Google has published research on a new method for measuring heart rate during everyday smartphone use, requiring no active participation from the user. By tapping the front-facing camera and deep learning, the technology — known as Passive Heart Rate Monitoring (PHRM) — could bring wearable-level health insights to the roughly five billion people who already own a capable smartphone.
Developing the Monitoring Model
Resting heart rate is a well-established biomarker of cardiovascular health and long-term risk, with elevated resting rates linked to adverse cardiovascular events. Google previously demonstrated in 2022 that smartphones could measure heart rate when a user places a finger over the camera; this new research extends that into passive, background monitoring during normal phone use.
The system uses the front-facing camera to capture facial video, with deep learning estimating heart rate at a mean absolute percentage error of less than 10% — meeting industry accuracy standards across all skin tones tested.
"As the only rPPG (remote photoplethysmography) method to meet heart rate accuracy standards for people of all skin tones – even in unpredictable real-world conditions – it sets a new standard for the field. It also represents the first use of rPPG to estimate daily RHR, achieving wearable-level accuracy across all skin tones."— Eric S Teasley, Product Manager, and Ming-Zer Poh, Staff Research Scientist, Google Research
How the Technology Works
PHRM works through remote photoplethysmography (rPPG), sensing subtle fluctuations in how light interacts with skin each time blood pulses through it. On-device software processes short clips of facial video, with temporal shift convolutional neural networks predicting heart rate from the processed footage.
A key challenge the team addressed was representation: previous rPPG studies underrepresented people with dark skin tones, since melanin makes pulse signals harder for cameras to detect. To correct this, Google built its dataset using the Monk Skin Tone Scale, ensuring light skin tones made up at least 25% of the data, medium tones at least 25%, and dark skin tones at least 33% — what the company calls the largest and most diverse rPPG study to date.
What's Next for PHRM
Google has identified areas for further refinement, including optimising camera exposure and improving accuracy during excessive head movement. The company also plans to make its data and modelling resources available to qualified researchers, a move intended to accelerate further work in the field.
With billions of compatible devices already in pockets worldwide, passive heart rate monitoring points to a future where cardiovascular health tracking is built quietly into the smartphones people already use — no wearable purchase required.