Common heart rate monitoring in wearable technologies such as smartwatches and wireless earbuds relies, at least in part, on photoplethysmography (PPG), which uses pulses of light to measure blood activity. It usually works well, but it does have limitations. Written by Google scientists New research blog discovered by 9to5Google Yesterday, they announced that they had tried a different approach called audioplethysmography (APG), which uses ultrasound to measure heart rate. And they did it with off-the-shelf active noise canceling (ANC) earphones and a software update.
This trick works by using a small microphone that reflects low-intensity ultrasound signals from inside the ear canal and helps the ANC work to detect disturbances on the skin surface as the blood pumps. Masu. According to the blog, the technology was “resilient” even with poor ear seals, different ear canal sizes, and darker skin tones. That last point is worth noting, as heart rate accuracy for people with naturally dark skin tones or those with tattoos has been an ongoing issue with smartwatches and other wearable devices.
Google researchers also found that the ultrasound approach worked well during music playback, but had problems in noisy environments, stating that “APG signals are sometimes very noisy and can be significantly affected by body movement.” There is a possibility of interference.” However, it turns out that the motion problem can be overcome by using multiple frequencies and extracting the most accurate signal among them.
In addition to commercially available earphones, the researchers also used a purpose-built prototype to test the effects of microphone placement. The field study was conducted with 153 participants. The researchers said the median error rates for heart rate and heart rate variability were 3.21 percent and 2.70 percent, respectively.