Google Unveils Smartphone Camera-based System for Heart Rate Monitoring
Advertisement
The technology, known as Passive Heart Rate Monitoring During Smartphone Use in Everyday Life (PHRM), was detailed in a study published in Nature.
Google has introduced a research system that leverages a smartphone's front-facing camera and artificial intelligence designed to estimate heart rate and resting heart rate during routine device use.
The technology, known as Passive Heart Rate Monitoring During Smartphone Use in Everyday Life (PHRM), was detailed in a study published in Nature.
The system leverages a smartphone's front-facing camera to capture video of the user's face in the seconds after they unlock their phone.
It uses a photoplethysmography technique that detects subtle changes in light reflected from the skin as blood flows through facial blood vessels. These signals are then analysed using AI models to estimate heart rate.
Researchers developed an on-device software pipeline that processes eight-second facial videos using computationally efficient deep-learning AI models. The system generates heart rate estimates along with confidence scores, which are then combined throughout the day using filtering techniques to estimate a user's daily resting heart rate.
As per Google, the PHRM system was developed using more than 350,000 video clips from nearly 700 research participants in laboratory and real-world settings.
The study included participants representing a broad range of skin tones, with performance assessed using the Monk Skin Tone scale.
Google reported that the technology achieved a mean percentage error of less than ten per cent compared to a direct ECG recording.
In addition, for resting heart rate, the company said the system demonstrated performance comparable to wearable fitness trackers, with a mean absolute error of less than five beats per minute.
The tech giant said the system works across a wide range of skin tones.
The company is further planning to optimise camera exposure, trigger additional sampling attempts, improve video stabilisation, and accelerometer-based gating to prioritise opportune at-rest moments.
Stay tuned for more such updates on Digital Health News
