phat 15(1): e2

Research Article

Cardiac and Respiratory Parameter Estimation Using Head-mounted Motion-sensitive Sensors

Download289 downloads
  • @ARTICLE{10.4108/phat.1.1.e2,
        author={J.  Hernandez and Y.  Li and J. M.  Rehg and R. W.  Picard},
        title={Cardiac and Respiratory Parameter Estimation Using Head-mounted Motion-sensitive Sensors},
        journal={EAI Endorsed Transactions on Pervasive Health and Technology},
        volume={1},
        number={1},
        publisher={ICST},
        journal_a={PHAT},
        year={2015},
        month={5},
        keywords={Ballistocardiography (BCG), blood volume pulse (BVP), heart rate, respiration rate, head-mounted wearable device, gyroscope, accelerometer, camera, daily life monitoring.},
        doi={10.4108/phat.1.1.e2}
    }
    
  • J. Hernandez
    Y. Li
    J. M. Rehg
    R. W. Picard
    Year: 2015
    Cardiac and Respiratory Parameter Estimation Using Head-mounted Motion-sensitive Sensors
    PHAT
    EAI
    DOI: 10.4108/phat.1.1.e2
J. Hernandez1,*, Y. Li2, J. M. Rehg2, R. W. Picard1
  • 1: Media Lab, Massachusetts Institute of Technology, Cambridge, MA 02139 USA
  • 2: Center for Behavior Imaging and the School of Interactive Computing, Georgia Institute of Technology, Atlanta, GA 30332 USA
*Contact email: javierhr@mit.edu

Abstract

This work explores the feasibility of using motion-sensitive sensors embedded in Google Glass, a head-mounted wearable device, to robustly measure physiological signals of the wearer. In particular, we develop new methods to use Glass’s accelerometer, gyroscope, and camera to extract pulse and respiratory waves of 12 participants during a controlled experiment. We show it is possible to achieve a mean absolute error of 0.82 beats per minute (STD: 1.98) for heart rate and 0.6 breaths per minute (STD: 1.19) for respiration rate when considering different observation windows and combinations of sensors. Moreover, we show that a head-mounted gyroscope sensor shows improved performance versus more commonly explored sensors such as accelerometers and demonstrate that a head-mounted camera is a novel and promising method to capture the physiological responses of the wearer. These findings included testing across sitting, supine, and standing postures before and after physical exercise.