Somerville, Mass.-based Mass General Brigham researchers found that using an AI face analysis tool can help predict biological age and survival outcomes for patients with cancer.
FaceAge, a deep learning algorithm, was trained on 8,851 photos of presumed healthy individuals from public datasets, and 6,196 cancer patients from two centers using photographs routinely taken at the start of radiotherapy treatment, according to a May 8 system news release.
The tool found that patients with cancer appear an average of five years older than their chronological age. People who appeared older had worse overall survival outcomes across multiple cancer types. FaceAge also outperformed clinicians in predicting short-term life expectancies of patients receiving palliative radiotherapy.
“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,” senior co-author Hugo Aerts, PhD, director of the Artificial Intelligence in Medicine program at Mass General Brigham, said in the release. “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.”
The study results were published May 8 in The Lancet Digital Health.
