Health monitoring is about to take an unprecedented leap, moving from wearables directly to smartphones. Google Research has introduced an innovative system called PHRM (Passive Heart Rate Monitoring), capable of measuring heart rate and resting heart rate in a completely passive way through the phone’s front camera.
Heart rate from the smartphone: what it is and how Google’s PHRM technology works
Resting heart rate is a fundamental signal for long-term cardiovascular health, and its monitoring can help predict significant health risks. While wearables have already paved the way in this field, the goal of Google’s new technology is democratize access to health monitoring, reaching billions of people who own a smartphone but do not necessarily have a dedicated device.
The PHRM system operates quietly in the background by leveraging the brief moments following the device’s facial unlock. Using deep learning algorithms and remote photoplethysmography (rPPG) techniques, the software analyzes small color changes in the user’s face, invisible to the human eye, to estimate beats per minute with an extremely low margin of error.
During tests conducted in real-world scenarios, the system demonstrated accuracy comparable to industry standards, maintaining a mean absolute error of less than 5 beats per minute compared to wearables.
An inclusive approach for every skin tone
One of the most innovative aspects of the research concerns design inclusivity, addressing a long-standing issue of light-based technologies that often struggle to provide accurate data for darker skin tones due to melanin.
To overcome this obstacle, Google’s team developed PHRM using the largest and most diverse video database ever collected, ensuring that performance was accurate and reliable for the full spectrum of skin tones based on the Monk scale.
Prevention becomes everyday and accessible
The practical implications of this technology are decidedly profound, as they transform a common gesture like unlocking your own phone into an opportunity for continuous preventive screening.
The passively collected data not only reflect heart rate but show a direct correlation with other risk factors such as body mass index and overall cardiovascular fitness level.
There are still some open challenges related to video stability and subject movement, but in the future—thanks to PHRM technology—our everyday devices could become true active sensors for wellbeing, thus eliminating the need for additional accessories (not welcomed by everyone).
