Unbiased validation of connectivity-based cohesive brain parcellation
Poster No:
2011
Submission Type:
Abstract Submission
Authors:
Ajay Nemani1, Mark Lowe1
Institutions:
1The Cleveland Clinic, Cleveland, OH
First Author:
Co-Author:
Introduction:
We have previously introduced cohesive parcellation, a functional connectivity-based approach that was shown to be optimal for downstream, exemplar-based network analysis1. However, this results in many more parcels than traditional parcellations. Most connectivity-based metrics of parcellations are sensitive to the size and number of parcels, complicating direct comparison. Recently, the distance-controlled boundary coeffcient (DCBC) was introduced, an unbiased, parcel resolution invariant measure incorporating both connectivity and distance2. We evaluate cohesive parcellation using DCBC and compare to other parcellations.
Methods:
18 healthy subjects (age 20-36, 8 female) were scanned on a 7T Siemens Magnetom (Erlangen, Germany) with a 32 channel Nova head coil (Massachusetts, USA). Whole-brain, rsfMRI were acquired (SMS multiband=3, 81 1.5mm slices, 128+4 volumes, TE/TR=21/2800 ms, 70° flip, FOV=192 mm2, resolution=1.2x1.2x1.5 mm3). T1w images were acquired for anatomical context as well as duel-echo B0 field maps. rsfMRI data were corrected for B0 distortions, slice timing, motion3, and physiologically-based nuisances4. Subject-specific cohesive parcellation was performed based on Pearson correlations of the grey matter rsfMRI data using a 0.5 cohesion threshold1. Cohesion, homogeneity, and Silhouette were calculated at the voxelwise level, grouped by parcel, and a parcel size-weighted average determined. For DCBC, a distance-based graph was assembled from the grey matter voxels based on spatial adjacency and corrected for gyral and sulcal folds using FreeSurfer-derived white and pial surface reconstructions. For each subject, DCBC was calculated for each parcel using voxel-to-voxel correlations and geodesic distances, with a final parcel size-weighted average extracted as above2.
Results:
Cohesive parcellation and associated parcel-based Silhouette and DCBC distributions are shown for a representative subject (figure 1, 2530 parcels; 441-3579 across all subjects). The Silhouette and DCBC for this subject was 0.329 and 0.0973, respectively. The Silhouette coefficient shows strong dependence on parcel size, while the DCBC is largely independent, except in the smallest parcels. The corresponding distributions for all measures are shown for all subjects (figure 2). Over all subjects, the size-weighted averages were 0.520±0.0038, 0.484±0.0038, 0.292±0.033, 0.0845±0.012 for cohesion, homogeneity, silhouette, and DCBC, respectively. For cohesion, homogeneity, and silhouette, these results are consistent with previous findings1. The DCBC for cohesive parcellation compares favorably to previously published DCBC results across a number of traditional parcellations, including the Yeo5 (0.0213), Power6 (0.0261), and Gordon7 (0.0236) parcellations2.
Conclusions:
Cohesive parcellation has been shown to compare favorably to other connectivity-based parcellations across several measures at both the individual subject1 and group8 levels while providing optimal parcel exemplars for downstream brain modelling. Because cohesive parcellation is focused on generating optimal exemplars, it produces higher parcel counts, complicating evaluation with traditional metrics. The DCBC metric is invariant to parcel count, which showed that cohesive parcellation performs better than other data-driven functional parcellations. Connectivity-based brain modeling is based on exemplars derived from simple averages of underlying voxel members. Incorporating spatial context into network analysis may improve these models9,10. Cohesive parcellation utilizes a flexible hierarchical platform for optimization. Future work will look to incorporate spatial context into this framework in a DCBC-like manner in order to further optimize parcels for downstream brain modeling.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 2
fMRI Connectivity and Network Modeling
Segmentation and Parcellation 1
Task-Independent and Resting-State Analysis
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
Data analysis
FUNCTIONAL MRI
HIGH FIELD MR
Modeling
Other - Parcellation
1|2Indicates the priority used for review
Provide references using author date format
2. Zhi, D. (2022), 'Evaluating brain parcellations using the distance-controlled boundary coefficient', Human Brain Mapping, vol. 43, pp. 3706-3720.
3. Beall, E.B. (2014), 'SimPACE: generating simulated motion corrupted BOLD data with synthetic-navigated acquisition for the development and evaluation of SLOMOCO: a new, highly effective slicewise motion correction', Neuroimage, vol. 101, pp. 21-34.
4. Beall, E.B. (2007), 'Isolating physiologic noise sources with independently determined spatial measures', Neuroimage, vol. 37, no. 4, pp. 1286-1300.
5. Yeo, B.T.T. (2011), 'The organization of the human cerebral cortex estimated by intrinsic functional connectivity', Journal of Neurophysiology, vol. 106, pp. 1125-1165.
6. Power, J. (2011), 'Functional network organization of the human brain', Neuron, vol. 72, no. 4, pp. 665-678.
7. Gordon, E.M. (2016), 'Generation and evaluation of a cortical area parcellation from resting-state correlations', Cerebral Cortex, vol. 26, no. 1, pp. 288-303.
8. Nemani, A. (2021), 'Group cohesive parcellation results in superior functional-based parcellation with greater parcel-level parsimony than current approaches', Proceedings of the 30th annual meeting of the International Society of Magnetic Resonance in Medicine, #3140.
9. Zalesky, A. (2012), 'On the use of correlation as a measure of network connectivity', NeuroImage, vol. 60, pp. 2096-2106.
10. Welton, T. (2015), 'Reproducibility of graph-theoretic brain network metrics: a systematic review', Brain Connectivity, vol. 5, pp. 193-202.