Contrast-invariant registration of cardiac and renal MR perfusion images

Abstract

Automatic registration of dynamic MR perfusion images is a challenging task due to the rapid changes of the image contrast caused by the wash-in and wash-out of the contrast agent. In this paper we introduce a contrast-invariant similarity metric and propose a common framework to perform affine registration on both cardiac and renal MR perfusion images. First, large-scale translational motion is identified by tracking a selected region of interest with integer pixel shifts. Then, we estimate the affine transformation of the organ for each frame. We have tested the proposed algorithm on real cardiac and renal MR perfusion scans and obtained encouraging registration results. © Springer-Verlag Berlin Heidelberg 2004.