On-Field Evaluation of Mouthpiece-and-Helmet-Mounted Sensor Data from Head Kinematics in Football

• Published in: Annals of biomedical engineering Vol. 52; no. 10; pp. 2655 - 2665
• Main Authors: Holcomb, Ty D., Marks, Madison E., Pritchard, N. Stewart, Miller, Logan E., Rowson, Steve, Bullock, Garrett S., Urban, Jillian E., Stitzel, Joel D.
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01-10-2024; Springer Nature B.V
• Subjects: Athletes; Biochemistry; Biological and Medical Physics; Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Biophysics; Classical Mechanics; Error analysis; Football; Head; Helmets; High school football; Kinematics; Mouthguards; Protective equipment; S.I. : Concussions II; Sensors; Telemetry; Head kinematics; Head Impact Telemetry System; Head impact exposure; American Football; Instrumented mouthguard; Video review
• ISSN: 0090-6964; 1573-9686; 1573-9686
• DOI: 10.1007/s10439-024-03583-0

Purpose
Wearable sensors are used to measure head impact exposure in sports. The Head Impact Telemetry (HIT) System is a helmet-mounted system that has been commonly utilized to measure head impacts in American football. Advancements in sensor technology have fueled the development of alternative sensor methods such as instrumented mouthguards. The objective of this study was to compare peak magnitude measured from high school football athletes dually instrumented with the HIT System and a mouthpiece-based sensor system.
Methods
Data was collected at all contact practices and competitions over a single season of spring football. Recorded events were observed and identified on video and paired using event timestamps. Paired events were further stratified by removing mouthpiece events with peak resultant linear acceleration below 10 g and events with contact to the facemask or body of athletes.
Results
A total of 133 paired events were analyzed in the results. There was a median difference (mouthpiece subtracted from HIT System) in peak resultant linear and rotational acceleration for concurrently measured events of 7.3 g and 189 rad/s
2
. Greater magnitude events resulted in larger kinematic differences between sensors and a Bland Altman analysis found a mean bias of 8.8 g and 104 rad/s
2
, respectively.
Conclusion
If the mouthpiece-based sensor is considered close to truth, the results of this study are consistent with previous HIT System validation studies indicating low error on average but high scatter across individual events. Future researchers should be mindful of sensor limitations when comparing results collected using varying sensor technologies.