Google researchers have developed an AI system that can estimate body composition using photographs captured with ordinary smartphones. Smartphone Body Analysis could offer users a simpler alternative to traditional methods used for measuring body fat. The experimental technology, called PhotoScan, analyzes multiple two-dimensional images to estimate several aspects of body composition. Meanwhile, researchers trained the system using information gathered from advanced medical imaging techniques.
The team initially used data from dual-energy X-ray absorptiometry scans, commonly known as DXA. These scans helped researchers measure overall body fat and examine how fat distributes across different areas. In addition, the researchers examined ratios involving abdominal, waist, and leg fat during development. They also studied differences between visceral and subcutaneous fat levels.
Furthermore, Google researchers incorporated information from magnetic resonance imaging scans into the development process. They then refined the system with additional photographs captured through smartphone cameras. This process helped the model connect visible physical characteristics with measurements obtained through established medical imaging methods.
According to the research, PhotoScan produced more accurate body fat estimates than bioelectrical impedance technology used by many wearable devices. That technology estimates body composition by sending a small electrical current through the body. However, wearable measurements can vary because hydration, device placement, and other factors influence results.
PhotoScan also estimated certain fat distribution ratios that conventional wearable sensors cannot directly measure. Consequently, the system could provide users with more detailed information than basic weight or body fat percentage readings. However, the technology remains a research project rather than a publicly available consumer product.
Researchers also examined whether the system could identify signals associated with insulin resistance. This potential application could expand the technology beyond body composition measurements. In particular, the researchers explored whether visual information could help identify metabolic risks that traditional weight measurements may overlook.
The findings highlight growing interest in using smartphones for health-related analysis. Modern phones already contain cameras capable of capturing detailed images under suitable conditions. Therefore, researchers continue exploring ways to use those cameras for measurements that previously required specialized equipment.
Nevertheless, the approach raises important privacy questions because users would need to provide photographs of their bodies. Such images could contain highly sensitive personal information, particularly if companies store or process them remotely. Therefore, privacy protections would become an important consideration if Google eventually develops PhotoScan into a consumer service.
The technology also faces another challenge because accurate information does not automatically change people’s behavior. Even when users understand their body composition, they still need to make meaningful changes involving nutrition, exercise, sleep, and other lifestyle factors. Consequently, accurate measurements represent only one part of improving long-term health outcomes.
For now, Google has not announced plans to release PhotoScan as a commercial product. However, the research demonstrates how artificial intelligence could expand smartphone-based health monitoring. Smartphone Body Analysis could eventually become part of broader digital health services if researchers can address privacy, accuracy, and user adoption concerns.
The development also reflects a wider movement toward convenient health tools that reduce reliance on specialized clinical equipment. As a result, smartphone cameras could eventually support new forms of personal health assessment. Still, further research will determine whether these experimental measurements can deliver reliable results across diverse populations and everyday conditions.
Smartphone Body Analysis therefore remains an emerging research concept rather than an established medical measurement. However, Google’s findings suggest that ordinary phone cameras may provide valuable information about body composition. If future testing confirms the results, similar technology could eventually make certain health assessments easier to access.
