An AI tool predicted that patients with cancer who aged rapidly between photos taken during their treatment had lower chances of survival.
FaceAge uses deep learning to determine a person’s biological age from a face photo.
The Mass General Brigham research team behind the tool reported in a new study that estimating biological age from multiple photos taken over time can provide even more information about how well a person with cancer will respond to treatment.
Their results, published in Nature Communications, suggest that the face ageing rate (FAR), which uses photos to measure changes in biological age over time, can serve as a non-invasive biomarker for cancer prognosis.
The new study analysed two photographs from each of 2,279 patients with cancer, taken at different time points during treatment. The researchers found that a higher FAR was significantly associated with a lower survival probability.
Corresponding author Raymond Mak, a radiation oncologist at Mass General Brigham Cancer Institute and a faculty member in the Artificial Intelligence in Medicine (AIM) programme at Mass General Brigham, said: ‘Our study suggests that measuring FaceAge over time may refine personalised treatment planning, improve patient counselling, and help guide the frequency and intensity of follow-up in oncology.’
In this study, the researchers sought to determine what information FaceAge could provide when applied to multiple photos of the same person taken over time.
They examined facial photos from a cohort of patients with various types of cancer who received at least two courses of radiation therapy at Brigham and Women’s Hospital between 2012 and 2023.
The photos were taken as part of the routine clinical workflow for each course of radiation therapy.
FAR was calculated as the change in FaceAge between these two time points, divided by the time interval. The researchers also calculated FaceAge Deviation (FAD), which estimates how biologically old or young a patient appears in a single face photo relative to their chronological age.
Median FAR results indicated that patients' facial ageing outpaced their chronological ageing by 40%. Higher FAR, or accelerated ageing, was associated with lower survival, with the effect strongest when the interval between photos was two years or more.
Additionally, patients with both high FAD and FAR values were significantly more likely to have poorer survival probabilities.
However, FAR was more stable in predicting survival outcomes over longer intervals than FAD – indicating that dynamic measurements might be more reliable than single-timepoint readings.
The authors suggest that integrating FAR with baseline FAD could yield a more nuanced and informative measure of an individual’s evolving health status. Further research is needed to evaluate FaceAge and FAR across more diverse populations.
