Face-reading AI predicts how old your body really is

Image credit: amyjoyhumphries via Unsplash

Researchers at Mass General Brigham in Boston have developed an AI tool called FaceAge, which predicts a person’s biological age by analyzing a selfie. The algorithm assesses facial features like skin texture and eye shape to estimate how the body is aging relative to actual age. In tests on 6,200 cancer patients using photos taken before radiotherapy, FaceAge found that patients appeared biologically about five years older than their chronological age on average. Notably, those with older-looking faces had a poorer survival outlook, suggesting potential use in prognosis, early health assessment, and personalized wellness planning.

Source

Bontempi, Dennis & Zalay, Osbert & Bitterman, Danielle & Birkbak, Nicolai & Shyr, Derek & Haugg, Fridolin & Qian, Jack & Roberts, Hannah & Perni, Subha & Prudente, Vasco & Pai, Suraj & Dekker, André & Haibe-Kains, Benjamin & Guthier, Christian & Balboni, Tracy & Warren, Laura & Krishan, Monica & Kann, Benjamin & Swanton, Charles & Aerts, Hugo. (2025). FaceAge, a deep learning system to estimate biological age from face photographs to improve prognostication: a model development and validation study. The Lancet Digital Health. 100870. 10.1016/j.landig.2025.03.002.

Additional Reading

https://www.washingtonpost.com/science/2025/05/12/ai-tool-biological-age-faceage/

https://www.massgeneralbrigham.org/en/about/newsroom/press-releases/ai-face-photos-tool-estimate-age-predict-cancer-outcomes

https://news.harvard.edu/gazette/story/2025/05/new-ai-tool-predicts-biological-age-by-looking-at-a-face/

Next
Next

CT scan radiation linked to long term cancer risk