Alibaba DAMO AI model detects 15 missed liver cancer cases in CT scans

August 26, 2026 at 05:35 PM2 min

This digest was compiled by AI from multiple sources — links to the originals are below.

An AI model developed by Alibaba's DAMO Academy identified 15 liver lesions initially missed by doctors on CT scans. The findings were published in Nature Medicine and led to treatment adjustments for the affected patients. The model also reduced scan analysis time by 27% and increased malignant tumor detection sensitivity by 11.5%.

Key Facts

The DAMO LiON model analyzed CT scans of over 10,000 patients during a two-month clinical study.
The AI system detected 15 metastatic lesions that had been missed by physicians, leading to changes in treatment plans.
Using the AI reduced image analysis time by 27% and increased sensitivity for detecting malignant tumors by 11.5%.
The model was developed by Alibaba's DAMO Academy in collaboration with Shengjing Hospital of China Medical University and other institutions.
DAMO Academy is also developing AI models for pancreatic, gastric, and colorectal cancer screening.

Clinical Study Results

The DAMO LiON model was tested in a two-month clinical study involving CT scans from more than 10,000 patients. It identified 15 metastatic liver lesions that radiologists had initially missed. The additional diagnoses led doctors to adjust treatment plans for those patients. The study results were published in Nature Medicine.

Technical Performance

The model analyzes contrast-enhanced CT images and can detect both primary malignant liver tumors and small metastases. Such metastases are difficult to diagnose because they may be about one centimeter in size, weakly differ from surrounding tissue, or appear in atypical anatomical locations. In experiments, use of the AI reduced image analysis time by 27%. Sensitivity for detecting malignant tumors increased by 11.5% when the AI was used.

Developer and Future Applications

DAMO LiON was developed by Alibaba's DAMO Academy together with Shengjing Hospital of China Medical University and other institutions. DAMO Academy is expanding its AI diagnostic tools to other cancers, including pancreatic, gastric, and colorectal cancer screening models. These systems are viewed primarily as physician support tools that speed up medical image analysis and reduce the risk of missing subtle pathologies.