أداة ذكاء اصطناعي تتنبأ بفرص نجاة مرضى السرطان من صورة وحدة لوجههمAn AI tool predicts cancer patients' chances of survival from a single photo of their face

Experts at Mass General Brigham, a nonprofit academic healthcare system in Massachusetts, have built an AI tool that can predict a cancer patient's chances of survival from a single photo of their face, The Guardian reported. Details of the study were published in the Lancet Digital Health journal.

The model, FAHR-Face, was trained on data from more than 40 million facial photographs across a range of ethnicities and ages. It produced two tools: FAHR-FaceAge estimates "biological age", meaning the age you look rather than your actual age, and predicts survival, while FAHR-FaceSurvival predicts survival directly from the photo.

Among the key findings: among cancer patients treated with radiotherapy in the US, those who looked five or more years older than their actual age had a 21% higher risk of dying than the rest of the group over an average four-year follow-up, while those who looked five or more years younger had a 17% lower risk. The researchers also found that habits such as smoking were linked to looking biologically older.

The researchers say the goal is to help doctors assess how well a patient can handle intensive treatments such as radiation and surgery, not to replace a doctor's judgement. Doctors cautioned that more research is needed before the tool can be widely used.