Google Tests AI PhotoScan to Track Insulin Resistance From 2 Smartphone Photos
Updated
Updated · TechRepublic · Aug 20
Google Tests AI PhotoScan to Track Insulin Resistance From 2 Smartphone Photos
3 articles · Updated · TechRepublic · Aug 20
Summary
Google is testing PhotoScan, an AI system that uses front and side smartphone photos to estimate body fat and fat distribution as signals of insulin resistance.
Image-derived body measurements improved prediction versus models using age, sex and BMI, and came close to clinical DXA body-scan models, suggesting phone cameras could approximate specialized equipment.
Pixel phones were used in part of the study, while eligible Pixel Watch and Fitbit devices are also gaining Insulin Resistance Trends based on longer-term wearable data.
Google says a phone scan would add a different layer from CGMs and watches—capturing body-composition changes at specific points rather than glucose, sleep or activity patterns over time.
PhotoScan remains a research project, but Google is positioning it as part of a broader at-home metabolic-health system that could later combine camera, wearable, CGM and blood-test data.