Dual-HINet: Dual Hierarchical Integration Network of Multigraphs for Connectional Brain Template Learning

Duran F. S., Beyaz A., Rekık I.

25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Singapore, Singapore, 18 - 22 September 2022, vol.13431, pp.305-314 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 13431
  • Doi Number: 10.1007/978-3-031-16431-6_29
  • City: Singapore
  • Country: Singapore
  • Page Numbers: pp.305-314
  • Keywords: Brain multigraph population, Connectional brain templates, Graph neural networks, Hierarchical multigraph embedding
  • Istanbul Technical University Affiliated: Yes


A connectional brain template (CBT) is a normalized representation of a population of brain multigraphs, where two anatomical regions of interests (ROIs) are connected by multiple edges. Each edge captures a particular type of interaction between pairs of ROIs (e.g., structural/functional). Learning a well-centered and representative CBT of a particular brain multigraph population (e.g., healthy or atypical) is a means of modeling complex and varying ROI interactions in a holistic manner. Existing methods generate CBTs by locally integrating heterogeneous multi-edge attributes (e.g., weights and features). However, such methods are agnostic to brain network modularity as they ignore the hierarchical structure of neural interactions. Furthermore, they only perform node-level integration at the individual level without learning the multigraph representation at the group level in a layer-wise manner. To address these limitations, we propose Dual Hierarchical Integration Network (Dual-HINet) for connectional brain template estimation, which simultaneously learns the node-level and cluster-level integration processes using a dual graph neural network architecture. We also propose a novel loss objective to jointly learn the clustering assignment across different edge types and the centered CBT representation of the population multigraphs. Our Dual-HINet significantly outperforms state-of-the-art methods for learning CBTs on a large-scale multigraph connectomic datasets. Our source code can be found at https://github.com/basiralab/Dual-HINet.