Sports concussions offer a unique opportunity to study head kinematics associated with mild traumatic brain injury. In this study, a model-based image matching (MBIM) approach was employed to analyze video footage of 57 concussions which occurred in National Football League (NFL) games. By utilizing at least two camera views, higher frame rate footage (> 60 images s⁻¹), and laser scans of the field and helmets involved in each case, it was possible to calculate the change in velocity of the helmet during impact in six degrees of freedom. The average impact velocity for these concussive events was 8.9 ± 2.0 m s⁻¹. The average changes in translational and rotational velocity for the concussed players’ helmets were 6.6 ± 2.1 m s⁻¹ and 29 ± 13 rad s⁻¹, respectively. The average change in translational velocity was higher for helmet-to-ground (n = 16) impacts compared to helmet-to-helmet (n = 30) or helmet-to-shoulder (n = 11) events (p < 0.001), while helmet-to-shoulder impacts had a smaller change in rotational velocity compared to the other impact sources (p < 0.001). By quantifying the impact velocities and locations associated with concussive impacts in professional American football, this study provides information that may be used to improve upon current helmet testing methodologies.

运动脑震荡提供了一个独特的机会来研究与轻度创伤性脑损伤相关的头部运动学。在这项研究中，采用基于模型的图像匹配 (MBIM) 方法来分析美国国家橄榄球联盟 (NFL) 比赛中发生的 57 次脑震荡的视频片段。通过使用至少两个摄像机视图、更高帧速率的镜头（> 60 个图像 s ^-1）以及对每种情况下所涉及的场地和头盔的激光扫描，可以计算出撞击期间头盔速度的变化 6自由度。这些震荡事件的平均撞击速度为 8.9 ± 2.0 m s ^-1。脑震荡球员头盔的平移和旋转速度的平均变化为 6.6 ± 2.1 m s ^-1和 29 ± 13 rad s^-1，分别。与头盔对头盔 ( n  = 30) 或头盔对肩膀 ( n  = 11) 事件相比，头盔对地面 ( n  = 16) 撞击的平移速度的平均变化更高( p  < 0.001)，而与其他撞击源相比，头盔对肩撞击的旋转速度变化较小（p  < 0.001）。通过量化与职业美式足球中的震荡冲击相关的冲击速度和位置，本研究提供了可用于改进当前头盔测试方法的信息。