Related Experiment Video
Updated: Dec 21, 2025
10:14
Author Spotlight: Ex Vivo OCT-Based Multimodal Imaging of Human Donor Eyes for Research into Age-Related Macular Degeneration
Published on: May 26, 2023
4.0K
Predicting conversion to wet age-related macular degeneration using deep learning.
Jason Yim1, Reena Chopra1,2, Terry Spitz3
1DeepMind, London, UK.
Nature Medicine
|May 20, 2020
Summary
An artificial intelligence system predicts progression to exudative age-related macular degeneration (exAMD) in the second eye. This AI tool aids in early detection and intervention for visual deterioration.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Exudative age-related macular degeneration (exAMD) significantly causes vision loss.
- Early detection of exAMD in the second eye is crucial for timely intervention.
Purpose of the Study:
- To develop and validate an AI system for predicting exAMD progression in the fellow eye.
- To assess the AI system's performance in identifying high-risk patients.
Main Methods:
- Utilized 3D optical coherence tomography (OCT) images and automatic tissue maps.
- Developed a combined AI model for exAMD progression prediction within a 6-month timeframe.
- Evaluated AI performance using sensitivity and specificity metrics.
Main Results:
- The AI system achieved 80% sensitivity at 55% specificity, and 34% sensitivity at 90% specificity.
- Automatic tissue segmentation identified pre-conversion anatomical changes and high-risk subgroups.
- The AI system outperformed five out of six human experts in prediction accuracy.
Conclusions:
- AI-powered prediction of exAMD progression is feasible and accurate.
- The developed AI system can aid in clinical decision-making for exAMD management.
- AI has the potential to reduce interobserver variability in exAMD diagnosis.
