On-Body AI: Flexible Cardiac Patch Detects Arrhythmias with 99.6% Accuracy — No Cloud Required

Researchers at the University of Chicago have developed a flexible computing patch that detects life-threatening heart arrhythmias with 99.6% accuracy — and does it entirely on the device, with no cloud round-trip.

That architectural choice is the headline. Most connected cardiac monitors stream raw signals to remote servers for analysis, introducing latency that can matter enormously during a cardiac event. By processing data directly on the body, the patch eliminates the delay — and keeps working when connectivity doesn't.

The edge-AI moment in wearables

The patch arrives amid a broader shift toward on-device intelligence in health wearables. Global wearable shipments now exceed 136.5 million units per quarter, and the innovation frontier has moved from counting steps to clinical-grade sensing: hormone-tracking biosensor bands like the one Clair Health just raised $11.6 million to build, vocal-biomarker analysis riding along on telehealth calls, and stretchable patches monitoring cardiac parameters continuously.

From research to reimbursement

Timing may work in the technology's favor. The CPT 2026 code set added codes for remote monitoring and AI diagnostics, and Medicare RPM spending has already passed $500 million annually. If on-body AI monitors can pair clinical accuracy with a clean billing pathway, the route from lab bench to care plan looks shorter than it has ever been.