Google Research: ذكاء اصطناعي على الساعة الذكية يكتشف مؤشرات ارتفاع ضغط الدم بدون معايرة بالكفة، بدراسة في NatureGoogle Research: smartwatch AI detects signs of hypertension without cuff calibration, in a Nature study

Google Research has published work on an AI system that detects signs of hypertension from a consumer smartwatch without cuff calibration, according to a company blog post on 5 November. The study appears in Nature and was validated against the clinical gold standard of 24-hour ambulatory blood-pressure monitoring.

The system does not try to output a moment-to-moment blood-pressure number. It aggregates physiological signals over several days to estimate a persistent hypertensive state. It uses the watch's optical PPG sensor for wrist pulse waves and the accelerometer for subtle ballistocardiogram vibrations, plus height as a supporting input.

In a prospective clinical study of 196 participants across five US sites, without known cardiovascular disease and not on blood-pressure medication, initial screening reached about 65.8% sensitivity and 90% specificity. In confirmatory testing after an elevated office reading, sensitivity was about 78.4% and specificity 90.9%. Google says the results are comparable to traditional cuff-based screening paths in the same framework.

The company stresses this is early feasibility research and is not meant to replace clinic cuffs or give a standalone medical diagnosis. The idea is to nudge at-risk people toward formal clinical assessment. The post does not say the feature is a finished product on every watch, or that it has regulatory approval as a diagnostic device.