Alibaba Group Holding’s research arm, Damo Academy, has officially open-sourced its new vision-language AI model, Damo Radar. The technology is specifically engineered to process contrast-enhanced computed tomography (CT) scans, covering 18 distinct abdominal organs. According to the company, the model is capable of identifying nearly 150 different medical conditions, ranging from common abnormalities to malignant cancers.

This release represents a significant expansion of Alibaba’s ongoing initiatives in the medical artificial intelligence sector. By making the model open-source, the company allows researchers and medical professionals to access the underlying technology, potentially accelerating the development of diagnostic tools that can assist radiologists in identifying diseases more efficiently. The model functions by integrating vision and language processing to interpret complex medical imaging data, aiming to provide a comprehensive overview of a patient's abdominal health based on scan results.

While the announcement highlights the potential for improved diagnostic speed and accuracy, the practical implementation of such AI tools in clinical settings remains subject to regulatory oversight and validation by medical institutions. Alibaba’s move follows a broader industry trend of major technology firms investing in specialized AI applications for healthcare, though the specific performance metrics and clinical trial data for Damo Radar have not been fully detailed in this initial open-source release.