Login / Signup

Laboratory Evaluation of the gForce Tracker™, a Head Impact Kinematic Measuring Device for Use in Football Helmets.

Kody R CampbellMeagan J WarnicaIris C LevineJeffrey S BrooksAndrew C LaingTimothy A BurkhartJames P Dickey
Published in: Annals of biomedical engineering (2015)
This study sought to compare a new head impact-monitoring device, which is not limited to specific helmet styles, against reference accelerometer measurements. Laboratory controlled impacts were delivered using a linear pneumatic impactor to a Hybrid III headform (HIII) fitted with a football helmet and the impact monitoring device (gForce Tracker-GFT) affixed to the inside of the helmet. Linear regression analyses and absolute mean percent error (MAPE) were used to compare the head impact kinematics measured by the GFT to a reference accelerometer located at the HIII's center of mass. The coefficients of determination were strong for the peak linear acceleration, peak rotational velocity, and HIC15 across all impact testing locations (r(2) = 0.82, 0.94, and 0.70, respectively), but there were large MAPE for the peak linear acceleration and HIC15 (MAPE = 49 ± 21% and 108 ± 58%). The raw GFT was accurate at measuring the peak rotational velocity at the center of mass of the HIII (MAPE = 9%). Results from the impact testing were used to develop a correction algorithm. The coefficients of determination for all impact parameters improved using the correction algorithm for the GFT (r(2) > 0.97), and the MAPE were less than 14%. The GFT appears to be a suitable impact-monitoring device that is not limited to specific styles of football helmets, however, correction algorithms will need to be developed for each helmet style.
Keyphrases
  • machine learning
  • physical activity
  • deep learning
  • high resolution
  • blood flow
  • molecularly imprinted