Quantifying Morphology-Related Deviations in Brain Strain Using an Automated Mesh Morphing Method

Finite element head models (FEHMs) have been widely used to study the biomechanics in traumatic brain injury (TBI). Most FEHMs are constructed to reflect the average head shape, which inevitably leads to the omission of individual brain morphology. In this study, we developed an automated mesh morphing method based on radial basis function-thin plate spline (RBF-TPS) with automated landmark extraction and projection. Five representative subject-specific head models and the baseline model were subjected to the head kinematics from six different datasets covering diverse impact scenarios. Results showed that morphology-related deviations increase with loading severity, reaching up to 0.21 for MPS95 and 0.14 s^-1 for MPSR95 (defined as the peak values of 95th percentile of maximum principal strain and strain rates, respectively). Based on logistic regression, TBI risk thresholds varied by approximately 19.4 % for MPS95 and 11.4 % for MPSR95 across the representative models. It was found that the strain and strain-rate magnitudes generally increased with brain volume. However, exceptions were observed where smaller brains exhibited higher MPS95 and MPSR95 across all models. Moreover, MPS95 exhibits an approximately linear relationship between deviation and the mean value, whereas MPSR95 shows no comparable trend, suggesting more complex mechanisms. These findings indicate that the influence of subject-specific morphology on strain response cannot be fully explained by size scaling alone, underscoring the importance of incorporating individual morphological characteristics into brain injury prediction models. Our automated mesh morphing method is openly available at: https://github.com/Yuzhe-Liu-Lab/Brain-RBF .