Brain Network Transformer
Трансформер для сетей мозга
2022-01-01
SCID: 54.1/shrwdemf
Discuss with AI
ABIDE datasetBrain Network Transformerbrain network analysisorthonormal clustering readoutpair-wise connection strengths
Figures from the paper
Abstract (AI)
Human brains are commonly modeled as networks of Regions of Interest (ROIs) and their connections for the understanding of brain functions and mental disorders. Recently, Transformer-based models have been studied over different types of data, including graphs, shown to bring performance gains widely. In this work, we study Transformer-based models for brain network analysis. Driven by the unique properties of data, we model brain networks as graphs with nodes of fixed size and order, which allows us to (1) use connection profiles as node features to provide natural and low-cost positional information and (2) learn pair-wise connection strengths among ROIs with efficient attention weights across individuals that are predictive towards downstream analysis tasks. Moreover, we propose an Orthonormal Clustering Readout operation based on self-supervised soft clustering and orthonormal projection. This design accounts for the underlying functional modules that determine similar behaviors among groups of ROIs, leading to distinguishable cluster-aware node embeddings and informative graph embeddings. Finally, we re-standardize the evaluation pipeline on the only one publicly available large-scale brain network dataset of ABIDE, to enable meaningful comparison of different models. Experiment results show clear improvements of our proposed Brain Network Transformer on both the public ABIDE and our restricted ABCD datasets. The implementation is available at https://github.com/Wayfear/BrainNetworkTransformer.
Key Findings
1
Efficient attention weights learn pairwise ROI connection strengths across individuals and provide representations predictive of downstream brain-network analysis tasks.
2
Experiments show clear performance improvements over existing models on both the public ABIDE dataset and the restricted ABCD dataset.
3
The Brain Network Transformer models ROI brain networks as fixed-size, ordered graphs, enabling connection profiles to serve as low-cost positional information.
4
The authors re-standardize evaluation on the publicly available large-scale ABIDE dataset to enable meaningful comparisons among brain-network models.
5
The proposed Orthonormal Clustering Readout uses self-supervised soft clustering and orthonormal projection to produce cluster-aware node and informative graph embeddings.
Research Object
human brain networks modeled as graphs of ordered Regions of Interest (ROIs) and their connections
Research Subject
Transformer-based analysis of brain networks, including pairwise ROI connection-strength learning, cluster-aware node and graph representations, and predictive performance on downstream tasks
Publication Details
Publication Date
2022-01-01
Journal
Publisher
ISSN
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest