Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification
Accumulation of head impacts may contribute to acute and long-term brain trauma. Wearable sensors can measure impact exposure, yet current sensors do not have validated impact detection methods for accurate exposure monitoring. Here we demonstrate a head impact detection method that can be implement...
Main Authors: | , , , , , , , , , , , |
---|---|
Other Authors: | |
Language: | en |
Published: |
NATURE PUBLISHING GROUP
2017
|
Online Access: | http://hdl.handle.net/10150/627166 http://arizona.openrepository.com/arizona/handle/10150/627166 |
id |
ndltd-arizona.edu-oai-arizona.openrepository.com-10150-627166 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-arizona.edu-oai-arizona.openrepository.com-10150-6271662018-04-02T03:00:29Z Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification Wu, Lyndia C. Kuo, Calvin Loza, Jesus Kurt, Mehmet Laksari, Kaveh Yanez, Livia Z. Senif, Daniel Anderson, Scott C. Miller, Logan E. Urban, Jillian E. Stitzel, Joel D. Camarillo, David B. Univ Arizona Accumulation of head impacts may contribute to acute and long-term brain trauma. Wearable sensors can measure impact exposure, yet current sensors do not have validated impact detection methods for accurate exposure monitoring. Here we demonstrate a head impact detection method that can be implemented on a wearable sensor for detecting field football head impacts. Our method incorporates a support vector machine classifier that uses biomechanical features from the time domain and frequency domain, as well as model predictions of head-neck motions. The classifier was trained and validated using instrumented mouthguard data from collegiate football games and practices, with ground truth data labels established from video review. We found that low frequency power spectral density and wavelet transform features (10 similar to 30 Hz) were the best performing features. From forward feature selection, fewer than ten features optimized classifier performance, achieving 87.2% sensitivity and 93.2% precision in cross-validation on the collegiate dataset (n = 387), and over 90% sensitivity and precision on an independent youth dataset (n = 32). Accurate head impact detection is essential for studying and monitoring head impact exposure on the field, and the approach in the current paper may help to improve impact detection performance on wearable sensors. 2017-12-21 Article Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification 2017, 8 (1) Scientific Reports 2045-2322 10.1038/s41598-017-17864-3 http://hdl.handle.net/10150/627166 http://arizona.openrepository.com/arizona/handle/10150/627166 Scientific Reports en http://www.nature.com/articles/s41598-017-17864-3 © The Author(s) 2018. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License. NATURE PUBLISHING GROUP |
collection |
NDLTD |
language |
en |
sources |
NDLTD |
description |
Accumulation of head impacts may contribute to acute and long-term brain trauma. Wearable sensors can measure impact exposure, yet current sensors do not have validated impact detection methods for accurate exposure monitoring. Here we demonstrate a head impact detection method that can be implemented on a wearable sensor for detecting field football head impacts. Our method incorporates a support vector machine classifier that uses biomechanical features from the time domain and frequency domain, as well as model predictions of head-neck motions. The classifier was trained and validated using instrumented mouthguard data from collegiate football games and practices, with ground truth data labels established from video review. We found that low frequency power spectral density and wavelet transform features (10 similar to 30 Hz) were the best performing features. From forward feature selection, fewer than ten features optimized classifier performance, achieving 87.2% sensitivity and 93.2% precision in cross-validation on the collegiate dataset (n = 387), and over 90% sensitivity and precision on an independent youth dataset (n = 32). Accurate head impact detection is essential for studying and monitoring head impact exposure on the field, and the approach in the current paper may help to improve impact detection performance on wearable sensors. |
author2 |
Univ Arizona |
author_facet |
Univ Arizona Wu, Lyndia C. Kuo, Calvin Loza, Jesus Kurt, Mehmet Laksari, Kaveh Yanez, Livia Z. Senif, Daniel Anderson, Scott C. Miller, Logan E. Urban, Jillian E. Stitzel, Joel D. Camarillo, David B. |
author |
Wu, Lyndia C. Kuo, Calvin Loza, Jesus Kurt, Mehmet Laksari, Kaveh Yanez, Livia Z. Senif, Daniel Anderson, Scott C. Miller, Logan E. Urban, Jillian E. Stitzel, Joel D. Camarillo, David B. |
spellingShingle |
Wu, Lyndia C. Kuo, Calvin Loza, Jesus Kurt, Mehmet Laksari, Kaveh Yanez, Livia Z. Senif, Daniel Anderson, Scott C. Miller, Logan E. Urban, Jillian E. Stitzel, Joel D. Camarillo, David B. Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification |
author_sort |
Wu, Lyndia C. |
title |
Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification |
title_short |
Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification |
title_full |
Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification |
title_fullStr |
Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification |
title_full_unstemmed |
Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification |
title_sort |
detection of american football head impacts using biomechanical features and support vector machine classification |
publisher |
NATURE PUBLISHING GROUP |
publishDate |
2017 |
url |
http://hdl.handle.net/10150/627166 http://arizona.openrepository.com/arizona/handle/10150/627166 |
work_keys_str_mv |
AT wulyndiac detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT kuocalvin detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT lozajesus detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT kurtmehmet detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT laksarikaveh detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT yanezliviaz detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT senifdaniel detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT andersonscottc detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT millerlogane detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT urbanjilliane detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT stitzeljoeld detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification AT camarillodavidb detectionofamericanfootballheadimpactsusingbiomechanicalfeaturesandsupportvectormachineclassification |
_version_ |
1718617760844480512 |