3D Multi-Modality Medical Image Registration Using Feature Space Clustering

Abstract

Abstract. In this paper, 3D voxel-similarity-based VB registration algorithms that optimize a feature-space clustering measure are proposed to combine the segmentation and registration process. We present a unifying de nition and a classi cation scheme for existing VB matching criteria and propose a new matching criterion: the entropy of the grey-level scatter-plot. This criterion requires no segmentation or feature extraction and no a priori knowledge of photometric model parameters. The effects of practical implementation issues concerning grey-level resampling, scatter-plot binning, parzen-windowing and resampling frequencies are discussed in detail and evaluated using real world data CT and MRI.

Non-linear Local Registration of Functional Data

“…-a global rigid method (GR): the registration by maximization of mutual information [9], [10]; -a global piecewise affine registration (PS): the Talairach Proportional Squaring [11]; -a non-linear global registration (NLG): a method based on optical flow and a robust optimization scheme [12]; -a local rigid method (LR): the local registration realized through the transformation described by the matrix M (see Sect. 2.1).…”

Fast CT/PET registration based on partial volume matching

“…Mutual information, originating in the information theory, is a voxel-based similarity measure of the statistical dependency between two datasets, which has been independently proposed by Collignon et al [3] and Viola and Wells [4]. It evaluates the amount of information that one variable contains about the other.…”

Reconstruction of 3D building models from aerial images and maps

“…The metric is based on a formulation of the mutual information between the object model and the images. In computer vision, mutual information has been used for relational matching (Vosselman, 1992) and for medical image registration (Viola and Wells, 1997;Collignon et al, 1995) among others.…”