Alibaba DAMO Academy's DAMO RADAR Medical Imaging Model Published in Science and Open-Sourced

Alibaba DAMO Academy, together with the First Affiliated Hospital of Zhejiang University and other global hospitals, published its general abdominal CT diagnostic model DAMO RADAR in Science and open-sourced it, calling it the world's first expert-level general medical imaging model. Weights are available on Hugging Face for research and application use.

Confirmed

• Covers 18 abdominal organs and detects 146 pathologies (some posts describe it as nearly 150 abdominal diseases and abnormalities), including malignant tumors
• Analyzes contrast-enhanced CT, abandoning the one-model-per-disease paradigm via vision-language learning that aligns CT images with radiology reports at the anatomical-structure level, learning directly from clinical reports without manual annotation
• Per victormustar, trained on over 400,000 contrast-enhanced abdominal CT exams and 15 million anatomy-aware image-text pairs
• Across nearly 40,000 real-world exams, performance on 146 clinical findings matched or exceeded most radiologists; per DrSingularity, average AUC was 0.913; QbitAI reported internal validation involving nearly 30,000 cases, with the paper as the authoritative source
• The model may be extended to other imaging modalities in the future

Why it matters

• It validates the feasibility of general-purpose imaging models over fragmented one-model-per-disease approaches
• Broad disease coverage, expert-level performance, and open weights make it practically valuable for clinical assistance and further research
• giveen sees it as a positive example of AI in healthcare; DrSingularity argues the AI race is shifting toward medical transformation and that AI may conquer all diseases in the 2030s (personal opinions)

2026-09-18 ~ 2026-09-20 · 8 related posts

• [source] DAMO's Abdominal CT AI Hits Science: One Model Detects 146 Diseases, Beats 23 of 26 Radiologists — 量子位 · 2026-09-18
• [source] Alibaba DAMO Academy's DAMO RADAR hits Science: expert-level universal abdominal CT model, open-sourced — 机器之心 · 2026-09-18
• Alibaba Open-Sources Medical AI Model Detecting Cancer and Nearly 150 Conditions — yogthos · 2026-09-19
• Alibaba open-sources medical AI model that detects cancer and nearly 150 conditions — giveen · 2026-09-19
• AI race is turning into a race to transform medicine, argues Dr_Singularity — Dr_Singularity · 2026-09-20
• [source] Alibaba open-sources RADAR, an expert-level abdominal CT AI trained on 400K scans — victormustar · 2026-09-20

2 near-duplicate retellings: AIFlow_ML · Dr_Singularity