AI Tool Uses Face Photos to Estimate Cancer Survival — SkimNews
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- FaceSurvival AI tool estimates patient survival by analyzing facial features near the eyes and bridge of the nose, trained on 35,000 cancer patient photos and millions of healthy faces.
- Dr Raymond Mak leads the research at Mass General Brigham, stating the goal is to make face-based health assessments more quantitative and scalable to support personalized treatment decisions.
- FaceAge AI tool, also under development, identifies nasolabial folds, marionette lines, forehead, temples, and cheeks as key indicators of biological aging, with consistency across lighting, expression, and minor plastic surgery.
- Researchers stress the need for more diverse facial data across ethnic groups to improve the tools' accuracy and generalizability, as both systems remain in early development.
- Prof Miroslaw Bober cautions that while the AI is a promising proof of concept, it is not ready for clinical use and must be tested against physician judgment across varied hospital settings.
Why it matters: The tools could help oncologists assess frailty and treatment suitability in aging populations, but without broader validation across ethnicities and clinical environments, their real-world impact remains limited. The pending patent and public data collection signal a push toward scalability, though expert skepticism underscores the gap between algorithmic prediction and medical explanation.
Ask SkimNews
