Version 1
: Received: 22 April 2024 / Approved: 22 April 2024 / Online: 23 April 2024 (03:17:24 CEST)
How to cite:
White, M.; De Lazzari, B.; Bezodis, N.; Camomilla, V. Wearable Sensors for Athletic Performance: A Comparison of Discrete and Continuous Feature Extraction Methods for Prediction Models. Preprints2024, 2024041417. https://doi.org/10.20944/preprints202404.1417.v1
White, M.; De Lazzari, B.; Bezodis, N.; Camomilla, V. Wearable Sensors for Athletic Performance: A Comparison of Discrete and Continuous Feature Extraction Methods for Prediction Models. Preprints 2024, 2024041417. https://doi.org/10.20944/preprints202404.1417.v1
White, M.; De Lazzari, B.; Bezodis, N.; Camomilla, V. Wearable Sensors for Athletic Performance: A Comparison of Discrete and Continuous Feature Extraction Methods for Prediction Models. Preprints2024, 2024041417. https://doi.org/10.20944/preprints202404.1417.v1
APA Style
White, M., De Lazzari, B., Bezodis, N., & Camomilla, V. (2024). Wearable Sensors for Athletic Performance: A Comparison of Discrete and Continuous Feature Extraction Methods for Prediction Models. Preprints. https://doi.org/10.20944/preprints202404.1417.v1
Chicago/Turabian Style
White, M., Neil Bezodis and Valentina Camomilla. 2024 "Wearable Sensors for Athletic Performance: A Comparison of Discrete and Continuous Feature Extraction Methods for Prediction Models" Preprints. https://doi.org/10.20944/preprints202404.1417.v1
Abstract
Wearable sensors have become increasingly popular for assessing athletic performance, but the optimal methods for processing and analysing the data remain unclear. This study investigates the efficacy of discrete and continuous feature extraction methods, separately and in combination, for modelling countermovement jump performance using wearable sensor data. We demonstrate that continuous features, particularly those derived from Functional Principal Component Analysis, outperform discrete features in terms of model performance, robustness to variations in data distribution and volume, and consistency across different datasets. Our findings underscore the importance of methodological choices, such as signal type, alignment methods, and model selection, in developing accurate and generalisable predictive models. We also highlight the potential pitfalls of relying solely on domain knowledge for feature selection and the benefits of data-driven approaches. Furthermore, we discuss the implications of our findings for the broader field of sports biomechanics and provide practical recommendations for researchers and practitioners aiming to leverage wearable sensor data for athletic performance assessment. Our results contribute to the development of more reliable and widely applicable predictive models, facilitating the use of wearable technology for optimising training and enhancing athletic outcomes across various sports disciplines.
Keywords
countermovement jump; jump power; inertial measurement units; smartphone; accelerometer; wearables; functional principal component analysis; feature extraction; signal alignment; sport
Subject
Computer Science and Mathematics, Signal Processing
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
