A Fine-grained Cortical and Subcortical Parcellation with High Flexibility and Adaptability
Poster No:
1804
Submission Type:
Abstract Submission
Authors:
Tongyu Zhang1, Ang Li1, Yingjie Peng1, Shangzheng Huang1
Institutions:
1State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, Beijing, China
First Author:
Tongyu Zhang
State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences
Beijing, China
State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences
Beijing, China
Co-Author(s):
Ang Li
State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences
Beijing, China
State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences
Beijing, China
Yingjie Peng
State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences
Beijing, China
State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences
Beijing, China
Shangzheng Huang
State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences
Beijing, China
State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences
Beijing, China
Introduction:
Various brain atlases are widely used in neuroscience to define regional boundaries. These atlases can be parcellated based on structural, functional, or multimodal information, covering tens to hundreds of regions (Lawrence, 2021). Discrepancies on region boundaries have led to a long-standing problem: the challenge in mapping regional measures or connectivity across atlases, which hindered the generalization and comparison of findings. Previous studies have attempted cortical subdivisions (Prieto, 2024); however, existing atlases are insufficient in granularity, while vertex- or voxel-level computations require excessive time and storage costs. Hence, we propose a fine-grained parcellation composed of nearly 9,000 sub-centimeter regions, enabling flexible cross-atlas mapping through region aggregation while ensuring efficiency for fine-scale analyses.
Methods:
For cortex, we subdivided the Desikan-Killiany (DK) atlas (Desikan, 2006) based on the symmetric fs_LR_164k surface, to preserve both anatomical boundaries and topological properties. Specifically, for each DK region, we first computed the pair-wise Euclidean distances between vertices on the inflated mesh, and constructed a distance matrix. Then, an improved Equal-Size Spectral Clustering method (Anamabo, 2023) was employed for subregion parcellation, balancing each subcluster by adjusting labels of boundary neighbors. Finally, we obtained 7,340 subregions (3,670 for each hemisphere), with an average area of 15 mm². For ease of use, the derived dense parcellation was also mapped to the fsaverage and MNI152 spaces. (Fig. 1a)
For subcortex, we selected the Brainnetome (BN) atlas (Fan, 2016) as a reference. Analogously, we performed clustering based on the voxel coordinates in MNI152 1mm space, resulting in 1,614 clusters (807 for each hemisphere), each with a volume of approximately 54 mm³. (Fig. 1b)
Finally, we constructed a highly fine-grained atlas with 8,954 regions, which can be reaggregated into existing atlases based on the overlap relationships between regions. Fig. 1c and 1d visualize the cortical parcellation in surface and volume spaces. Fig. 1e show the subcortical regions. Atlas resources and label mapping tables are available at: https://github.com/tyzhang97/Fine-grained-Brain-Atlas.
For subcortex, we selected the Brainnetome (BN) atlas (Fan, 2016) as a reference. Analogously, we performed clustering based on the voxel coordinates in MNI152 1mm space, resulting in 1,614 clusters (807 for each hemisphere), each with a volume of approximately 54 mm³. (Fig. 1b)
Finally, we constructed a highly fine-grained atlas with 8,954 regions, which can be reaggregated into existing atlases based on the overlap relationships between regions. Fig. 1c and 1d visualize the cortical parcellation in surface and volume spaces. Fig. 1e show the subcortical regions. Atlas resources and label mapping tables are available at: https://github.com/tyzhang97/Fine-grained-Brain-Atlas.
·Fig. 1. Fine-grained brain parcellation.
Results:
As validation, we computed the Dice similarity coefficient between the reaggregated atlas and the ground truth target atlases, including AAL (Tzourio-Mazoyer, 2002), (Glasser, 2016), (Schaefer, 2018) and Tianye (Tian, 2020), etc. The merged regions show a high degree of overlap with the ground truth, with an average Dice coefficient ranging from 0.81 to 0.97 (Fig. 2a and 2b). This demonstrates the reliability of spatial aggregation, indicating that structural phenotypes at different scales can be obtained by integrating measurements from fine-grained regions.
For functional data, we randomly selected 100 unrelated individuals from the HCP-YA dataset (Van, 2013). Raw fMRI time series from fine-grained regions were averaged according to the mapping relationship and then we calculated the Pearson correlation between the averaged time series and ground truth time series extracted from coarse atlases. Results of the two methods are highly and significantly correlated, with average correlation coefficients over 0.9 (Fig. 2c and 2d). Additionally, regarding the extraction of whole-brain time series, the proposed fine-grained atlas can reduce 86% of time cost for a single subject (4 runs) compared to the total time required for 12 other atlases separately (Fig. 2e). Thus, performing computation at a fine scale once can reliably fit the results of other atlases, enabling efficient mapping across atlases.
For functional data, we randomly selected 100 unrelated individuals from the HCP-YA dataset (Van, 2013). Raw fMRI time series from fine-grained regions were averaged according to the mapping relationship and then we calculated the Pearson correlation between the averaged time series and ground truth time series extracted from coarse atlases. Results of the two methods are highly and significantly correlated, with average correlation coefficients over 0.9 (Fig. 2c and 2d). Additionally, regarding the extraction of whole-brain time series, the proposed fine-grained atlas can reduce 86% of time cost for a single subject (4 runs) compared to the total time required for 12 other atlases separately (Fig. 2e). Thus, performing computation at a fine scale once can reliably fit the results of other atlases, enabling efficient mapping across atlases.
·Fig. 2. Validation of the fine-grained parcellation.
Conclusions:
We propose a novel fine-grained brain parcellation with high flexibility and adaptability. It provides fundamental units for analyzing structural, functional, or other multimodal data, facilitating the mapping across atlases or supporting any scenario requiring fine-scale region localization.
Modeling and Analysis Methods:
Segmentation and Parcellation 2
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Cortical Anatomy and Brain Mapping
Subcortical Structures
Neuroanatomy Other
Neuroinformatics and Data Sharing:
Brain Atlases 1
Keywords:
Atlasing
Cortex
FUNCTIONAL MRI
MRI
STRUCTURAL MRI
Sub-Cortical
Other - Brain Parcellation, Fine-grained, Spectral Clustering, Cross-atlas Mapping
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
2. Desikan, R. S., Ségonne, F., Fischl, B., Quinn, B. T., Dickerson, B. C., Blacker, D., ... & Killiany, R. J. (2006). An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage, 31(3), 968-980.
3. Fan, L., Li, H., Zhuo, J., Zhang, Y., Wang, J., Chen, L., ... & Jiang, T. (2016). The human brainnetome atlas: a new brain atlas based on connectional architecture. Cerebral cortex, 26(8), 3508-3526.
4. Glasser, M. F., Coalson, T. S., Robinson, E. C., Hacker, C. D., Harwell, J., Yacoub, E., ... & Van Essen, D. C. (2016). A multi-modal parcellation of human cerebral cortex. Nature, 536(7615), 171-178.
5. Lawrence, R. M., Bridgeford, E. W., Myers, P. E., Arvapalli, G. C., Ramachandran, S. C., Pisner, D. A., ... & Vogelstein, J. T. (2021). Standardizing human brain parcellations. Scientific data, 8(1), 78.
6. Prieto, Y., Molina, J., Otero, M., Hernández, C., Poupon, C., Mangin, J.-F., El-Deredy, W., & Guevara, P. (2024, June). Multiscale subdivision of an anatomical cortical parcellation based on geodesic distance. Organization for Human Brain Mapping Annual Meeting (OHBM), Seoul, South Korea. https://ww6.aievolution.com/hbm2401/Abstracts/viewAbs?abs=4063
7. Schaefer, A., Kong, R., Gordon, E. M., Laumann, T. O., Zuo, X. N., Holmes, A. J., ... & Yeo, B. T. (2018). Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI. Cerebral cortex, 28(9), 3095-3114.
8. Tzourio-Mazoyer, N., Landeau, B., Papathanassiou, D., Crivello, F., Etard, O., Delcroix, N., ... & Joliot, M. (2002). Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage, 15(1), 273-289.
9. Tian, Y., Margulies, D. S., Breakspear, M., & Zalesky, A. (2020). Topographic organization of the human subcortex unveiled with functional connectivity gradients. Nature neuroscience, 23(11), 1421-1432.
10. Van Essen, D. C., Smith, S. M., Barch, D. M., Behrens, T. E., Yacoub, E., Ugurbil, K., & Wu-Minn HCP Consortium. (2013). The WU-Minn human connectome project: an overview. Neuroimage, 80, 62-79.