A Review of AI-Wearable Technologies for Public Health Surveillance in the U.S: Challenges and Recommendations
Artificial intelligence (AI)–enhanced wearables generate continuous physiologic streams that can be aggregated for real-time public health surveillance. This review summarizes evidence from U.S. studies on device penetration, analytic performance, and operational value. Surveillance infrastructure has evolved from mailed case cards to cloud dashboards accepting patient-generated data; current machine-learning pipelines transform inertial, photoplethysmographic, and temperature signals into population biomarkers. Lead-time gains of two to six days over laboratory reporting have been documented for influenza, COVID-19, and heart failure admissions. However, uneven adoption, sensor bias, privacy regulation, and limited interoperability constrain scale-up. Interfacing solutions such as FHIR subscriptions and federated analytics are assessed, alongside emerging FDA guidance on AI lifecycle management. Strategic recommendations address standards consolidation, equitable subsidization of devices, algorithm auditability, and workforce training. With these measures, AI wearables could transition from consumer novelties to an integral layer of U.S. public-health intelligence, offering earlier outbreak detection, finer chronic-disease surveillance, and more precise resource allocation.
Paper
Full text
A Review of AI-Wearable Technologies for Public Health Surveillance in the U.S: Challenges and Recommendations
Semantic Scholar · 2025
Abstract
Artificial intelligence (AI)–enhanced wearables generate continuous physiologic streams that can be aggregated for real-time public health surveillance. This review summarizes evidence from U.S. studies on device penetration, analytic performance, and operational value. Surveillance infrastructure has evolved from mailed case cards to cloud dashboards accepting patient-generated data; current machine-learning pipelines transform inertial, photoplethysmographic, and temperature signals into population biomarkers. Lead-time gains of two to six days over laboratory reporting have been documented for influenza, COVID-19, and heart failure admissions. However, uneven adoption, sensor bias, privacy regulation, and limited interoperability constrain scale-up. Interfacing solutions such as FHIR subscriptions and federated analytics are assessed, alongside emerging FDA guidance on AI lifecycle management. Strategic recommendations address standards consolidation, equitable subsidization of devices, algorithm auditability, and workforce training. With these measures, AI wearables could transition from consumer novelties to an integral layer of U.S. public-health intelligence, offering earlier outbreak detection, finer chronic-disease surveillance, and more precise resource allocation.