Modeling of brain tissue retraction using intraoperative data

Abstract

We present a method for modeling tissue retraction during image-guided neurosurgery. A poroelastic brain model is driven by the stereoscopically- measured motion of a retractor to produce a full volume displacement field, which is used to update the preoperative MR images. Using the cortical surface surrounding the retractor as an independent evaluative landmark, we show that our approach is capable of capturing approximately 75% of the cortical deformation during tissue retraction. © Springer-Verlag Berlin Heidelberg 2004.