Multisubject non-rigid registration of Brain MRI using intensity and geometric features

Abstract

In this article we merge point feature and intensity-based registration in a single algorithm to tackle the problem of multiple brain registration. Because of the high variabilityof the shape of the cortex across individuals, there exist geometrical ambiguities in the registration process that an intensityme asure alone is unable to solve. This problem can be tackled using anatomical knowledge. First, we automaticallysegment and label the whole set of the cortical sulci, with a non-parametric approach that enables the capture of their highlyv ariable shape and topology. Then, we develop a registration energy that merges intensity and feature point matching. Its minimization leads to a linear combination of a dense smooth vector field and radial basis functions. We use and process differentlythe bottom line of the sulci from its upper border, whose localization is even more variable across individuals. We show that the additional sulcal energyim proves the registration of the cortical sulci, while still keeping the transformation smooth and one-to-one.