Related Experiment Video
Updated: Jun 26, 2026
Stereotaxic Surgical Approach to Microinject the Caudal Brainstem and Upper Cervical Spinal Cord via the Cisterna Magna in Mice
Published on: January 21, 2022
Geometric scaling factors for the pediatric brainstem
Kerry A Danelson1, Mao Yu, F Scott Gayzik
1Wake Forest University School of Medicine, Virginia Tech - Wake Forest University Center for Injury Biomechanics, Winston-Salem, NC, 27157, USA.
Insights
This study refines pediatric brainstem models for vehicle crash simulations. Geometric morphometrics identified age-related shape changes, improving injury prediction for child occupants in motor vehicle crashes (MVCs).
Area of Science:
- Biomechanical Engineering
- Pediatric Injury Biomechanics
- Computational Anatomy
Background:
- Motor vehicle crashes (MVCs) are a leading cause of death and head injury in children.
- Accurate finite element models of pediatric brains are crucial for predicting occupant responses.
- Current adult models require adaptation for pediatric use, particularly for the brainstem.
Purpose of the Study:
- To improve geometric scaling factors for the pediatric brainstem.
- To adapt existing adult brain models for pediatric occupants.
- To enhance the accuracy of injury prediction metrics for children in MVCs.
Main Methods:
- Geometric morphometric analysis using sliding landmarks and Generalized Procrustes Analysis.
- Magnetic resonance imaging (MRI) data from 59 individuals aged newborn to 21 years.
- Regression of landmark coordinates onto age to develop age-based predictive models.
Main Results:
- Quantified age-related shape changes in the pediatric brainstem.
- Developed a model predicting pediatric brainstem landmark locations based on age.
- Calculated geometric scaling factors for brainstem shape across pediatric age groups.
Conclusions:
- The study provides improved geometric scaling factors for pediatric brainstem models.
- These findings contribute to more accurate finite element models for pediatric occupant safety.
- Enhanced models can lead to better injury prediction and prevention strategies for child MVC occupants.
Abstract:
Injuries caused by motor vehicle crashes (MVCs) are the leading cause of death for children in the United States as well as the leading cause of head injury. Improved finite element models that integrate the correct shape of the pediatric brain with current injury prediction metrics would improve occupant response prediction for the pediatric occupant. The focus of this study is the improvement of geometric scaling factors for the brainstem to adapt current adult brain models to a pediatric model. The method used to assess shape change for this study was a geometric morphometric analysis technique. A sliding landmark form of a general Procrustes analysis was selected for its ability to compare curved structures with few true landmarks. The dataset consisted of fifty-nine individuals ranging in age from newborn to twenty-one years of age with groups specified at newborn, three months, six months, one year, three years, six years, ten years, fifteen years, and twenty-one years of age. Data was collected by outlining the structure on transverse and sagittal scans of magnetic resonance images and then creating a landmark dataset with a user-defined number of points for each individual. Once all individuals had the same number of landmarks, these points are allowed to slide on planes tangent to the surface until a value described as bending energy is minimized relative to an iteratively computed mean configuration from a Generalized Procrustes Analysis. A General Procrustes Analysis was completed for this data set to determine the shape differences between the age groups. Then, the coordinate locations were regressed onto age, and this analysis resulted in a model that predicted landmark locations based on age. From this model, the dimensions of the brainstem were calculated using the specified age groups. The final step was taking the dimensions of the predicted twenty-one year old model as the base and calculating a geometric scaling factor for shape, without including changes in size, for each age group. To assess the statistical significance of the process, permutation tests with 100,000 iterations were performed with resulting p-values of 0.17177 for a linear regression and 0.13467 for a quadratic regression of landmark location as a function of age.
