AI Can Spot Signs of Impending Death Among Cancer Patients in a Single Face Photo
Researchers have developed an AI model capable of predicting mortality risk in cancer patients by analyzing facial photographs. The system identifies subtle signs of biological aging and health deterioration that are often difficult for human doctors to quantify.
Why it matters
This technology could provide a more objective tool for medical professionals to assess patient frailty and determine the suitability of aggressive treatments.
Doctors have long studied their patients’ faces for signs of illness. But a new AI system has taken it a step further.
In a remarkable new study, researchers developed an artificial intelligence system that could identify cancer patients at unusually high risk of dying. Of the 122 patients assigned the highest risk scores, more than half died within just 90 days of having their picture taken. Among 975 patients in the lowest-risk group, only 20 died during the same period.
Our faces may carry subtle signs of deteriorating health that even experienced doctors struggle to quantify. Trained on more than 40 million facial photographs, the AI learned to recognize patterns associated with biological aging and cancer survival.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in