Wu, L.C. ; Zarnescu, L. ; Nangia, V. ; Cam, B. ; Camarillo, D.B., Stanford University
Volume: 61 , Issue: 11, Page(s): 2659 – 2668
Injury from blunt head impacts causes acute neurological deficits and repeated trauma may lead to chronic neurodegeneration. To prevent repeat injury, legislative efforts such as the Lystedt Law require athletes be removed from play if suspected of sustaining head trauma. However, such legislations cannot be fully enforced due to injury under-reporting in sports. A more objective method of injury screening may be head impact monitoring.
We developed a head impact detection system that has the potential to enable real-time trauma screening on the field. The simplest approach for head impact detection is an acceleration thresholding algorithm, which may falsely detect high-acceleration spurious events such as manual manipulation of the device. To allow for high-sensitivity detection while minimizing false positives, our system distinguishes head impacts from non-impacts through two subsystems. First, we use infrared proximity sensing to detect the presence of teeth in the mouthguard tray, and filter out all events where the mouthguard is not worn, or off-teeth. Second, on-teeth, non-impact events are rejected using a support vector machine classifier trained on frequency domain features of head center-of-gravity linear acceleration and rotational velocity. The remaining events are classified as head impacts.
In a controlled laboratory evaluation, the present system performed substantially better than a 10g acceleration threshold in head impact detection (98% sensitivity, 99.99% specificity, 99% accuracy, and 99.98% precision, compared to 92% sensitivity, 58% specificity, 65% accuracy, and 37% precision). Once adapted for field deployment by training and validation with field data, this system has the potential to effectively detect head trauma in sports, military service, and other high-risk activities.
Keywords: impact detection, support vector machines, infrared proximity sensing, traumatic brain injury
Group homepage: http://camlab.stanford.edu/