4th International Conference on Wireless Mobile Communication and Healthcare - "Transforming healthcare through innovations in mobile and wireless technologies"

Research Article

Performance of energy expenditure assessment using a chest-worn wireless patch sensor

Download549 downloads
  • @INPROCEEDINGS{10.4108/icst.mobihealth.2014.257292,
        author={Nandakumar Selvaraj and Toai Doan},
        title={Performance of energy expenditure assessment using a chest-worn wireless patch sensor},
        proceedings={4th International Conference on Wireless Mobile Communication and Healthcare - "Transforming healthcare through innovations in mobile and wireless technologies"},
        publisher={IEEE},
        proceedings_a={MOBIHEALTH},
        year={2014},
        month={12},
        keywords={energy expenditure heart rate actigraphy wearable sensors performance analysis},
        doi={10.4108/icst.mobihealth.2014.257292}
    }
    
  • Nandakumar Selvaraj
    Toai Doan
    Year: 2014
    Performance of energy expenditure assessment using a chest-worn wireless patch sensor
    MOBIHEALTH
    IEEE
    DOI: 10.4108/icst.mobihealth.2014.257292
Nandakumar Selvaraj1,*, Toai Doan1
  • 1: Vital Connect Inc, Campbell, CA 95008 USA
*Contact email: snandaa@gmail.com

Abstract

Traditional systems for energy expenditure (EE) assessment are impractical for continuous monitoring in free-living conditions. The study presents the performance of a chest-worn wireless HealthPatchTM sensor for the continuous estimation of EE rate and total energy expenditure (TEE) based on the heart rate and acceleration signals of upper torso. Volunteers (n=32) were attached with patch sensors at three locations on chest, a portable metabolic analyzer, three commercial devices: BodyMediaFIT, Nike+FuelBand and FitBitForce for comparative analysis. Participants carried out a protocol consisted of resting, mild, moderate and intense level of exercises that lasted for 90 min. Analyses of correlation, performance errors and agreement were carried out for the EE rate and TEE values of the patch sensor compared to the metabolic analyzer. The correlation coefficient and mean absolute error of patch sensor's EE rate were 0.94+/-0.04 and 0.67+/-0.24 (Kcal/min), respectively for the collective three patch locations. The patch sensor offered the most accurate estimates of TEE with least mean absolute percentage error of <15%, least bias (0.8 Kcal) and narrowest 95% limits of agreement (-79 to 81 Kcal) than the other consumer based wearable sensors.