The NIH/NIGMS
Center for Integrative Biomedical Computing

SCI Publications

2012


M. Datar, P. Muralidharan, A. Kumar, S. Gouttard, J. Piven, G. Gerig, R.T. Whitaker, P.T. Fletcher. “Mixed-Effects Shape Models for Estimating Longitudinal Changes in Anatomy,” In Spatio-temporal Image Analysis for Longitudinal and Time-Series Image Data, Lecture Notes in Computer Science, Vol. 7570, Springer Berlin / Heidelberg, pp. 76--87. 2012.
ISBN: 978-3-642-33554-9
DOI: 10.1007/978-3-642-33555-6_7

ABSTRACT

In this paper, we propose a new method for longitudinal shape analysis that ts a linear mixed-e ects model, while simultaneously optimizing correspondences on a set of anatomical shapes. Shape changes are modeled in a hierarchical fashion, with the global population trend as a xed e ect and individual trends as random e ects. The statistical signi cance of the estimated trends are evaluated using speci cally designed permutation tests. We also develop a permutation test based on the Hotelling T2 statistic to compare the average shapes trends between two populations. We demonstrate the bene ts of our method on a synthetic example of longitudinal tori and data from a developmental neuroimaging study.

Keywords: Computer Science