About | Contact Us | Register | Login
ProceedingsSeriesJournalsSearchEAI
Pervasive Computing Technologies for Healthcare. 17th EAI International Conference, PervasiveHealth 2023, Malmö, Sweden, November 27-29, 2023, Proceedings

Research Article

An IoT-Based Method for Collecting Reference Walked Distance for the 6-Minute Walk Test

Cite
BibTeX Plain Text
  • @INPROCEEDINGS{10.1007/978-3-031-59717-6_31,
        author={Sara Caramaschi and J\^{e}r\^{e}my Bezan\`{e}on and Carl Magnus Olsson and Dario Salvi},
        title={An IoT-Based Method for Collecting Reference Walked Distance for the 6-Minute Walk Test},
        proceedings={Pervasive Computing Technologies for Healthcare. 17th EAI International Conference, PervasiveHealth 2023, Malm\o{}, Sweden, November 27-29, 2023, Proceedings},
        proceedings_a={PERVASIVEHEALTH},
        year={2024},
        month={6},
        keywords={6MWT odometer walk distance IoT inertial sensors},
        doi={10.1007/978-3-031-59717-6_31}
    }
    
  • Sara Caramaschi
    Jérémy Bezançon
    Carl Magnus Olsson
    Dario Salvi
    Year: 2024
    An IoT-Based Method for Collecting Reference Walked Distance for the 6-Minute Walk Test
    PERVASIVEHEALTH
    Springer
    DOI: 10.1007/978-3-031-59717-6_31
Sara Caramaschi1,*, Jérémy Bezançon, Carl Magnus Olsson1, Dario Salvi1
  • 1: Internet of Things and People
*Contact email: sara.caramaschi@mau.se

Abstract

This paper addresses the need for accurate and continuous measurement of walked distance in applications such as indoor localisation, gait analysis or the 6-minute walk test (6MWT). We propose a method to continuously collect ground truth data of walked distance using an IoT-based trundle wheel. The wheel is connected via Bluetooth Low Energy to a smartphone application which allows the collection of inertial sensor data and GPS location information in addition to the reference distance. We prove the usefulness of this data collection approach in a use case where we derive walked distance from inertial data. We train a 1-dimensional CNN on inertial data collected by one researcher in 15 walking sessions of 1 km length at varying speeds. The training is facilitated by the continuous nature of the reference data. The accuracy of the algorithm is then tested on holdout data of a 6-min duration for which the error of the inferred distance is within clinically significant limits. The proposed approach is useful for the efficient collection of input and reference data for the development of algorithms used to estimate walked distance, such as for the 6MWT.

Keywords
6MWT odometer walk distance IoT inertial sensors
Published
2024-06-04
Appears in
SpringerLink
http://dx.doi.org/10.1007/978-3-031-59717-6_31
Copyright © 2023–2025 ICST
EBSCOProQuestDBLPDOAJPortico
EAI Logo

About EAI

  • Who We Are
  • Leadership
  • Research Areas
  • Partners
  • Media Center

Community

  • Membership
  • Conference
  • Recognition
  • Sponsor Us

Publish with EAI

  • Publishing
  • Journals
  • Proceedings
  • Books
  • EUDL