Med-RADIO: Reducing All Medical Domains Into One via Multi-Teacher Distillation

2026-09-29

AI Summary

This study addresses the limitations of medical generalist models, specifically their restricted training data scale and insufficient depth of domain knowledge, by proposing a multi-teacher knowledge distillation framework for constructing vision foundation models. The method compresses multi-source expert knowledge into a unified model via a modality-aligned distillation stream, incorporating a balanced loss mechanism and a data reorganization strategy to effectively integrate complementary cross-modal expertise. Experimental results demonstrate that the proposed model significantly outperforms existing strong generalist models across benchmarks spanning five medical imaging modalities, achieving performance levels approaching the upper bounds of individual single-modality experts. This work establishes a novel paradigm for developing high-performance, universal medical AI systems.

Abstract

The rapid expansion of large-scale medical datasets and computational resources has driven significant progress in medical foundation models. Given the inherent heterogeneity of medical imaging modalities, current research mainly follows two paths: specialized models optimized for specific modalities, and generalist models designed to handle multiple modalities. However, medical generalist models suffer from both insufficient training data scale relative to natural image generalists and inadequate domain-specific depth relative to medical specialists. Empirically, generalist models establish a cross-modality performance baseline, while specialists define the performance ceiling within their respective domains. To elevate this baseline toward these ceilings, we propose Med-RADIO, a medical multi-teacher distillation framework that Reduces All Domains Into One by compressing complementary expertise from multiple domain-specific teachers into a unified medical vision foundation model. Our method curates both generalist and specialist teachers, allocates modality-aligned distillation streams to reorganize generalist pretraining data so it matches specialist domains, and uses a balanced loss to prevent any single teacher from dominating the distillation process. On internal and external classification benchmarks spanning five modalities, Med-RADIO improves over strong medical generalists under linear probing and remains competitive with representative specialists on most evaluated modalities. Code is available at https://github.com/CAIR-HKISI/Med-RADIO.

Problem

• medical foundation models
• generalist models
• specialist models
• multi-modal medical imaging
• knowledge distillation

Innovation

• Multi-Teacher Distillation
• Medical Foundation Model
• Modality-Aligned Distillation
• Balanced Loss
• Generalist-Specialist Integration