Observed Signal · Aug 20, 2026 · Technical Release · Source: t3n · Impact: 4/5 · Sentiment: Neutral
Google Photoscan AI Estimates Body Fat from Phone Photos
Google Research announced Photoscan AI, a method that estimates body fat composition from simple 2D smartphone photos. The model was trained on clinical DXA X-ray data and MRI body scans, then fine-tuned with real smartphone images. According to Google Research, Photoscan AI achieves higher accuracy than bioelectrical impedance analysis (BIA) used in many wearables and approaches the precision of clinical DXA imaging. The team also evaluated whether Photoscan outputs could help predict insulin resistance and cardiometabolic risk, positioning the approach as a scalable tool for population-level body-composition and metabolic research.
Technical research release from a major platform (Google Research) introducing a scalable smartphone-based AI method for body-composition estimation; significant for AI capabilities and potential data/privacy implications.
Track Google Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- Google Research developed Photoscan AI to estimate body fat composition from smartphone photos.
- Photoscan was trained on DXA X-ray data and combined with MRI body data, then fine-tuned using real smartphone images.
- Google Research reports Photoscan achieves higher accuracy than BIA-based wearable sensors and approaches DXA accuracy.
- The researchers evaluated Photoscan's potential to help detect insulin resistance and cardiometabolic risk.
Connected Companies & Entities
4 Entities mapped“The Research team has developed Photoscan AI, a new approach that aims to precisely determine body fat percentage based on simple smartphone...”
“The article was published on t3n.de (t3n – digital pioneers)....”
“The article notes that external content from TargetVideo GmbH complements the editorial offering on t3n.de....”
“Photo credit in the article: beauty-box / Shutterstock....”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Google Health Guardian: Pixel Watch predicts diabetes risk
Google is rolling out its first 'Health Guardian' features for Pixel Watches (3rd to 5th generation) and the Fitbit Air band. Using AI analysis of sensor data, the Google Health app will now provide insights into blood pressure trends, insulin resistance tendencies, and sleep breathing quality. The insulin resistance analysis aims to detect gradual metabolic changes without blood draws, potentially flagging early signs of Type 2 diabetes. Blood pressure is estimated passively from pulse and movement patterns. The features are free on Pixel Watch 3-5, while Fitbit Air requires a Health Premium subscription (€9/month). In select European countries, the Pixel Watch 4 and 5 can also detect respiratory emergencies and automatically call emergency services. Google Research published a study in Nature.
Vision AI for Precise Food Volume and Calorie Estimation
This technical guide demonstrates a multimodal Vision AI pipeline for estimating food volume, weight and nutritional values from smartphone photos. The author combines the Segment Anything Model (SAM) for pixel-accurate segmentation, OpenCV for preprocessing, and the GPT-4o multimodal API to reason about depth, density and convert mask area into volume/weight estimates. The tutorial includes code examples for obtaining SAM masks, sending image+mask data to GPT-4o (expecting structured JSON), and wrapping the workflow in a FastAPI / Pydantic endpoint that returns calories, macros and a confidence score. Prerequisites listed include Python 3.10+, SAM weights (sam_vit_h_4b8939.pth), an OpenAI API key and FastAPI. The post also discusses production challenges (occlusion, lighting, GPU memory, concurrency) and links to additional scaling resources on the WellAlly blog.
Researcher Tests AI Diet Apps, Finds Accuracy and Privacy Issues
A nutrition expert at the University of Virginia, Sibylle Kranz, tested popular AI-powered diet apps and found that while they make food logging easier and more accessible, they raise significant concerns about data privacy and accuracy. Users often share extensive personal data without fully understanding how it's used, posing a privacy risk. Additionally, recent studies and Kranz's own tests show that AI systems sometimes underestimate calorie counts or provide incorrect dietary advice, requiring users to correct the information. Despite these drawbacks, Kranz views AI as a valuable supplement to professional nutritional counseling, not a replacement, if used under proper guidance.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.