Evaluation of the accuracy and reliability for photoplethysmography based heart rate and beat-to-beat detection during daily activities

  • Julia Pietilä
  • Saeed Mehrang
  • Johanna Tolonen
  • Elina Helander
  • Holly Jimison
  • Misha Pavel
  • Ilkka Korhonen
Conference paper
Part of the IFMBE Proceedings book series (IFMBE, volume 65)

Abstract

With the advances in sensor technology and the emergence of new sensor systems, it is important to assess the accuracy of these devices. In this paper, we describe an evaluation study for two wrist-worn devices, namely PulseOn (PO) and Empatica E4 (E4), measuring photoplethysmography – based heart rate (PPG HR) and inter-beat intervals (IBIs). The accuracy and reliability of PPG HR and beat-to-beat detection are evaluated with respect to electrocardiography (ECG) – based HR and IBIs during different daily activities, such as sitting, standing, household work and cycling. The evaluation study employed data from twenty male subjects. The absolute difference of PPG and ECG HR was less than 10 bpm for 90-99% and 81-97% of time for PO and E4, respectively. The accuracy and reliability of the devices were decreased during household work due to the excess hand movements. On average, the mean absolute error in HR was 2.5 bpm higher in PO and 3.7 bpm higher in E4 during household work than during sitting. The percentage of correctly detected heartbeats was 89% for PO and 68% for E4 during sitting but 76% for PO and only 9% for E4 during household work. PO showed better beat-to-beat detection accuracy than E4 in all activities. The errors in heart rate variability measure (HRV) of root mean square of successive inter-beat interval differences were 3.5±3.9 ms for PO and 10.2±6.7 ms for E4 during sitting, but 18.0±10.9 ms for PO and 48.7±21.8 ms for E4 during cycling. As a conclusion, PPG – based wrist-worn devices are accurate and reliable for HR and beat-to-beat detection when the amount of hand movements is not excess but HRV can be estimated from PPG IBI data reliably only during resting conditions. Moreover, there were significant differences in accuracy between different devices.

Keywords

wearable sensor PPG HR HRV 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1. Stahl SE, An H-S, Dinkel DM et al. (2016) How accurate are the wrist-based heart rate monitors during walking and running activities? Are they accurate enough? BMJ Open Sport Exerc Medicine 2: e000106.Google Scholar
  2. 2. Spierer DK, Rosen Z, Litman LL et al. (2015) Validation of photoplethysmography as a method to detect heart rate during rest and exercise. J Med Eng 39(5): 264-271.Google Scholar
  3. 3. Schäfer A, Vagedes J. (2013) How accurate is pulse rate variability as an estimate of heart rate variability?: A review on studies comparing photoplethysmographic technology with an electrocardiogram. International Journal of Cardiology 166(1): 15-29.Google Scholar
  4. 4. Lee J-M, An H-S, Kang S-K et al. (2016) Examining the validity of Fitbit Charge HR for measuring heart rate in free-living conditions. Research Presentations. 15.Google Scholar
  5. 5. Empatica. (2016) Utilizing the PPG/BVP signal. Available at: https://support.empatica.com/hc/en-us/articles/204954639-Utilizing-the-PPG-BVP-signal. Accessed: 27.02.2017.
  6. 6. PulseOn. (2016) Technology. Available at: http://pulseon.com/news/technology. Accessed: 27.02.2017.
  7. 7. Firstbeat Technologies Ltd. Firstbeat Bodyguard 2 guide. Available at: https://www.firstbeat.com/app/uploads/2015/10/150811_Bodyguard2_Guide_ENG.pdf. Accessed: 27.2.2017.
  8. 8. Empatica (2017) Data export and formatting from Empatica Connect. Available at: https://support.empatica.com/hc/en-us/articles/201608896-Data-export-and-formatting-from-Empatica-Connect-. Accessed 27.02.2017.
  9. 9. Empatica (2017) How is IBI.csv obtained? Available at: https://support.empatica.com/hc/en-us/articles/201912319-How-is-IBI-csv-obtained-. Accessed 10.04.2017.
  10. 10. Saalasti S, Seppänen M, Kuusela A. (2004) Artefact correction for heart beat interval data. Advanced Methods for Processing Bioelectrical Signals, ProBisi Meeting 2004.Google Scholar
  11. 11. Parak J, Tarniceriu A, Renevey P et al. (2015) Evaluation of the beat-to-beat detection accuracy of PulseOn wearable optical heart rate monitor. Conf Proc IEEE Eng Med Biol Soc, 2015, 8099-102.Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  • Julia Pietilä
    • 1
  • Saeed Mehrang
    • 1
  • Johanna Tolonen
    • 1
  • Elina Helander
    • 1
  • Holly Jimison
    • 1
    • 2
  • Misha Pavel
    • 1
    • 2
  • Ilkka Korhonen
    • 1
  1. 1.BioMediTech Institute and Faculty of Biomedical Sciences and EngineeringTampere University of TechnologyTampereFinland
  2. 2.College of Computer & Information Science and Bouvé College of Health SciencesNortheastern UniversityBostonUSA

Personalised recommendations