Abstract
Injuries resulting from sports and physical activities can be persistent and pose a substantial problem for player’s economic wellbeing and quality of life. The recent proliferation of internet-of-things and wearable technologies to measure health and activity are enabling researchers and practitioners to collect more precise data than ever before on an increasingly wide range of sport-related activities [1, 2]. Despite this impressive growth of evidence, the use of health analytics in sports has been largely descriptive with less focus attributed to predictive models and real-time analysis of data derived from inertial measurement units (IMUs) and wearable sensors for player’s performance management and injury prevention. The aim of this study is to investigate potential injury prevention values of wearable technology data. We focus on the stochastic nature of injuries to identify major risk factors and to develop injury prediction models. Using Zephyr BioHarness Wearable technology , we collected comprehensive data from student-athletes and cadets to generate insights that allowed us to combine various physiological and motion-based parameters in order to predict injuries and determine the extent to which modifiable risk factors can be altered for optimal training. Our research provides insight for athletic trainers to aid in how to best design physical training to reduce injury risk and maximize performance through improved adaptability to the rapidly changing environmental demands.
Recommended Citation
Zadeh, Amir and Bruce, Scott, "Wearable Technology and Data Science for Injury Prediction in Sports" (2019). AMCIS 2019 Proceedings. 46.
https://aisel.aisnet.org/amcis2019/treo/treos/46
Wearable Technology and Data Science for Injury Prediction in Sports
Injuries resulting from sports and physical activities can be persistent and pose a substantial problem for player’s economic wellbeing and quality of life. The recent proliferation of internet-of-things and wearable technologies to measure health and activity are enabling researchers and practitioners to collect more precise data than ever before on an increasingly wide range of sport-related activities [1, 2]. Despite this impressive growth of evidence, the use of health analytics in sports has been largely descriptive with less focus attributed to predictive models and real-time analysis of data derived from inertial measurement units (IMUs) and wearable sensors for player’s performance management and injury prevention. The aim of this study is to investigate potential injury prevention values of wearable technology data. We focus on the stochastic nature of injuries to identify major risk factors and to develop injury prediction models. Using Zephyr BioHarness Wearable technology , we collected comprehensive data from student-athletes and cadets to generate insights that allowed us to combine various physiological and motion-based parameters in order to predict injuries and determine the extent to which modifiable risk factors can be altered for optimal training. Our research provides insight for athletic trainers to aid in how to best design physical training to reduce injury risk and maximize performance through improved adaptability to the rapidly changing environmental demands.