A new AI tool can predict cancer survival from a single patient’s selfie, research suggests AI (artificial intelligence)

Doctors have developed an artificial intelligence tool that can predict a cancer patient’s chances of survival from a single photo of their face.
The AI system was custom-built by experts at Mass General Brigham, a non-profit, integrated health care system in Massachusetts, to estimate biological age and directly predict survival outcomes in patients diagnosed with cancer.
It’s one of the most comprehensive demonstrations to date of how AI analysis of a selfie can help assess a patient’s health and guide their medical care, and was trained on data from more than 40 million facial images across a range of races and ages.
Details of the breakthrough have been published in the Lancet Digital Health journal.
The system can estimate a person’s age – the age a person looks at, rather than their actual, chronological age – and predict cancer patients’ chances of survival from a standard photo of their face, according to the results of a study that tested the model.
Doctors use technology to create two complementary tools. FAHR-FaceAge estimates a patient’s biological age, and predicts cancer survival. FAHR-FaceSurvival predicts survival from the face image directly, rather than life in biological years.
Although FAHR-FaceAge and FAHR-FaceSurvival are derived from the same model, they capture different and complementary information about a patient’s health.
Each algorithm also relied on different parts of the face to make successive predictions of age and risk of death. FAHR-FaceAge focuses on rays that run from the sides of the nose down to the corners of the mouth, the forehead area just above the eye sockets, and the temples. FAHR-FaceSurvival, on the other hand, examines the region below the eyes and around the bridge of the nose.
“FAHR-FaceSurvival advances the field of facial health recognition AI a step further by studying survival risk directly from facial images of cancer patients,” said Hugo Aerts, senior author of the study. “To our knowledge, this is the first indication that an AI model can directly predict cancer survival with a simple image.”
Among cancer patients treated with radiotherapy in the US, people, according to FAHR-FaceAge, who looked five or more years older than their actual age had a 21% higher risk of dying during an average of four years of follow-up than the rest of the group. Patients who looked five years younger or older had a 17% lower risk.
“An important technological development is that FAHR-Face learns a general representation of the face from more than 40 million images, which can be converted into different health outcomes such as biological age and life expectancy,” said Aerts. “This model-based approach gives us a more efficient way to develop more biomarkers of facial health than training each model from scratch.”
The results of the study suggest that the average biological age of people is closely related to their physical health, which can indicate their ability to survive certain treatments.
The breakthrough could pave the way for AI facial analysis to assist doctors in making difficult decisions about the best way forward in treating their patients.
The tool can help better predict the risk of death or complications after major surgery or other cancer treatment.
Dr. Raymond Mak, a radiation oncologist at Mass General Brigham, said: “As people age, doctors need better ways to measure immune function and frailty before recommending more aggressive treatments such as radiation, surgery or systemic therapy.
“Our goal is to make facial imaging-based health screenings more versatile, consistent and scalable to support personalized treatment decisions instead of replacing the doctor.”
Researchers have found lifestyle factors to be associated with facial aging. For example, heavy alcohol use, drug use and smoking were linked to looking biologically older in AI facial image analysis.
“One of the surprising findings is how consistent FAHR-FaceAge is across a wide range of real-world images, including pose, lighting, facial expressions and makeup,” said Mak. “It was stable and of all the plastic surgery procedures tested, there was little change seen after rhinoplasty.”
Doctors cautioned that while single-face images could provide a simple, cost-effective way to support personalized treatment decisions, more research was needed before the tool could be widely used.
Source link



