{"slug": "google-research-introduces-sensorfm-a-wearable-health-foundation-model-on-one-of", "title": "Google Research Introduces SensorFM: A Wearable Health Foundation Model Pretrained on One Trillion Minutes of Sensor Data", "summary": "Google Research, Google DeepMind, and university collaborators introduced SensorFM, a wearable health foundation model pretrained on over one trillion minutes of unlabeled sensor data from 5 million participants. The model uses a ViT-1D masked-autoencoder backbone and outperformed feature-engineered baselines on 34 of 35 tasks using frozen embeddings with a PCA-50 linear probe. The research also included an agentic classroom that searched 30,516 prediction heads and a clinician evaluation for a Personal Health Agent.", "body_md": "SensorFM, a wearable health foundation model from Google Research, Google DeepMind, and university collaborators. We walk through its ViT-1D masked-autoencoder backbone, pretrained on more than one trillion minutes of unlabeled sensor signals from 5,000,000 consented participants. We examine the co-scaling results across four model sizes and four data volumes, including the case where capacity outruns data. We show how frozen embeddings plus a PCA-50 linear probe beat feature-engineered baselines on 34 of 35 tasks. We also review the agentic classroom that searched 30,516 prediction heads, and the clinician evaluation grounding a Personal Health Agent.\n\nThe post [Google Research Introduces SensorFM: A Wearable Health Foundation Model Pretrained on One Trillion Minutes of Sensor Data](https://www.marktechpost.com/2026/07/10/google-research-introduces-sensorfm-a-wearable-health-foundation-model-pretrained-on-one-trillion-minutes-of-sensor-data/) appeared first on [MarkTechPost](https://www.marktechpost.com).", "url": "https://wpnews.pro/news/google-research-introduces-sensorfm-a-wearable-health-foundation-model-on-one-of", "canonical_source": "https://www.marktechpost.com/2026/07/10/google-research-introduces-sensorfm-a-wearable-health-foundation-model-pretrained-on-one-trillion-minutes-of-sensor-data/", "published_at": "2026-07-10 08:52:04+00:00", "updated_at": "2026-07-10 09:16:32.181331+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-research", "ai-products", "ai-infrastructure"], "entities": ["Google Research", "Google DeepMind", "SensorFM", "MarkTechPost"], "alternates": {"html": "https://wpnews.pro/news/google-research-introduces-sensorfm-a-wearable-health-foundation-model-on-one-of", "markdown": "https://wpnews.pro/news/google-research-introduces-sensorfm-a-wearable-health-foundation-model-on-one-of.md", "text": "https://wpnews.pro/news/google-research-introduces-sensorfm-a-wearable-health-foundation-model-on-one-of.txt", "jsonld": "https://wpnews.pro/news/google-research-introduces-sensorfm-a-wearable-health-foundation-model-on-one-of.jsonld"}}