Home / Health / AI Reads Health From Your Face!
AI Reads Health From Your Face!
21 Oct
Summary
- AI models predict cancer patient's biological age from facial photos.
- Facial analysis could aid doctors in survival predictions and treatment.
- AI models trained on over 40 million facial images for analysis.
Researchers have developed artificial intelligence models capable of predicting a cancer patient's biological age and survival outcomes from a single facial photograph. Published in The Lancet Digital Health, this groundbreaking research suggests that facial analysis may soon become a valuable supplementary tool for clinicians.
The AI models, trained on more than 40 million facial images, examine distinct facial regions to estimate a patient's biological age, reflecting cellular health, and their risk of mortality. These AI assessments echo ancient observations by Hippocrates regarding facial signs of frailty.
One model, FAHR-FaceAge, estimates biological age, finding that individuals with biological ages over 20 years greater than their chronological age had significantly higher 90-day mortality rates. Another, FAHR-FaceSurvival, combines facial features with clinical data to predict survival, with high-risk patients showing a 52.5% mortality rate within 90 days.
While promising, the researchers emphasize these AI models are not intended to replace physician judgment but rather to inform assessments and treatment decisions. Further clinical validation and testing across diverse patient groups are necessary before widespread implementation.
