Related Experiment Video
Updated: Oct 26, 2025
07:30
A Test Bed to Examine Helmet Fit and Retention and Biomechanical Measures of Head and Neck Injury in Simulated Impact
Published on: September 21, 2017
9.1K
Effective Head Impact Kinematics to Preserve Brain Strain
Kianoosh Ghazi1, Shaoju Wu1, Wei Zhao1
1Department of Biomedical Engineering, Worcester Polytechnic Institute, 60 Prescott Street, Worcester, MA, 01605, USA.
Annals of Biomedical Engineering
|August 3, 2021
Summary
This study introduces a new method using convolutional neural networks to simplify head impact analysis, accurately preserving peak brain strain and location. This advances head injury biomechanics beyond traditional metrics.
Area of Science:
- Biomechanics
- Neuroscience
- Computational modeling
Background:
- Current brain injury metrics often simplify complex impacts, neglecting critical spatial strain information.
- Peak Maximum Principal Strain (MPS) is a key indicator, but its location is often ignored.
Purpose of the Study:
- To develop effective impact kinematics preserving both peak MPS and spatially detailed MPS.
- To automate head impact simplification using a convolutional neural network (CNN).
Main Methods:
- Developed a CNN trained on a large dataset (N=3069) to automate impact simplification.
- Utilized a pre-computed brain response atlas (pcBRA) for reference.
- Validated the method by matching impacts with pcBRA idealized impacts based on elementwise MPS.
Main Results:
- CNN-estimated effective peak rotational velocity achieved a high coefficient of determination (R² ≈ 0.96) compared to traditional metrics (R² ≈ 0.34).
- The method accurately characterized effective peak rotation velocity and rotational axis for 73.5% of impacts.
- Results demonstrated superior preservation of spatially detailed MPS compared to conventional metrics.
Conclusions:
- The developed CNN-based approach offers a more accurate and spatially detailed assessment of head impacts.
- This method advances head injury biomechanics by reducing arbitrary impacts into idealized 'impact modes'.
- Future work could enhance accuracy by expanding the pcBRA and focusing on region-wise strains.
