Continued developments of sensor technology including hardware miniaturization and increased sensitivity have enabled the development of less intrusive methods to monitor physiological parameters during daily life. In this work, we present methods to recover cardiac and respiratory parameters using accelerometer and gyroscope sensors on the wrist. We demonstrate accurate measurements in a controlled laboratory study where participants (n = 12) held three different positions (standing up, sitting down and lying down) under relaxed and aroused conditions. In particular, we show it is possible to achieve a mean absolute error of 1.27 beats per minute (STD: 3.37) for heart rate and 0.38 breaths per minute (STD: 1.19) for breathing rate when comparing performance with FDA-cleared sensors. Furthermore, we show comparable performance with a state-of-the-art wrist-worn heart rate monitor, and when monitoring heart rate of three individuals during two consecutive nights of in-situ sleep measurements.
CITATION STYLE
Hernandez, J., McDuff, D., & Picard, R. W. (2015). Biowatch: Estimation of heart and breathing rates from wrist motions. In Proceedings of the 2015 9th International Conference on Pervasive Computing Technologies for Healthcare, PervasiveHealth 2015 (pp. 169–176). Institute of Electrical and Electronics Engineers Inc. https://doi.org/10.4108/icst.pervasivehealth.2015.259064
Mendeley helps you to discover research relevant for your work.