OPERA: Offline Policy-guided Expert Routing and Adaptation for Universal Biomedical Image Analysis
One AI model that works across different hospitals without retraining
Researchers created OPERA, a system that combines multiple specialized AI models to analyze medical images from different scanners and hospitals without needing to retrain on new data. By learning how to route each image to the best-suited expert model during a short calibration phase, then adapting slightly at test time, OPERA maintained high accuracy across 9 different medical imaging datasets—including X-rays, CT scans, and MRI images—without the expensive cycle of retraining for each new setting.
Medical AI systems often fail when deployed to new hospitals or scanners because patient populations and imaging protocols differ, forcing expensive and time-consuming retraining. OPERA eliminates this bottleneck: hospitals can deploy the same system across different equipment and patient groups without collecting new labeled data or sharing sensitive patient information. This makes it practical to build robust diagnostic AI that actually works in the real world, where retraining is rarely an option.