Helmet-Head Decoupling in Ice Hockey Impacts: An In-lab Exploratory Study Using Autoregressive Modeling of...

Area of Science:

• Biomechanics
• Sports Engineering
• Injury Prevention

Background:

• Head linear acceleration estimation in ice hockey is crucial for understanding injury mechanisms.
• Decoupling between helmet and head motion complicates accurate measurement.
• Inertial Measurement Units (IMUs) in helmets offer a potential solution.

Purpose of the Study:

• To develop and validate a novel method for estimating head linear acceleration during ice hockey impacts.
• To utilize AutoRegressive (AR) modeling to address helmet-head decoupling.
• To assess the effectiveness of IMU-instrumented helmets for impact analysis.

Main Methods:

• Impacts were simulated on an Anthropometric Test Device (ATD) using a pendulum impactor.
• Testing covered multiple impact directions and energies.
• An AR-based transfer function was applied to IMU data for acceleration estimation.
• Performance was compared against raw and filtered data and evaluated on different helmet types.

Main Results:

• The AR transfer function significantly reduced Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE) compared to low-pass filtered signals.
• Error reduction was consistent across various impact directions.
• The model's performance decreased when applied to a different helmet model, indicating generalization limitations.

Conclusions:

• The proposed AR modeling methodology enhances head linear acceleration estimation accuracy in ice hockey impacts.
• Helmet-specific calibration or model retraining is necessary for optimal performance across different helmet designs.
• This approach holds promise for improving head impact assessment in sports.