Multi-Level Parcellation of the Cerebral Cortex Using Resting-State fMRI
Cortical parcellation is one of the core steps for identifying the functional architecture of the human brain. Despite the increasing number of attempts at developing parcellation algorithms using resting-state fMRI, there still remain challenges to be ov
- PDF / 1,557,631 Bytes
- 8 Pages / 439.363 x 666.131 pts Page_size
- 108 Downloads / 185 Views
Abstract. Cortical parcellation is one of the core steps for identifying the functional architecture of the human brain. Despite the increasing number of attempts at developing parcellation algorithms using restingstate fMRI, there still remain challenges to be overcome, such as generating reproducible parcellations at both single-subject and group levels, while sub-dividing the cortex into functionally homogeneous parcels. To address these challenges, we propose a three-layer parcellation framework which deploys a different clustering strategy at each layer. Initially, the cortical vertices are clustered into a relatively large number of supervertices, which constitutes a high-level abstraction of the rs-fMRI data. These supervertices are combined into a tree of hierarchical clusters to generate individual subject parcellations, which are, in turn, used to compute a groupwise parcellation in order to represent the whole population. Using data collected as part of the Human Connectome Project from 100 healthy subjects, we show that our algorithm segregates the cortex into distinctive parcels at different resolutions with high reproducibility and functional homogeneity at both single-subject and group levels, therefore can be reliably used for network analysis.
1
Introduction
Parcellation of the cerebral cortex constitutes one of the core steps to reveal the functional organization of the brain. It is usually followed by network analyses devised to generate graphical models of the connections between the parcellated regions. Such analyses have the potential to uncover the neural mechanisms behind the human behavior and to help understand neurological disorders [8]. It is of great importance to obtain reliable parcellations, since errors at this stage propagate into the subsequent analysis and consequently affect the final results. Among others, there are two notable attributes that define “reliability” in the context of a cortical parcellation: 1) parcellated sub-regions should be functionally consistent and comprise similar vertices, since network nodes are typically represented by a single entity (such as the average time series of the constituent vertices) and 2) both individual subject and groupwise parcellations should be reproducible to some extent, that is, multiple parcellations obtained from different datasets of the same subject as well as groupwise parcellations computed from the subsets of the same population should exhibit functional and structural similarity. c Springer International Publishing Switzerland 2015 N. Navab et al. (Eds.): MICCAI 2015, Part III, LNCS 9351, pp. 47–54, 2015. DOI: 10.1007/978-3-319-24574-4_6
48
S. Arslan and D. Rueckert
With this motivation, we propose a whole-cortex parcellation framework based on resting-state functional magnetic resonance imaging (rs-fMRI). The brain is still functional in the absence of external stimuli, thus rs-fMRI time series can be utilized to parcellate the cortical surface into functionally homogeneous sub-regions. The majority of the literature o
Data Loading...
