Clustering probabilistic tractograms using independent component analysis applied to the thalamus

The connectivity information contained in diffusion tensor imaging (DTI) has previously been used to parcellate cortical and subcortical regions based on their connectivity profiles. The aim of the current study is to investigate the utility of a novel approach to connectivity based parcellation of...

Full description

Saved in:
Bibliographic Details
Published in:NeuroImage (Orlando, Fla.) Fla.), 2011-02, Vol.54 (3), p.2020-2032
Main Authors: O'Muircheartaigh, Jonathan, Vollmar, Christian, Traynor, Catherine, Barker, Gareth J., Kumari, Veena, Symms, Mark R., Thompson, Pam, Duncan, John S., Koepp, Matthias J., Richardson, Mark P.
Format: Article
Language:English
Subjects:
Adult
Biomedical research
Brain
Brain Mapping
Cluster Analysis
Diffusion Tensor Imaging
Female
Functional Laterality - physiology
Humans
Male
Medical research
Methods
Middle Aged
Multivariate analysis
NMR
Nuclear magnetic resonance
Principal Component Analysis
Probability
Spinothalamic Tracts - anatomy & histology
Studies
Thalamus - anatomy & histology
Young Adult
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The connectivity information contained in diffusion tensor imaging (DTI) has previously been used to parcellate cortical and subcortical regions based on their connectivity profiles. The aim of the current study is to investigate the utility of a novel approach to connectivity based parcellation of the thalamus using probabilistic tractography and independent component analysis (ICA). We use ICA to identify spatially coherent tractograms as well as their underlying seed regions, in a single step. We compare this to seed-based tractography results and to an established and reliable approach to parcellating the thalamus based on the dominant cortical connection from each thalamic voxel (Behrens et al., 2003a,b). The ICA approach identifies thalamo-cortical pathways that correspond to known anatomical connections, as well as parcellating the underlying thalamus in a spatially similar way to the connectivity based parcellation. We believe that the use of such a multivariate method to interpret the complex datasets created by probabilistic tractography may be better suited than other approaches to parcellating brain regions. ► Independent component analysis used to segment the thalamus ► Promising technique to parcel out consistent fibre tracts ► Independent components consistent with tractography from the same region ► Independent components consistent with other clustering methods ► Multivariate method applied to segment probabilistic tractography
ISSN:1053-8119
1095-9572
DOI:10.1016/j.neuroimage.2010.09.054