A Biomechanical Model of Soft Tissue Deformation, with Applications to Non-rigid Registration of Brain Images with Tumor Pathology

The finite element method is applied to the biomechanics of brain tissue deformation. Emphasis is given to the deformations induced by the growth of tumors, and to the deformable registration of anatomical atlases with patient images. A uniform contraction of the tumor is first used to obtain an estimate of the shape of the brain prior to the growth of the tumor. A subsequent nonlinear regression method is used to improve on the above estimate. The resulting deformation mapping is finally applied to an atlas, yielding the registration of the atlas with the tumor-deformed anatomy. A preliminary 2D implementation that includes inhomogeneity and a nonlinear elastic material model is tested on simulated data as well as a patient image. The long-term scope of this work is its application to surgical planning systems.