Probabilistic Clustering and Shape Modelling of White Matter Fibre Bundles using Regression Mixtures

Ratnarajah, Nagulan and Simmons, Andy and Davydov, Oleg and Hojjatoleslami, Ali (2011) Probabilistic Clustering and Shape Modelling of White Matter Fibre Bundles using Regression Mixtures. In: MICCAI 2011, 18-22 September 2011, Toronto, Canada. (Unpublished) (Full text available)

PDF
Download (479kB)
[img]
Preview

Abstract

We present a novel approach for probabilistic clustering of white matter fibre pathways using curve-based regression mixture modelling techniques in 3D curve space. The clustering algorithm is based on a principled method for probabilistic modelling of a set of fibre trajectories as individual sequences of points generated from a finite mixture model consisting of multivariate polynomial regression model components. Unsupervised learning is carried out using maximum likelihood principles. Specifically, conditional mixture is used together with expectation-maximisation (EM) algorithm to estimate cluster membership. The result of clustering is the probabilistic assignment of fibre trajectories to each cluster and an estimate of the cluster parameters. A statistical model is calculated for each clustered fibre bundles using fitted parameters of the probabilistic clustering. We illustrate the potential of our clustering approach on synthetic data and real data.

Item Type: Conference or workshop item (Paper)
Subjects: R Medicine > RC Internal medicine > RC0321 Neuroscience. Biological psychiatry. Neuropsychiatry
Q Science > QA Mathematics (inc Computing science) > QA801 Analytic mechanics
Q Science > QA Mathematics (inc Computing science) > QA273 Probabilities
Q Science > QA Mathematics (inc Computing science) > QA440 Geometry > QA611 Topology
Q Science > QM Human anatomy
Divisions: Faculties > Science Technology and Medical Studies > School of Biosciences > Biomedical Research Group
Depositing User: Sayed Ali Hojjatoleslami
Date Deposited: 12 May 2011 17:10
Last Modified: 22 May 2014 09:53
Resource URI: http://kar.kent.ac.uk/id/eprint/27765 (The current URI for this page, for reference purposes)
  • Depositors only (login required):

Downloads

Downloads per month over past year