Alibaba Group’s Damo Academy has open-sourced Damo Radar, an artificial intelligence model that can identify nearly 150 abdominal diseases, including cancers, from CT scans.. Damo Radar analyzes contrast-enhanced CT scans across 18 abdominal organs and is being released for hospitals, researchers and developers to inspect and build on..
In testing on nearly 40,000 real-world examinations, the model achieved an average area under the curve score of 0.913 across 146 clinical findings, according to a study published in Science. An AUC score of 1.0 indicates perfect diagnostic accuracy.. In a study involving 26 radiologists from multiple hospitals, Damo Radar outperformed 23 of them on average.
When radiologists used the model as an assistant, missed diagnoses fell by 10% and diagnosis time dropped by more than 30%.. The model was trained on CT scans paired with clinical reports, using vision-language learning to connect medical images with text descriptions. The research team described Damo Radar as “the world’s first expert-level generalist medical imaging model.”.
The development team said the training method could be extended beyond abdominal CT scans to other forms of medical imaging.. Damo Academy has also developed other medical AI tools, including Damo LiON, a liver cancer diagnostic model that identified 15 malignant tumors missed by physicians in a prospective clinical trial..
Featured image credit. Tags: AlibabaDamo Radar
