Image-guided surgery has been a standard technique in the field of neurosurgery. However, brain shift throughout the surgical procedure has been a major issue affecting the spatial accuracy of conventional neuronavigation system. This phenomenon is not negligible in the currently available navigation system because the imaging data used in such system is based on the preoperative data. This study aims at developing the anatomically detailed three-dimensional finite element brain model and demonstrating the prediction capability of gravity-induced brain shift using the developed finite element brain model without additional acquisition of intraoperative imaging data. Although precise anatomical structure such as gyri and sulci on the surface, falx, tentorium cerebelli and ventricle may influence the brain tissue deformation as a mechanical boundary condition of the deformation field, most of the previously published brain finite element models did not include these anatomies. In this study, the patient-specific finite element brain model with brain substructures, falx cerebri with tentorium, as anatomical constraint was developed by manual segmentation of magnetic resonance imaging, and the brain model coregistered with rigid skull model segmented by computed tomography images. The model of brain parenchyma consisted of 253, 278tetrahedral elements whose material property was hyper-viscoelastic model presented in the literature. The computation of the gravity-induced brain shift by assuming that the patient was positioned lateral (left fronto-temporal craniotomy) was conducted using ABAQUS/Explicit. Deformation of the brain surface due to gravity after dura opening was demonstrated in 3D view with color-coded surface rendering and good agreement with the clinically experienced brain shift reported in the literature. The deformed gyri and sulc i on the surface also displayed with keeping high spatial resolution. The deep structure exhibited smaller amount of shift than the surface structure. The illustrative results successfully demonstrated the gravity-induced brain shift after left front-temporal craniotomy in three-dimension using the developed model.
