Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring

Aref Smiley; Ihor Havrylchuk; Te Yi Tsai; Elena Zakashansky; Aileen Gabriel; Taulant Xhakli; Wanting Cui; Irena Parvanova; Hu Cui; Joseph Finkelstein

doi:10.3233/SHTI230335

Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring

Aref Smiley
, Ihor Havrylchuk
, Te Yi Tsai
, Elena Zakashansky
, Aileen Gabriel
, Taulant Xhakli
, Wanting Cui
, Irena Parvanova
, Hu Cui
, Joseph Finkelstein

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Scopus citations

Abstract

This study aimed to build machine learning (ML) algorithms for the automated classification of cycling exercise exertion levels using data from wearable devices. The best predictive features were selected using the minimum redundancy maximum relevance algorithm (mRMR). Top selected features were then used to build and assess the accuracy of five ML classifiers to predict the level of exertion. The Naïve Bayes showed the best F1 score of 79%. The proposed approach may be used for real-time monitoring of exercise exertion.

Original language	English (US)
Title of host publication	Caring is Sharing - Exploiting the Value in Data for Health and Innovation - Proceedings of MIE 2023
Editors	Maria Hagglund, Madeleine Blusi, Stefano Bonacina, Lina Nilsson, Inge Cort Madsen, Sylvia Pelayo, Anne Moen, Arriel Benis, Lars Lindskold, Parisis Gallos
Publisher	IOS Press BV
Pages	1023-1024
Number of pages	2
ISBN (Electronic)	9781643683881
DOIs	https://doi.org/10.3233/SHTI230335
State	Published - May 18 2023
Externally published	Yes
Event	33rd Medical Informatics Europe Conference: Caring is Sharing - Exploiting the Value in Data for Health and Innovation, MIE2023 - Gothenburg, Sweden Duration: May 22 2023 → May 25 2023

Publication series

Name	Studies in Health Technology and Informatics
Volume	302
ISSN (Print)	0926-9630
ISSN (Electronic)	1879-8365

Conference

Conference	33rd Medical Informatics Europe Conference: Caring is Sharing - Exploiting the Value in Data for Health and Innovation, MIE2023
Country/Territory	Sweden
City	Gothenburg
Period	5/22/23 → 5/25/23

Keywords

Aerobic exercise
exertion level
machine learning
wearable devices

ASJC Scopus subject areas

Biomedical Engineering
Health Informatics
Health Information Management

Access to Document

10.3233/SHTI230335

Cite this

Smiley, A., Havrylchuk, I., Tsai, T. Y., Zakashansky, E., Gabriel, A., Xhakli, T., Cui, W., Parvanova, I., Cui, H., & Finkelstein, J. (2023). Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring. In M. Hagglund, M. Blusi, S. Bonacina, L. Nilsson, I. C. Madsen, S. Pelayo, A. Moen, A. Benis, L. Lindskold, & P. Gallos (Eds.), Caring is Sharing - Exploiting the Value in Data for Health and Innovation - Proceedings of MIE 2023 (pp. 1023-1024). (Studies in Health Technology and Informatics; Vol. 302). IOS Press BV. https://doi.org/10.3233/SHTI230335

Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring. / Smiley, Aref; Havrylchuk, Ihor; Tsai, Te Yi et al.
Caring is Sharing - Exploiting the Value in Data for Health and Innovation - Proceedings of MIE 2023. ed. / Maria Hagglund; Madeleine Blusi; Stefano Bonacina; Lina Nilsson; Inge Cort Madsen; Sylvia Pelayo; Anne Moen; Arriel Benis; Lars Lindskold; Parisis Gallos. IOS Press BV, 2023. p. 1023-1024 (Studies in Health Technology and Informatics; Vol. 302).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Smiley, A, Havrylchuk, I, Tsai, TY, Zakashansky, E, Gabriel, A, Xhakli, T, Cui, W, Parvanova, I, Cui, H & Finkelstein, J 2023, Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring. in M Hagglund, M Blusi, S Bonacina, L Nilsson, IC Madsen, S Pelayo, A Moen, A Benis, L Lindskold & P Gallos (eds), Caring is Sharing - Exploiting the Value in Data for Health and Innovation - Proceedings of MIE 2023. Studies in Health Technology and Informatics, vol. 302, IOS Press BV, pp. 1023-1024, 33rd Medical Informatics Europe Conference: Caring is Sharing - Exploiting the Value in Data for Health and Innovation, MIE2023, Gothenburg, Sweden, 5/22/23. https://doi.org/10.3233/SHTI230335

Smiley A, Havrylchuk I, Tsai TY, Zakashansky E, Gabriel A, Xhakli T et al. Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring. In Hagglund M, Blusi M, Bonacina S, Nilsson L, Madsen IC, Pelayo S, Moen A, Benis A, Lindskold L, Gallos P, editors, Caring is Sharing - Exploiting the Value in Data for Health and Innovation - Proceedings of MIE 2023. IOS Press BV. 2023. p. 1023-1024. (Studies in Health Technology and Informatics). doi: 10.3233/SHTI230335

Smiley, Aref ; Havrylchuk, Ihor ; Tsai, Te Yi et al. / Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring. Caring is Sharing - Exploiting the Value in Data for Health and Innovation - Proceedings of MIE 2023. editor / Maria Hagglund ; Madeleine Blusi ; Stefano Bonacina ; Lina Nilsson ; Inge Cort Madsen ; Sylvia Pelayo ; Anne Moen ; Arriel Benis ; Lars Lindskold ; Parisis Gallos. IOS Press BV, 2023. pp. 1023-1024 (Studies in Health Technology and Informatics).

@inproceedings{0c46c5b1ae9147f58b3cbab11d4d53c4,

title = "Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring",

abstract = "This study aimed to build machine learning (ML) algorithms for the automated classification of cycling exercise exertion levels using data from wearable devices. The best predictive features were selected using the minimum redundancy maximum relevance algorithm (mRMR). Top selected features were then used to build and assess the accuracy of five ML classifiers to predict the level of exertion. The Na{\"i}ve Bayes showed the best F1 score of 79\%. The proposed approach may be used for real-time monitoring of exercise exertion.",

keywords = "Aerobic exercise, exertion level, machine learning, wearable devices",

author = "Aref Smiley and Ihor Havrylchuk and Tsai, \{Te Yi\} and Elena Zakashansky and Aileen Gabriel and Taulant Xhakli and Wanting Cui and Irena Parvanova and Hu Cui and Joseph Finkelstein",

note = "Publisher Copyright: {\textcopyright} 2023 European Federation for Medical Informatics (EFMI) and IOS Press.; 33rd Medical Informatics Europe Conference: Caring is Sharing - Exploiting the Value in Data for Health and Innovation, MIE2023 ; Conference date: 22-05-2023 Through 25-05-2023",

year = "2023",

month = may,

day = "18",

doi = "10.3233/SHTI230335",

language = "English (US)",

series = "Studies in Health Technology and Informatics",

publisher = "IOS Press BV",

pages = "1023--1024",

editor = "Maria Hagglund and Madeleine Blusi and Stefano Bonacina and Lina Nilsson and Madsen, \{Inge Cort\} and Sylvia Pelayo and Anne Moen and Arriel Benis and Lars Lindskold and Parisis Gallos",

booktitle = "Caring is Sharing - Exploiting the Value in Data for Health and Innovation - Proceedings of MIE 2023",

}

TY - GEN

T1 - Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring

AU - Smiley, Aref

AU - Havrylchuk, Ihor

AU - Tsai, Te Yi

AU - Zakashansky, Elena

AU - Gabriel, Aileen

AU - Xhakli, Taulant

AU - Cui, Wanting

AU - Parvanova, Irena

AU - Cui, Hu

AU - Finkelstein, Joseph

PY - 2023/5/18

Y1 - 2023/5/18

N2 - This study aimed to build machine learning (ML) algorithms for the automated classification of cycling exercise exertion levels using data from wearable devices. The best predictive features were selected using the minimum redundancy maximum relevance algorithm (mRMR). Top selected features were then used to build and assess the accuracy of five ML classifiers to predict the level of exertion. The Naïve Bayes showed the best F1 score of 79%. The proposed approach may be used for real-time monitoring of exercise exertion.

AB - This study aimed to build machine learning (ML) algorithms for the automated classification of cycling exercise exertion levels using data from wearable devices. The best predictive features were selected using the minimum redundancy maximum relevance algorithm (mRMR). Top selected features were then used to build and assess the accuracy of five ML classifiers to predict the level of exertion. The Naïve Bayes showed the best F1 score of 79%. The proposed approach may be used for real-time monitoring of exercise exertion.

KW - Aerobic exercise

KW - exertion level

KW - machine learning

KW - wearable devices

UR - https://www.scopus.com/pages/publications/85159764669

UR - https://www.scopus.com/pages/publications/85159764669#tab=citedBy

U2 - 10.3233/SHTI230335

DO - 10.3233/SHTI230335

M3 - Conference contribution

C2 - 37203570

AN - SCOPUS:85159764669

T3 - Studies in Health Technology and Informatics

SP - 1023

EP - 1024

BT - Caring is Sharing - Exploiting the Value in Data for Health and Innovation - Proceedings of MIE 2023

A2 - Hagglund, Maria

A2 - Blusi, Madeleine

A2 - Bonacina, Stefano

A2 - Nilsson, Lina

A2 - Madsen, Inge Cort

A2 - Pelayo, Sylvia

A2 - Moen, Anne

A2 - Benis, Arriel

A2 - Lindskold, Lars

A2 - Gallos, Parisis

PB - IOS Press BV

T2 - 33rd Medical Informatics Europe Conference: Caring is Sharing - Exploiting the Value in Data for Health and Innovation, MIE2023

Y2 - 22 May 2023 through 25 May 2023

ER -

Automated Classification of Exercise Exertion Levels Based on Real-Time Wearable Physiological Signal Monitoring

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this