Assistant Professor The University of British Columbia, United States
Introduction:: Soccer has a relatively high injury rate compared to other contact sports, including volleyball, basketball, field hockey and rugby [1]. It has been shown that mental fatigue increases over the course of a 16-week pre-season in elite female netball players [1]. Fatigue has also been shown to induce biomechanical patterns that lead to increased risk of hamstring injury in soccer players [2]. As players progress through the season, rather than looking at fatigue directly, using typical subjective metrics such as the visual analogue scale [1], [3], there may be value in looking at the changes in the kinematics of play, especially those relevant to injury. Muscular fatigue has been shown to increase the number of accelerations undertaken by soccer players [3]. Because inertial measurement units (IMUs) can collect linear acceleration and angular velocity data on-field, an IMU is ideal for exploring kinematic changes. These metrics can be collected and compared over a series of events, rather than at one or two timepoints. Additionally, data such as the lateral acceleration, frontal plane and horizontal plane angular acceleration can be obtained, which are associated with movements with an increased risk of injury, such as cutting and pivoting. Therefore, this abstract takes a preliminary look at IMU data collected during events, to investigate the change in athlete kinematics over the course of a competition season as part of a study using IMUs for injury occurrence characterization in soccer.
Materials and Methods:: During this pilot field deployment, a single IMU (ICM 20649, with range ±30g and ±4000deg/s), sampling at 1000Hz was deployed to 15 female collegiate soccer players over 257 player events (including 46 player games and 211 player practices) at the University of British Columbia during the Fall 2022 season. The sensor was securely mounted between the shoulder blades of each player using hypoallergenic tape. Video cameras were also deployed to the pitch to record athlete movements. Participants then proceeded to play as normal. Full 6-degree-of-freedom linear acceleration and angular velocity was recorded.
Data processing was performed using MATLAB R2021b software (The Mathworks Inc., Natick, MA, USA). Signals were processed using a zero-lag 2nd order low-pass Butterworth filter with cut-off 100 Hz. A custom peak-detection method on the anteroposterior axis was used to identify foot strikes with peak magnitude >25m/s2. Angular acceleration was determined from angular velocity numerically, using the 5-point stencil to take the derivative. Peak mediolateral accelerations, frontal plane angular accelerations, and horizontal plane angular accelerations were assessed during practices over the course of the season in a single player, totalling 23 player practices over the course of 70 days. The median peak acceleration per player event along the access of interest was compared to assess change. A linear regression, with 95% confidence intervals and Pearson’s correlation coefficient were reported relating the metric of interest and time.
Results, Conclusions, and Discussions:: Results: The aim of this preliminary analysis was to determine whether a change could be observed in the median peak acceleration of high-impact foot strikes over the course of the competition season. There is a positive, albeit weak, correlation between season day and the number of high-impact foot strikes. The median peak lateral linear acceleration and median peak frontal plane angular acceleration during these foot strikes also increased throughout the season. This contrasts with an observed decrease in the median peak horizontal plane angular acceleration, suggesting a change in athlete behaviour over time.
Discussion and Conclusion: The increase in the magnitude of the lateral linear acceleration and frontal plane angular acceleration peaks could suggest that players are increasing the intensity with which they perform cutting and bending motions. Based on this preliminary analysis, the angular acceleration in the frontal plane and lateral linear acceleration may prove relevant for determining quantitative metrics with respect to movements with an increased risk of injury. It is also worth exploring the decrease in horizontal plane acceleration for these high-impact movements. However, a detailed investigation of the movement type is necessary to investigate the changing kinematics with respect to activity exposure over the course of the season, which is the focus of ongoing data analysis. In doing so, the aim is to separate exposure hours on field by activity type, as not all athlete hours carry the same risk of injury, and separating exposure hours by activity type can elucidate patterns in injury occurrence [4].
Acknowledgements (Optional): : We would like to acknowledge the UBC Women's Soccer team and coaching team for their enthusiasm towards sports research.
References (Optional): : [1] [1] S. Russell, D. G. Jenkins, S. L. Halson, and V. G. Kelly, “Mental fatigue increases across a 16-week pre-season in elite female athletes,” J Sci Med Sport, vol. 25, no. 4, pp. 356–361, Apr. 2022, doi: 10.1016/J.JSAMS.2021.12.002.
[2] [2] K. Small, L. McNaughton, M. Greig, and R. Lovell, “The effects of multidirectional soccer-specific fatigue on markers of hamstring injury risk,” J Sci Med Sport, vol. 13, no. 1, pp. 120–125, Jan. 2010, doi: 10.1016/J.JSAMS.2008.08.005.
[3] [3] D. Coutinho, B. Gonçalves, D. P. Wong, B. Travassos, A. J. Coutts, and J. Sampaio, “Exploring the effects of mental and muscular fatigue in soccer players’ performance,” Hum Mov Sci, vol. 58, pp. 287–296, Apr. 2018, doi: 10.1016/J.HUMOV.2018.03.004.
[4] [4] M. K. Krill, J. R. Borchers, J. T. Hoffman, R. L. Tatarski, and T. E. Hewett, “A 4-year Single Program Analysis, The Physician and Sportsmedicine,” vol. 45, no. 1, pp. 26–30, 2017, doi: 10.1080/00913847.2017.1279520.