Citation

Campbell KR, Marshall SW, Luck JF, Pinton GF, Stitzel JD, Boone JS, Guskiewicz KM, Mihalik JP. Med. Sci. Sports Exerc. 2020; 52(10): 2198-2206.

Copyright

(Copyright © 2020, Lippincott Williams and Wilkins)

DOI

10.1249/MSS.0000000000002371

PMID

32936594

Abstract

PURPOSE: This study aimed to quantify the Head Impact Telemetry (HIT) System's impact detection and location measurement accuracy using an impact biomechanics data set paired with video of high school football special teams plays.

METHODS: The head impact biomechanics data set and video were collected from 22 high school football players, wearing HIT System instrumented helmets, competing in 218 special teams plays over a single high school football season. We used two separate video analysis approaches. To quantify the impact detection accuracy, we evaluated the video for head impacts independently of the impact data collection triggers collected by the HIT System. Video-observed impacts matched to valid and invalid head impacts by the HIT System algorithm were categorized as true positives, false positives, false negatives, and true negatives. To quantify impact location accuracy, we analyzed video-synchronized head impacts for impact location independent of the HIT System's impact location measurement and quantified the estimated percent agreement of impact location between the HIT System recorded impact location and the impact location observed on video.

RESULTS: The HIT System's impact-filtering algorithm had 69% sensitivity, 72% specificity, and 70% accuracy in categorizing true and non-head impact data collection triggers. The HIT System agreed with video-observed impact locations on 64% of the 129 impacts we analyzed (unweighted k = 0.43, 95% confidence interval = 0.31-0.54).

CONCLUSION: This work provides data on the HIT System's impact detection and location accuracy during high school football special teams plays using game video analysis that has not been previously published. Based on our data, we believe that the HIT System is useful for estimating population-based impact location distributions for special teams plays.

Language: en