This AI tool can predict side-effects in breast cancer patients
A team of medical professionals and researchers has developed an artificial intelligence (AI) tool that can predict risks of side-effects in breast cancer patients after treatment. Currently undergoing trials in the UK, France, and the Netherlands, the technology aims to deliver better personalized healthcare. Tim Rattay, a consultant breast surgeon at the University of Leicester, believes this tool could provide valuable insights to both doctors and patients regarding the side-effects associated with surgery and radiotherapy for breast cancer.
Training AI to foresee lymphoedema post-treatment
The AI tool underwent training using data from 6,361 breast cancer patients. Its purpose is to predict lymphoedema (painful swelling of the arm) up to three years post surgery and radiotherapy. If patients are found to be at a higher risk of arm swelling, alternative treatments could be offered to them. "The final, best-performing model makes predictions using 32 different patient and treatment features," said Dr Guido Bologna, part of the team who developed the AI tool.
AI's predictive accuracy and future applications
The AI tool showcased a predictive accuracy rate of 73.4%, accurately forecasting lymphoedema in an average of 81.6% cases. It also correctly identified patients who wouldn't develop it in an average of 72.9% cases. Rattay suggested that high-risk patients could benefit from additional supportive measures like wearing an arm compression sleeve during treatment. The research team plans to enroll 780 patients for a clinical trial called the "Pre-Act" project with aspirations to broaden the tool's predictive capabilities for other side-effects.
