Using step counts and heart rate readings from Google’s Fitbit devices, a machine learning model accurately predicted the risk of hospitalization, a study presented at Heart Rhythm 2025 found. The study used data from participants in the National Institutes of Health’s All of Us research program.

This year saw continued advances in smartwatches as they increasingly move from wellness assistants to medical monitors. Nowhere has that been clearer than in cardiovascular health, where multiple wearables now allow users to quickly detect atrial fibrillation, a notoriously shy condition previously only detectable in early stages by chance in a physician’s office or by wearing a cumbersome Holter monitor for 24 to 48 hours.

Fitbit Inc. and Apple Inc. picked up the pace in their race to put health monitoring apps on wrists everywhere with Fitbit gaining 510(k) clearance from the U.S. FDA for its new ECG app for the Fitbit Sense and Apple revealing the blood oxygen sensor built into its Series 6 watch this week. The new apps join a growing array of technological advances that permit wearable devices to track and record a range of health metrics.

The case of Fitbit Inc. v. Valencell Inc. has seen the usual number of twists and turns as it wended its way through Article III courts and an inter partes review (IPR), but a U.S. Supreme Court case that requires that all claims in an IPR be reviewed was decided during the Fitbit IPR.

Wearables giant Fitbit Inc. has kicked off its first large-scale, virtual study to evaluate how well its wrist-worn technology can spot episodes of irregular heart rhythm that might indicate atrial fibrillation (AF). The study, part of a broader plan to equip Fitbit devices with new tools to speed the detection of a range of conditions, is open to people in the U.S. ages 22 and older who own a Fitbit that tracks heart rate.