Joint Registration and Clustering of Images

M. Sabuncu, M. E. Shenton, P. Golland
miccai
Volume 4791, Pages 47-54
October, 2007

Download full paper

Abstract

We demonstrate an EM-based algorithm that jointly registers and clusters a group of images using an affine transformation model. The output is a small number of prototype images that represent the different modes of the population. The proposed algorithm can be viewed as a generalization of other well-known atlas construction algorithms, where the collection of prototypes represent multiple atlases for that population. Our experiments indicate that the employment of multiple atlases improves the localization of the underlying structure in a new subject.


Reference

Sabuncu M, Shenton ME, Golland P. Joint registration and clustering of images. In miccai, volume 4791 of Lecture Notes in Computer Science. Heidelberg, Germany: Springer, 2007;47-54.

Research area

registration
© 2013 Psychiatry Neuroimaging Laboratory | Last updated 04.15.2013