Does the smartwatch need a blood pressure cuff to calibrate?

No. The AI relies on two sensors already built into modern watches — a pulse sensor and a motion sensor — and needs no cuff for calibration.

Can I use this to diagnose high blood pressure now?

No. Google says the method is still a research project and doesn't replace a diagnosis; the findings have been folded into the blood pressure trends feature in the Google Health app, with the goal of nudging people to get checked.

How was the AI trained without diagnoses?

A neural network learned from nearly 500,000 hours of sensor data by comparing short segments and figuring out which belonged to the same person, picking up typical pulse and heartbeat patterns without any diagnoses to guide it.

AI in HealthcareTHE DECODER1d agoNov 7, 2026, 19:01 JST

Google AI spots high blood pressure from a watch, 65.8% hit rate

Send on LINE

3 Key Points

What happened
In a study of 196 people across five US research sites, Google Research's AI flagged 65.8 percent of those with high blood pressure, versus 55.3 percent for the standard in-office cuff measurement.
Why it matters
The system needs no cuff for calibration and uses only the pulse and motion sensors already in modern smartwatches, so screening could reach people who never get measured — though Google says it is still a research project and does not replace a diagnosis.

WHO IT HITSPrimary-care and cardiology teams, plus makers of consumer wearables, may see a path to cuff-free hypertension screening; Google stresses the method is still research and not a diagnostic replacement.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Get these free every morning

Context & Analysis

Google's approach sidesteps a long-standing constraint in wearable health: cuff-based blood pressure measurement is bulky and intermittent, and calibrating a watch usually requires one. The research team instead let a neural network learn from nearly 500,000 hours of sensor data by matching short segments to the same person, picking up pulse and heartbeat patterns without labeled diagnoses.

The system then averages those patterns over several days and produces a probability for high blood pressure. In the 196-person, five-site US study, that translated into catching 65.8 percent of hypertension cases, compared with 55.3 percent for the standard in-office measurement.

Google frames the work as a nudge rather than a diagnosis, noting that more than half of people with high blood pressure don't know they have it. The findings are already reflected in the blood pressure trends feature of the Google Health app, but the method remains a research project. Google's health model SensorFM, which also learns from smartwatch data, is part of the broader effort.

FAQ

Does the smartwatch need a blood pressure cuff to calibrate?: No. The AI relies on two sensors already built into modern watches — a pulse sensor and a motion sensor — and needs no cuff for calibration.
Can I use this to diagnose high blood pressure now?: No. Google says the method is still a research project and doesn't replace a diagnosis; the findings have been folded into the blood pressure trends feature in the Google Health app, with the goal of nudging people to get checked.
How was the AI trained without diagnoses?: A neural network learned from nearly 500,000 hours of sensor data by comparing short segments and figuring out which belonged to the same person, picking up typical pulse and heartbeat patterns without any diagnoses to guide it.

THE DECODERRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.