Estimating oxygen uptake in simulated team sports using machine learning models and wearable sensor data: A pilot study.

Accurate assessment of training status in team sports is crucial for optimising performance and reducing injury risk. This pilot study investigates the feasibility of using machine learning (ML) models to estimate oxygen uptake (VO2) with wearable sensors during team sports activities. Six healthy m...

Full description

Saved in:
Bibliographic Details
Main Authors: Dermot Sheridan, Arne Jaspers, Dinh Viet Cuong, Tim Op De Beéck, Niall M Moyna, Toon T de Beukelaar, Mark Roantree
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0319760
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850176186663043072
author Dermot Sheridan
Arne Jaspers
Dinh Viet Cuong
Tim Op De Beéck
Niall M Moyna
Toon T de Beukelaar
Mark Roantree
author_facet Dermot Sheridan
Arne Jaspers
Dinh Viet Cuong
Tim Op De Beéck
Niall M Moyna
Toon T de Beukelaar
Mark Roantree
author_sort Dermot Sheridan
collection DOAJ
description Accurate assessment of training status in team sports is crucial for optimising performance and reducing injury risk. This pilot study investigates the feasibility of using machine learning (ML) models to estimate oxygen uptake (VO2) with wearable sensors during team sports activities. Six healthy male team sports athletes participated in the study. Data were collected using inertial measurement units (IMU), heart rate monitors, and breathing rate sensors during incremental fitness tests. The performance of different ML models, including multiple linear regression (MLR), XGBoost, and deep learning models (LSTM, CNN, MLP), was compared using raw and engineered features from IMU data. Results indicate that while LSTM models with raw IMU data provided the most accurate predictions (RMSE: 4.976, MAE: 3.698 [Formula: see text]), MLR models remained competitive, especially with engineered features. Multi-sensor configurations, particularly those including sensors on the torso and limbs, enhanced prediction accuracy. The findings demonstrate the potential of ML models to monitor VO2 noninvasively in real-time, offering valuable insights into the internal physiological demand during team sports activities.
format Article
id doaj-art-8fb0fe74dd1d47f387edf0c9208738c8
institution OA Journals
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-8fb0fe74dd1d47f387edf0c9208738c82025-08-20T02:19:18ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01204e031976010.1371/journal.pone.0319760Estimating oxygen uptake in simulated team sports using machine learning models and wearable sensor data: A pilot study.Dermot SheridanArne JaspersDinh Viet CuongTim Op De BeéckNiall M MoynaToon T de BeukelaarMark RoantreeAccurate assessment of training status in team sports is crucial for optimising performance and reducing injury risk. This pilot study investigates the feasibility of using machine learning (ML) models to estimate oxygen uptake (VO2) with wearable sensors during team sports activities. Six healthy male team sports athletes participated in the study. Data were collected using inertial measurement units (IMU), heart rate monitors, and breathing rate sensors during incremental fitness tests. The performance of different ML models, including multiple linear regression (MLR), XGBoost, and deep learning models (LSTM, CNN, MLP), was compared using raw and engineered features from IMU data. Results indicate that while LSTM models with raw IMU data provided the most accurate predictions (RMSE: 4.976, MAE: 3.698 [Formula: see text]), MLR models remained competitive, especially with engineered features. Multi-sensor configurations, particularly those including sensors on the torso and limbs, enhanced prediction accuracy. The findings demonstrate the potential of ML models to monitor VO2 noninvasively in real-time, offering valuable insights into the internal physiological demand during team sports activities.https://doi.org/10.1371/journal.pone.0319760
spellingShingle Dermot Sheridan
Arne Jaspers
Dinh Viet Cuong
Tim Op De Beéck
Niall M Moyna
Toon T de Beukelaar
Mark Roantree
Estimating oxygen uptake in simulated team sports using machine learning models and wearable sensor data: A pilot study.
PLoS ONE
title Estimating oxygen uptake in simulated team sports using machine learning models and wearable sensor data: A pilot study.
title_full Estimating oxygen uptake in simulated team sports using machine learning models and wearable sensor data: A pilot study.
title_fullStr Estimating oxygen uptake in simulated team sports using machine learning models and wearable sensor data: A pilot study.
title_full_unstemmed Estimating oxygen uptake in simulated team sports using machine learning models and wearable sensor data: A pilot study.
title_short Estimating oxygen uptake in simulated team sports using machine learning models and wearable sensor data: A pilot study.
title_sort estimating oxygen uptake in simulated team sports using machine learning models and wearable sensor data a pilot study
url https://doi.org/10.1371/journal.pone.0319760
work_keys_str_mv AT dermotsheridan estimatingoxygenuptakeinsimulatedteamsportsusingmachinelearningmodelsandwearablesensordataapilotstudy
AT arnejaspers estimatingoxygenuptakeinsimulatedteamsportsusingmachinelearningmodelsandwearablesensordataapilotstudy
AT dinhvietcuong estimatingoxygenuptakeinsimulatedteamsportsusingmachinelearningmodelsandwearablesensordataapilotstudy
AT timopdebeeck estimatingoxygenuptakeinsimulatedteamsportsusingmachinelearningmodelsandwearablesensordataapilotstudy
AT niallmmoyna estimatingoxygenuptakeinsimulatedteamsportsusingmachinelearningmodelsandwearablesensordataapilotstudy
AT toontdebeukelaar estimatingoxygenuptakeinsimulatedteamsportsusingmachinelearningmodelsandwearablesensordataapilotstudy
AT markroantree estimatingoxygenuptakeinsimulatedteamsportsusingmachinelearningmodelsandwearablesensordataapilotstudy