A reproducible MRI-to-FE workflow for generating population-specific and subject-specific finite element head models

ABSTRACT Purpose Finite element (FE) models are widely used to study traumatic brain injury, but most are based on single-subject scans or generic atlases that may not represent specific cohorts, introducing variability in biomechanical outputs. This study presents a reproducible framework for generating a cohort-specific MRI-derived template brain and constructing a representative FE head model. Methods A cohort-specific template was created from T1-weighted and diffusion-weighted MRI scans of collision sport athletes (n = 78). An FE head model was then generated from this template, with material properties mapped using MRI-derived microstructural information. The anatomical accuracy of the template was evaluated using deformation-based morphometry against standard brain atlases. The FE model was validated against cadaveric head impact experiments. A subject-specific FE model was also developed to test the pipeline’s applicability and model performance. Results Registration of an external non-collision sport controls (n = 20) to the template showed reduced deformation compared with registration to the MNI152 and an existing adolescent atlas, indicating improved cohort representation. Validation against cadaveric data demonstrated small to moderate correlations between simulated and experimental nodal displacements. Subject-specific and fiber dispersion-informed models showed significantly different strain distributions compared with isotropic models under low-intensity impacts, but no significant differences under high-intensity impacts. Conclusion The proposed framework enables reproducible construction of cohort-specific FE brain models using open-access tools. Results highlight the importance of subject-and cohort-specific modelling for accurately characterizing brain biomechanics. This approach improves the anatomical and microstructural realism of FE models and supports more precise mapping of mechanical strain to clinical outcomes in TBI. STATEMENTS AND DECLARATIONS We declare that we have no financial or personal relationships with individuals, companies, or organisations that could inappropriately influence the content of this work. We also confirm that we hold no professional or personal interests in any product, service, or company that could be perceived as affecting the content and quality of this manuscript.