Wearable technology, such as smart watches and wireless earbuds, typically relies on photoplethysmography (PPG) for heart rate monitoring. PPG uses light pulses to measure blood activity, but it has its limitations. According to Google scientists, a new approach called audioplethysmography (APG) that uses ultrasound to measure heart rate has been tested and implemented on off-the-shelf active noise-canceling (ANC) earbuds through a software update. This innovative technique provides more accurate results compared to PPG and addresses issues with accuracy for individuals with darker skin tones or tattoos when wearing smartwatches and other wearables. The ultrasound signal is bounced off the inside of the ear canal, and the small microphone used for ANC detects skin surface perturbations caused by blood flow.
The researchers also found that the ultrasound approach remained effective even when music was playing. However, it had some limitations in noisy environments and could be affected by body motion. To overcome these challenges, the researchers used multiple frequencies and identified the most accurate signal among them. Additionally, purpose-built prototypes were utilized alongside commercially available earbuds to test the impact of microphone placement. A field study involving 153 participants resulted in a median error rate of 3.21 percent for heart rate and 2.70 percent for heart rate variability.