Google health’s team has been fine-tuning the health parameters collected by your smartphone for some time now and evaluating the accuracy of these measurements.
The new work supports the use of consumer-grade smartphones for measuring HR(Heart rate) and RR(Respiratory rate).
One application of these measurements is in fitness and wellness for the general consumer user. Specifically, an elevated resting HR or slower heart rate recovery after exercise has been linked to lower physical fitness and a higher risk of all-cause mortality.
Results of the new study, “Prospective validation of smartphone-based heart rate and respiratory rate measurement algorithms” were published in Nature journal yesterday
The main objective of the study was to see if your smartphone camera could be used to measure heart rate and respiratory rate accurately.
The team tested the accuracy of two computational approaches that determined HR and RR from the videos obtained using a smartphone.
Changes in blood flow through the finger were used to determine HR; similar results were seen for people with different skin tones.
Chest movements were used to determine RR; similar results were seen between people with and without chronic lung conditions.
This study demonstrates that smartphones can be used to measure HR and RR accurately.
There are already smartphone apps that can do this and are available in the play store. However, these apps seldom undergo rigorous clinical validation for accuracy and generalizability to important populations and patient subgroups.
Google’s Smartphone algorithms for Heart Rate and Respiratory Rate
The first algorithm leverages photoplethysmography (PPG) acquired using smartphone cameras for HR measurement.
PPG signals are recorded by placing a finger over the camera lens, and the color changes captured in the video are used to determine the oscillation of blood volume after each heartbeat. 
The second algorithm leverages upper-torso videos obtained via the front-facing smartphone camera to track the physical motion of breathing to measure RR.
Efficacy of smartphone-based health measurements
The team reported the results of two prospective clinical studies validating the performance of smartphone algorithms to estimate HR and RR.
Both algorithms showed high accuracy compared to the reference standard vital sign measurements, with the mean MAPE for HR <2% and the mean MAE for RR <1 breath/min (both significantly below the prespecified targets).
In addition, the HR estimation was robust across the full range of skin tones, and the RR estimation generalized to participants with common chronic respiratory conditions: COPD and asthma.
Because skin tone can be a potential source of bias in medical devices, and the accuracy of PPG-based HR estimation can be affected by melanin’s light-absorbing property, the study enrolled participants with diverse skin tones to validate the robustness of their HR estimation algorithm across skin tones.
