Google Research AI detects signs of hypertension using a standard smartwatch without a wrist cuff
The study published by Google Research in Nature shows that its AI uses built-in pulse and motion sensors on smartwatches to detect signs of hypertension without the need for cuff calibration. Neural networks learned pulse and heartbeat patterns unsupervised from nearly 500,000 hours of sensor data. In a study covering 196 participants across five research sites in the US, the AI identified 65.8% of patients with hypertension, outperforming the 55.3% detection rate of standard office measurements.
Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open AIHOT original