TY - GEN
T1 - Intrinsic Patch-Based Cortical Anatomical Parcellation Using Graph Convolutional Neural Network on Surface Manifold
AU - Wu, Zhengwang
AU - Zhao, Fenqiang
AU - Xia, Jing
AU - Wang, Li
AU - Lin, Weili
AU - Gilmore, John H.
AU - Li, Gang
AU - Shen, Dinggang
N1 - Funding Information:
Acknowledgements. This work was partially supported by NIH grants (MH107815, MH116225, and MH117943).
Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - Automatic parcellation of cortical surfaces into anatomically meaningful regions of interest (ROIs) is of great importance in brain analysis. Due to the complex shape of the convoluted cerebral cortex, conventional methods generally require three steps to obtain the parcellations. First, the original cortical surface is iteratively inflated and mapped onto a spherical surface with minimal metric distortion, for providing a simpler shape for analysis. Then, a registration or learning-based labeling method is adopted to parcellate ROIs on the mapped spherical surface. Finally, parcellation labels on the spherical surface are mapped back to the original cortical surface. Despite great success, spherical mapping of the original cortical surface is inherently sensitive to topological noise and cannot deal with the impaired brains that violate spherical topology. To address these issues, in this paper, we propose to directly parcellate the cerebral cortex on the original cortical surface manifold without requiring spherical mapping, by leveraging the strong learning ability of the graph convolutional neural networks. Also, we extend the convolution to the surface manifold using the kernel strategy, which enables us to overcome the notorious shape difference issue (e.g., different vertex number and connections) across different subjects. Our method aims to learn the highly nonlinear mapping between cortical attribute patterns (on local intrinsic surface patches) and parcellation labels. We have validated our method on a normal cortical surface dataset and a synthetic dataset with impaired brains, which shows that our method achieves comparable accuracy to the methods using spherical mapping, and works well on cortical surfaces violating the spherical topology.
AB - Automatic parcellation of cortical surfaces into anatomically meaningful regions of interest (ROIs) is of great importance in brain analysis. Due to the complex shape of the convoluted cerebral cortex, conventional methods generally require three steps to obtain the parcellations. First, the original cortical surface is iteratively inflated and mapped onto a spherical surface with minimal metric distortion, for providing a simpler shape for analysis. Then, a registration or learning-based labeling method is adopted to parcellate ROIs on the mapped spherical surface. Finally, parcellation labels on the spherical surface are mapped back to the original cortical surface. Despite great success, spherical mapping of the original cortical surface is inherently sensitive to topological noise and cannot deal with the impaired brains that violate spherical topology. To address these issues, in this paper, we propose to directly parcellate the cerebral cortex on the original cortical surface manifold without requiring spherical mapping, by leveraging the strong learning ability of the graph convolutional neural networks. Also, we extend the convolution to the surface manifold using the kernel strategy, which enables us to overcome the notorious shape difference issue (e.g., different vertex number and connections) across different subjects. Our method aims to learn the highly nonlinear mapping between cortical attribute patterns (on local intrinsic surface patches) and parcellation labels. We have validated our method on a normal cortical surface dataset and a synthetic dataset with impaired brains, which shows that our method achieves comparable accuracy to the methods using spherical mapping, and works well on cortical surfaces violating the spherical topology.
KW - Cortical surface parcellation
KW - Graph CNN
UR - http://www.scopus.com/inward/record.url?scp=85075694436&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-32248-9_55
DO - 10.1007/978-3-030-32248-9_55
M3 - Conference contribution
AN - SCOPUS:85075694436
SN - 9783030322472
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 492
EP - 500
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings
A2 - Shen, Dinggang
A2 - Yap, Pew-Thian
A2 - Liu, Tianming
A2 - Peters, Terry M.
A2 - Khan, Ali
A2 - Staib, Lawrence H.
A2 - Essert, Caroline
A2 - Zhou, Sean
PB - Springer Science and Business Media Deutschland GmbH
T2 - 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Y2 - 13 October 2019 through 17 October 2019
ER -