Registration of 3D intraoperative MR images of the brain using a finite element biomechanical model

Abstract

We present a new algorithm for the non-rigid registration of 3D Magnetic Resonance (MR) intraoperative image sequences showing brain shift. The algorithm tracks key surfaces (cortical surface and the lateral ventricles) in the image sequence using an active surface algorithm. The volumetric deformation field of the objects the surfaces are embedded in is then inferred from the displacements at the boundary surfaces using a biomechanical finite element model of these objects. The biomechanical model allows us to analyse characteristics of the deformed tissues, such as stress measures. Initial experiments on an intraoperative sequence of brain shift show a good correlation of the internal brain structures after deformation using our algorithm, and a good capability of measuring surface as well as subsurface shift. We measured distances between landmarks in the deformed initial image and the corresponding landmarks in the target scan. The surface shift was recovered from up to 1cm down to less than 1mm, and subsurface shift from up to 6mm down to 3mm or less.