Google’s AI Detects High Blood Pressure Using Smartwatch Sensors

Google Research has developed an AI capable of detecting signs of high blood pressure using a standard smartwatch, eliminating the need for a blood pressure cuff. This technology utilizes two sensors commonly found in modern watches: a pulse sensor and a motion sensor, as detailed in a Google blog post.

The AI was trained on nearly 500,000 hours of sensor data, learning to identify typical pulse and heartbeat patterns without relying on prior diagnoses. It averages these patterns over several days to calculate the likelihood of high blood pressure. Additionally, Google’s health model, SensorFM, also learns from smartwatch data.

In a study involving 196 participants across five research sites in the United States, the smartwatch identified hypertension with a detection rate of 65.8%, compared to 55.3% for traditional cuff measurements.

Google has integrated these findings into the blood pressure trends feature of the Google Health app. However, the method remains a research project and is not intended to replace formal diagnoses. The initiative aims to encourage individuals to seek medical evaluations, as more than half of those with high blood pressure are unaware of their condition, according to Google.

Source: The Decoder