Chinese scientists develop AI tool to predict liver cancer recurrence

HEFEI, March 16 (Xinhua) -- A Chinese research team has developed an AI tool that predicts liver cancer recurrence risk with 82.2 percent accuracy, according to a study recently published in the journal Nature.

Liver cancer, the third leading cause of cancer-related deaths worldwide, has a postoperative recurrence rate as high as 70 percent. Accurately predicting recurrence was a critical challenge.

Researchers from the University of Science and Technology of China, led by Sun Cheng, have developed a scoring system named TIMES, which quantifies spatial distribution patterns of immune cells within the tumor microenvironment to assess relapse likelihood. The system is the world's first liver cancer recurrence prediction tool integrating spatial immune data.

The study demonstrated that immune cell spatial organization, not just their quantity, determines clinical outcomes. By combining spatial transcriptomics, proteomics, multispectral immunohistochemistry, and AI-driven spatial analysis, the team established a novel method for tumor microenvironment assessment. The system was trained using liver cancer tissue samples from 61 patients.

The researchers opened a free online version of TIMES, allowing global users to upload pathological staining images for instant risk evaluation.

The team aims to provide a revolutionary decision-making tool to help doctors optimize personalized treatments, especially in resource-limited settings, Sun said, adding that they are actively collaborating with industry partners to standardize clinical applications.