An Anatomically Accurate Finite Element Brain Model: Development, Validation and Comparison to Existing Models

The objective of this study was two-fold. The first objective was to develop and validate a high resolution, anatomically accurate brain finite element (FE) model from the International Consortium for Brain Mapping (ICBM) brain atlas using a voxel-based mesh generation approach. The second objective was to quantitatively compare performance of six validated brain FE models in three validation conditions against localized brain motion data. The ABM was developed from the ICBM brain atlas by converting each voxel into an element using a custom code developed in MATLAB (Mazziotta et al. 1995, 2001). The brain material properties were optimized using a Latin hypercube design (LHD) method. The ABM was validated against three experimental cadaver tests conducted by Hardy et al. (2001; 2007) through FE simulation in LS-DYNA. The three experimental tests considered for validation were: C755-T2 (occipital impact), C383-T1 (frontal impact), and C291-T1 (parietal impact) (Hardy et al. 2001; Hardy 2007). The five additional FE models considered in the current study are the Simulated Injury Monitor (SIMon), the Global Human Body Models Consortium (GHBMC) head model, the Total Human Model for Safety (THUMS) head model, the Kungliga Tekniska Högskolan (KTH) model, and the Dartmouth Head Injury Model (DHIM) (Kleiven and von Holst 2002; Takhounts et al. 2003; Kimpara et al. 2006; Kleiven 2007; Mao et al. 2013; Ji et al. 2014a). Validation results for the SIMon, GHBMC, and THUMS models were also obtained through direct simulation in LS-DYNA. Results for the remaining models were obtained from published literature. To evaluate model performance, the error between experimental and predicted displacements was quantified using a relatively new metric called CORA (CORrelation and Analysis) (Gehre et al. 2009). The ABM shows good agreement with experimental validation data. Additionally, looking at each model’s average CORA score between the three impacts, the ABM scores the best CORA rating. This result indicates that of the models considered, the ABM demonstrates the strongest ability to predict local brain deformations under a range of impact severities and directions.