Researchers develop face recognition tool to predict cancer patient's physiological age

New Delhi, May 9 (PTI) Researchers have developed a face recognition tool that could predict a cancer patient's physiological age, also known as biological age which refers to the health of an individual's body and organs The AI-based model, named 'FaceAge' and described in a paper in The Lancet Digital Health journal, found that patients with cancer appeared about five years older than their chronological age -- years lived since birth.

The patients also suffered worse survival outcomes across multiple cancer types, the researchers said.

"This work demonstrates that a photo like a simple selfie contains important information that could help to inform clinical decision-making and care plans for patients and clinicians," said co-senior author Hugo Aerts, director of the Artificial Intelligence in Medicine program at Mass General Brigham, a US-based healthcare system.

"How old someone looks compared to their chronological age really matters -- individuals with FaceAges that are younger than their chronological ages do significantly better after cancer therapy," Aerts said.

The AI algorithm was trained on nearly 59,000 photos of healthy individuals taken from public datasets and tested on about 6,200 cancer patients. Photographs routinely taken at the start of radiotherapy treatment were used during testing.

"Looking older was correlated with worse overall survival. We found that, on average, patients with cancer looked older than their chronological age," the authors wrote.

The algorithm could also support a physician in gauging a patient's overall health and vitality, thereby helping determine the best course of treatment, the team said.

They asked 10 clinicians and researchers to predict short-term life expectancy from 100 photos of patients receiving palliative (pain-related) radiotherapy -- their predictions were found to be slightly better than a "coin flip".

However, with the support of FaceAge tool, the predictions and assessments improved significantly, the authors said.

"We found that FaceAge can improve physicians' survival predictions in patients with incurable cancer receiving palliative treatments, highlighting the clinical use of the algorithm to support end-of-life decision making," the team wrote.

They said the tool is being tested for predicting one's overall health status, diseases and lifespan and more research is needed before being deployed in a real-world setting. PTI KRS NB