Alibaba DAMO Academy Launches Liver Cancer AI Model, Detects 15 Missed Cases in 2 Mths
I'm LongbridgeAI, I can summarize articles.Alibaba DAMO Academy launched the DAMO LiON AI model for liver cancer detection, developed with Shengjing Hospital. In a two-month trial, it identified 15 missed malignant tumors, primarily small metastases. The AI improved diagnostic accuracy over radiologists, reduced review time by 27%, and enhanced sensitivity by 11.5%. Findings were published in Nature Medicine.
Alibaba DAMO Academy announced that it has jointly developed the liver cancer diagnostic AI model DAMO LiON with institutions including Shengjing Hospital of China Medical University. The model can identify tiny cancerous liver lesions through CT imaging and serve as an "AI safety assistant" to help doctors review scans. It can accurately identify primary liver cancer and is particularly effective at detecting easily overlooked liver metastases.
In a two-month real-world prospective clinical trial, the AI model identified 15 malignant tumors that had originally been missed, with most lesions measuring around 1 centimeter. The model helped patients receive timely surgery or drug treatment.
Experimental results showed that the AI model achieved a higher accuracy rate in identifying malignant tumors than radiologists. When doctors used the AI model to assist in scan reviews, review time was reduced by 27%, while sensitivity to malignant tumors improved by 11.5%, effectively reducing missed diagnoses. With AI assistance, junior doctors were able to achieve the performance level of senior doctors.
The related paper was published in the top international academic journal Nature Medicine.