Ai predicts reappearance of liver cancer post-transplant, free online system

2023-12-04 15:47:49

An artificial intelligence model capable of predicting the reappearance of post-transplant liver cancer. This was carried out by an international study, coordinated by the Department of General and Specialist Surgery of the Sapienza University of Rome, which collected data from approximately 4,000 patients from North America, Europe and Asia to develop a calculator capable of predicting the risk of recurrence of hepatocellular carcinoma. The system, available on a completely free online page, will allow better management and care of patients.

Liver cancer or hepatocarcinoma – recalls a note – represents one of the most common indications for transplantation: in Italy, more than half of the over 1,500 liver transplants carried out every year have hepatocarcinoma as the main cause. In this context, it is of fundamental importance to foresee the possibility that the tumor may recur, on the one hand avoiding subjecting high-risk patients to this operation and on the other improving the care and management of transplant patients. Artificial intelligence was used to develop this calculator, thanks to the help of engineers from the Polytechnic University of Turin. Using this innovative and sophisticated system, the new calculator developed has proven to be more reliable than existing ones, thus increasing the possibility of improving the care of all patients undergoing transplantation for liver cancer.

“The developed score was called Train-Ai (Time-Response-AlphafetoproteIN-Artificial Intelligence)”, explains Quirino Lai of Sapienza. “All the variables that make up the score are easily obtainable before the transplant in order to allow its calculation practically in every part of the world, therefore based on user-friendly parameters. Another important innovation was to develop a calculator based on thousands of patients from all over the world, while the existing scores were based on regional, or at most national, decidedly more limited realities”. The possibility of predicting the risk of developing a recurrence – underlines the note – is of enormous importance for the patient for two reasons: it can allow the identification of a class at unacceptably high risk of recurrence which can therefore be excluded from the transplant itself (frutile transplant due to oncology); it may allow us to study the patient post-transplant with greater attention (closer checks, reduction of immunosuppressive therapy) to prevent relapse in patients at increased risk.