feasibility and clinical potential of pre-treatment OCT-based precision stratification and personalized therapeutic decision-making.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Humans and AI: Planning and Decision Support for Human-Machine Teams

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Diabetic macular edema (DME) significantly contributes to visual impairment in diabetic patients. Treatment responses to intravitreal therapies vary, highlighting the need for patient stratification to predict therapeutic benefits and enable personalized strategies. To our knowledge, this study is the first to explore pre-treatment stratification for predicting DME treatment responses. To advance this research, we organized the 2nd Asia-Pacific Tele-Ophthalmology Society (APTOS) Big Data Competition in 2021. The competition focused on improving predictive accuracy for anti-VEGF therapy responses using ophthalmic OCT images. We provided a dataset containing tens of thousands of OCT images from 2,000 patients with labels across four sub-tasks. This paper details the competition's structure, dataset, leading methods, and evaluation metrics. The competition attracted strong scientific community participation, with 170 teams initially registering and 41 reaching the final round. The top-performing team achieved an AUC of 80.06%, highlighting the potential of AI in personalized DME treatment and clinical decision-making.

Problem

• Predicting DME treatment responses using OCT images
• Stratifying patients for personalized anti-VEGF therapy
• Improving AI accuracy for clinical decision-making in DME

Innovation

• Uses OCT images for DME treatment prediction
• Organized APTOS competition for AI solutions
• Top team achieved 80.06% AUC score