Modeling surgical loads to account for subsurface tissue deformation during stereotactic neurosurgery

摘要

Abstract: For more than a decade, surgical procedures have benefited significantly from the advent of OR (operating room) coregistered preoperative CT (computed tomographic) and MR (magnetic resonance) imaging. Despite advances in imaging and image registration, one of the most challenging problems is accounting for intraoperative tissue motion resulting from surgical loading conditions. Due to the considerable expense and cumbersome nature of intraoperative MR/CT scanners and the lack of high spatial definition of intracranial anatomy with ultrasound, we have elected to pursue a physics-based computational approach to account for tissue deformation in the context of frameless steroetactic neurosurgery. We have developed a computational model of the brain based on porous media physics and have begun to quantify subsurface deformation due to comparable surgical loads using an in vivo porcine model. Templates of CT-observable markers are implanted in a grid-like fashion in the pig brian to quantify tissue motion. Preliminary results based on the simplest of model assumptions are encouraging and have predicted displacement within 15% of measured values. In this paper, a series of computations is compared to experimental data to further understand the impact of material properties and pressure gradients within a homogeneous model of brain deformation. The results show that the best fits are obtained with Young's moduli and Poisson's ratio which are smaller than those values typically reported in the literature. As the Poisson ratio decreases towards 0.4 the corresponding Young's modulus increases towards the low end of the values contained in the literature. The optimal pressure gradient is found to be within physiological limits but generally higher than literature values would suggest for a given level of imparted loading, although differences between our experiments and those in the literature with respect to tissue loading conditions are noted. !39