On-Field Performance of an Instrumented Mouthguard for Detecting Head Impacts in American Football

Abstract

Wearable sensors that accurately record head impacts experienced by athletes during play can enable a wide range of potential applications including equipment improvements, player education, and rule changes. One challenge for wearable systems is their ability to discriminate head impacts from recorded spurious signals. This study describes the development and evaluation of a head impact detection system consisting of a mouthguard sensor and machine learning model for distinguishing head impacts from spurious events in football games. Twenty-one collegiate football athletes participating in 11 games during the 2018 and 2019 seasons wore a custom-fit mouthguard instrumented with linear and angular accelerometers to collect kinematic data. Video was reviewed to classify sensor events, collected from instrumented players that sustained head impacts, as head impacts or spurious events. Data from 2018 games were used to train the ML model to classify head impacts using kinematic data features (127 head impacts; 305 non-head impacts). Performance of the mouthguard sensor and ML model were evaluated using an independent test dataset of 3 games from 2019 (58 head impacts; 74 non-head impacts). Based on the test dataset results, the mouthguard sensor alone detected 81.6% of video-confirmed head impacts while the ML classifier provided 98.3% precision and 100% recall, resulting in an overall head impact detection system that achieved 98.3% precision and 81.6% recall.

Ready for Impact? A validity and feasibility study of instrumented mouthguards (iMGs)

“…Further analysis of false-negatives for FRI revealed that the iMG algorithm incorrectly binned 17.4% of false-negatives. The FRI algorithm was trained on American Football data,[16] therefore optimising the algorithm using non-helmeted sports would likely improve false-negative performance. HitIQ false-negative performance was likely due to the 13 g recording threshold, which was different to the trigger threshold (Table 1).…”

Head Impact Exposure and Biomechanics in University Varsity Women’s Soccer

“…17 Custom-made mouthguards have a similar form factor as regular athletics mouthguards, and have been deployed in American football, Australian rugby union, and boxing. 2 , 12 , 16 , 20 , 28 More recently, custom-fitted mouthpieces with sensors embedded in acrylic and molded to the posterior side of the upper dentition have also been validated for head impact measurement, where peak linear acceleration, angular velocity, and angular acceleration measurements linearly correlate with reference measurements with r 2 ≥ 0.95 and slope falling within 1 ± 0.04. 41 Additionally, through video verification of on-field head impacts, the mouthpieces showed an overall sensitivity of 69.2% and a positive predictive value of 80.3% in head impact detection.…”

Value Analysis of Video Image Multiprocessing Based on Vision Sensor in the Field of Football Scientific Research

“…He studied the development and evaluation of head impact detection systems to distinguish between head hits in football matches and false incidents. Data from the 2018 game was used to train ML models to classify head hits using kinematic data features, and his research lacked data [1]. Suarez et al introduced a CMOS vision sensor chip using standard 0.18 μ M CMOS technology for Gaussian pyramid extraction.…”