Tumor Treating Fields (TTFields) are electric fields at an intermediate frequency approved for treatment of Glioblastoma Multiforme. Understanding how TTFields distribution in the brain influences disease progression can be studied using numerical simulations. Creation of computational patient models involves accurate segmentation of patient MRIs, a task that cannot be performed automatically, and is therefore time-consuming. We present a method for rapidly creating patient head models using a healthy head model as a deformable template. The method is robust even when MRI data quality is low. It is enabling a study correlating the spatial distribution of TTFields and patient outcome.