Heterogeneous Biomechanical Model on Correcting Brain Deformation Induced by Tumor Resection-camera-ready

In this paper we present a non-rigid registration method to align pre-operative MRI (preMRI) with resected intra-operative MRI (iMRI) to compensate for brain deformation during tumor resection. This method formulates the registration as a three-variable (point correspondence, deformation field and resection region) functional minimization problem, in which point correspondence is represented by a fuzzy assign matrix, deformation field is represented by a piecewise linear function regularized by the strain energy of a heterogeneous biomechanical model, and resection region is represented by a maximal connected tetrahedral mesh. A Nested Expectation and Maximization framework is developed to simultaneously resolve these three variables. This method accommodates a heterogeneous biomechanical model as the regularization term to realistically describe the underlying deformation field, and allows the removal of the tetrahedra from the model to simulate the tumor resection. A simple two tissue heterogeneous model (ventricle plus the rest of the brain) is used to evaluate this method on 14 clinical cases. The experimental results show the effectiveness of this method in correcting the deformation induced by resection. The comparison between the homogeneous model and the heterogeneous model demonstrates the statistical significance of the improvement brought by the heterogeneous model (P-value 0.04).