Background and Summary

Computer simulations using a computational model of the human body enables the study of complex biomedical problems by allowing access to a three-dimensional (3D) information, which would be difficult to extract from experimental measurements. Such models are mainly used for medical device design problems that cover different study areas, such as, fluid dynamics, solid mechanics, electromagnetics, and thermal propagation. In electromagnetic studies, computational models can be used to study problems, such as, tissue thermal effects during electromagnetic exposure [1, 2, 3], distribution of electric current in the brain during therapeutic electrical brain stimulation (EBS)  [4, 5, 6], and safety of therapeutic devices during pregnancy [7].


There are more than thirty computational human models reported in the literature [7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29]. The models were developed either from ex vivo axial anatomical image data sets or from 3D medical imaging data. For example, the Visible Photographic Man (VIP-man) model [27], which has 31 head structures and voxel resolution of 4X4X4  mm^3, was developed from the cryosection images of the Visible Human Project (VHP) [30]. Similarly, the Chinese electromagnetic human head model (CMODEL) [22], which has 49 structures and voxel resolution of 0.16X0.16X0.5 mm^3, was built from the axial anatomical image data set of the Chinese Visible Human (CVH) [12,  13]. On the other hand, models that were built from 3D medical imaging data include: HUGO model [28] (15 head structures and voxel resolution of 1 mm^3) from the MRI and CT of VHP data set, MRI-based head model developed by Makris et al. [29] (49 head structures and voxel resolution of 1 mm^3), and the MIDA head and neck model [23] (153 structures and voxel resolution of 0.5 mm^3).


The majority of the models reported in the literature are voxel models; however, some of them are available in both the voxel and the surface mesh formats [8, 11, 16, 20, 21, 23]. Surface mesh models have the added advantage of being compatible with the standard finite element method (FEM) analysis, being deformable, and being more accurate at surface approximation compared to the staircase approximation in voxel models [31]. However, only a single model [21], from the available surface mesh models, was tested for FEM meshability. Recently, we used this model to build a 30-weeks pregnant female model that is also FEM meshable [7]. For FEM meshability, the surface meshed based model must have water-tight surfaces (no holes), 2-manifold mesh, and no self-intersecting faces. A mesh is 2-manifold if every node of the mesh has a disk-shaped neighborhood of triangles. Additionally, only few models are freely available to the scientific community [8, 19, 20, 21]. To the best of the authors' knowledge, currently, there is no computational human model that is available in both voxel and mesh formats, tested for FEM compatibility, and freely available. Within this context, we have built a computational human head model, which is FEM compatible and available freely in both voxel and surface mesh formats.


We have built a human head computational model, which was named "Menelik" after an Ethiopian emperor, by using the Visible Human Project (VHP) male data set, which is freely available from the U.S. National Library of Medicine. The VHP male data set contains a set of cryosection photographs, MRI, and CT images. The model was presented in both surface mesh and voxel (a resolution of 0.5 mm^3) formats. It contains 33 different anatomical structures that were indexed to 23 voxel regions based on their electromagnetic material properties. The Menelik model was tested and validated by simulating transcranial Direct Current Stimulation (tDCS) and computing the specific absorption rate (SAR) from the transmit coil of MRI. The results were compared and contrasted with the results obtained from using the MIDA model [23].