Background: Head impacts in sports can produce brain injuries. The accurate quantification of head kinematics through instrumented mouthguards (iMG) can help identify underlying brain motion during injurious impacts. The aim of the current study is to assess the validity of an iMG across a large range of linear and rotational accelerations to allow for on-field head impact monitoring. Methods: Drop tests of an instrumented helmeted anthropometric testing device (ATD) were performed across a range of impact magnitudes and locations, with iMG measures collected concurrently. ATD and iMG kinematics were also fed forward to high-fidelity brain models to predict maximal principal strain. Results: The impacts produced a wide range of head kinematics (16–171 g, 1330–10,164 rad/s² and 11.3–41.5 rad/s) and durations (6–18 ms), representing impacts in rugby and boxing. Comparison of the peak values across ATD and iMG indicated high levels of agreement, with a total concordance correlation coefficient of 0.97 for peak impact kinematics and 0.97 for predicted brain strain. We also found good agreement between iMG and ATD measured time-series kinematic data, with the highest normalized root mean squared error for rotational velocity (5.47 ± 2.61%) and the lowest for rotational acceleration (1.24 ± 0.86%). Our results confirm that the iMG can reliably measure laboratory-based head kinematics under a large range of accelerations and is suitable for future on-field validity assessments.

Instrumented mouthguards (iMG) measure head kinematics in sport, but their measurements have not been validated at high levels of accelerations observed in those sports like rugby. In addition, the effects of filter cut-off frequency on the measured kinematics are still unknown. To address these questions, a drop testing helmeted head-form of an anthropometric testing device (ATD) was used to produce a range of accelerations and accurately control them. Peak linear acceleration (PLA), rotational velocity (PRV), rotational acceleration (PRA) and maximum principal strain (MPS) values were computed. The influence of filter cut-off frequency on peak kinematics was also calculated. Comparison of the peak values across ATD and iMG indicated high levels of agreement, with a total concordance correlation coefficient of 0.97 and intraclass correlation coefficients of 0.990 for PLA, 0.970 for PRV, 0.945 for PRA, and 0.970 for MPS. Cut-off frequencies of 100-300Hz did not significantly attenuate peak kinematics, but frequencies lower than 100Hz did. This is the first study to test an iMG under impact conditions seen in sport. The method presented can be used for in-lab validation of iMGs under head accelerations seen in sport. Furthermore, these results can contribute towards defining standards for filtering iMG data.