Google tests heart rate tracking using just your phone camera

Google thinks your next heart-rate monitor might already be sitting in your pocket. In a newly published research, the company has detailed a system that can estimate a person’s heart rate and resting heart rate using only a smartphone’s front-facing camera. The technology could eventually bring wearable-like health tracking to people who don’t own a smartwatch or fitness band.

For years, tracking heart rate has largely been the job of devices strapped to your wrist. Products like Fitbit trackers and smartwatches have made it easy to monitor your cardiovascular health throughout the day. The catch is that not everyone owns one, and many people never will. Smartphones, on the other hand, are nearly everywhere. That’s the opportunity Google is chasing.

The system, called Passive Heart Rate Monitoring (PHRM), uses the phone’s selfie camera to analyze skin changes as blood flows through the body. These tiny fluctuations are invisible to the human eye, but modern cameras and machine learning models can detect them. Google’s implementation captures an 8-second video clip after a user unlocks their phone with face authentication. An on-device AI model then analyzes the footage and estimates the user’s heart rate.

Phone-based heart-rate tracking isn’t entirely new. Some devices have previously used the rear camera and flash, while others have relied on optical fingerprint sensors to estimate heart rate. Google’s approach, however, takes a different route. Instead of requiring users to actively initiate a measurement, the system can estimate heart rate using the selfie camera during normal phone use and build a picture of resting heart rate over time.

The most impressive part of Google’s research is what the company claims works reliably in the real world. To train and validate the system, Google used more than 350,000 video clips collected from nearly 700 participants. Unlike many previous studies, the company intentionally included a diverse range of skin tones, a critical factor because darker skin can make camera-based blood-flow measurements more challenging.

The results appear promising. According to Google, the system achieved heart-rate accuracy that met industry standards across all tested skin-tone groups. When estimating resting heart rate, its performance came within five beats per minute of readings from a Fitbit Charge 6.

They also conducted an unusual real-world study in which participants used their own phones for more than a week while simultaneously wearing ECG equipment and a Fitbit tracker. The smartphone system reportedly continued to perform well outside controlled laboratory conditions.

Despite the encouraging results, Google isn’t pretending the technology is perfect. The system had difficulty consistently capturing measurements for people with darker skin tones, even though the readings it obtained were accurate. Researchers also found that factors such as talking, head movement, and other everyday behaviors could introduce errors.

Privacy is another obvious challenge. While the research relied on explicit consent and encrypted data collection, any future consumer implementation would need strong safeguards. Google suggests that on-device processing and face authentication could help address those concerns.

For now, PHRM remains a research project. Still, the implications are significant. If smartphones can reliably track cardiovascular metrics without requiring extra hardware, heart-health monitoring could become accessible to far more people than today’s smartwatch-centric approach allows.

Given that billions of people already carry smartphones every day, Google’s latest experiment may be one of the clearest examples yet of how the devices we use constantly could evolve into health-monitoring tools.