Comparison of functional thalamic segmentation from seed-based analysis and ICA

Joanne R. Hale, Stephen D. Mayhew, Karen Mullinger, Rebecca S. Wilson, Theodoros N. Arvanitis, S.T. Francis, Andrew P. Bagshaw

Research output: Contribution to journalArticlepeer-review

25 Citations (Scopus)
542 Downloads (Pure)


Information flow between the thalamus and cerebral cortex is a crucial component of adaptive brain function, but the details of thalamocortical interactions in human subjects remain unclear. The principal aim of this study was to evaluate the agreement between functional thalamic network patterns, derived using seed-based connectivity analysis and independent component analysis (ICA) applied separately to resting state functional MRI (fMRI) data from 21 healthy participants. For the seed-based analysis, functional thalamic parcellation was achieved by computing functional connectivity (FC) between thalamic voxels and a set of pre-defined cortical regions. Thalamus-constrained ICA provided an alternative parcellation. Both FC analyses demonstrated plausible and comparable group-level thalamic subdivisions, in agreement with previous work. Quantitative assessment of the spatial overlap between FC thalamic segmentations, and comparison of each to a histological “gold-standard” thalamic atlas and a structurally-defined thalamic atlas, highlighted variations between them and, most notably, differences with both histological and structural results. Whilst deeper understanding of thalamocortical connectivity rests upon identification of features common to multiple non-invasive neuroimaging techniques (e.g. FC, structural connectivity and anatomical localisation of individual-specific nuclei), this work sheds further light on the functional organisation of the thalamus and the varying sensitivities of complementary analyses to resolve it.
Original languageEnglish
Pages (from-to)448-465
Early online date17 Apr 2015
Publication statusPublished - 1 Jul 2015


  • Functional connectivity
  • Thalamocortical
  • Resting state
  • fMRI
  • ICA
  • Thalamus


Dive into the research topics of 'Comparison of functional thalamic segmentation from seed-based analysis and ICA'. Together they form a unique fingerprint.

Cite this