Alibaba's Damo Academy Open-Sources Damo Radar, a Medical AI Model Detecting Nearly 150 Abdominal Conditions
Damo Academy, a subsidiary of Alibaba Group, has introduced an open-source AI model named Damo Radar, capable of identifying nearly 150 abdominal diseases, including several forms of cancer, through the analysis of CT scan images. The model scrutinizes contrast-enhanced scans of 18 abdominal organs. After training with clinical data and close to 40,000 assessments, it achieved an average area under the curve of 0.913 for 146 medical conditions. Recent research published in Science highlighted its ability to surpass the diagnostic accuracy of many radiologists. This innovation represents a significant advancement in Alibaba's development of medical AI technology.
Key facts
- Alibaba Group Holding's research arm, Damo Academy, has open-sourced an AI model called Damo Radar.
- The model can identify nearly 150 abdominal conditions, including cancers, from CT scans.
- It analyzes contrast-enhanced CT scans covering 18 abdominal organs.
- The model was trained using CT scans paired with clinical reports.
- In nearly 40,000 real-world examinations, it achieved an average AUC of 0.913 across 146 clinical findings.
- An AUC of 1.0 represents perfect diagnostic accuracy.
- According to a new study published in Science, the model outperformed most radiologists.
- The research team said the training method could eventually be extended to other types of medical imaging.
Entities
Institutions
- Alibaba Group Holding
- Damo Academy
- Science