Chinese scientists use AI model in colorectal cancer treatment

GUANGZHOU, April 3 (Xinhua) -- Chinese researchers have used artificial intelligence (AI) to reduce unnecessary surgical resection in colorectal cancer patients after local excision, the South China University of Technology (SCUT), located in Guangdong Province in south China, said on Wednesday.

The team developed and validated an AI prediction model to assess the risk of recurrence following local excision of colorectal cancer, using pathological images of endoscopically or transanal surgically resected T1 rectal cancer (RC) specimens.

Colorectal cancer is the third most common cancer worldwide, accounting for around 10 percent of all cancer cases. For patients with early-stage colorectal cancer, local resection serves as the primary treatment. However, some high-risk patients still require additional surgeries to prevent recurrence, which not only increases the physical burden but also significantly diminishes their quality of life.

Researchers from the SCUT and Guangdong Provincial People's Hospital developed an artificial neural network model to predict recurrence risk in T1 RC patients -- thereby providing physicians and patients with guidance for post-excision decision-making.

The AI-based prediction model has helped reduce unnecessary additional surgeries by about 34.9 percent in all enrolled patients, compared with the current U.S. guideline in this regard.

The model not only aids physicians in decision-making but also demonstrates the remarkable performance and potential of histopathology image-based AI in predicting tumor behavior, researchers said.

The study has been published in the European Journal of Surgical Oncology.