Abstract:Graph neural networks (GNNs) have recently grown in popularity for disease prediction. Existing GNN-based methods primarily build the graph topological structure around a single modality and combine it with other modalities to acquire feature representations of acquisitions. The complicated relationship in each modality, however, may not be well highlighted due to its specificity. Further, relatively shallow networks restrict adequate extraction of high-level features, affecting disease prediction performance. Accordingly, this paper develops a new interactive deep cascade spectral graph convolutional network with multi-relational graphs (IDCGN) for disease prediction tasks. Its crucial points lie in constructing multiple relational graphs and dual cascade spectral graph convolution branches with interaction (DCSGBI). Specifically, the former designs a pairwise imaging-based edge generator and a pairwise non-imaging-based edge generator from different modalities by devising two learnable networks, which adaptively capture graph structures and provide various views of the same acquisition to aid in disease diagnosis. Again, DCSGBI is established to enrich high-level semantic information and low-level details of disease data. It devises a cascade spectral graph convolution operator for each branch and incorporates the interaction strategy between different branches into the network, successfully forming a deep model and capturing complementary information from diverse branches. In this manner, more favorable and sufficient features are learned for a reliable diagnosis. Experiments on several disease datasets reveal that IDCGN exceeds state-of-the-art models and achieves promising results.

Dual Feature Interaction-Based Graph Convolutional Network

A novel interactive deep cascade spectral graph convolutional network with multi-relational graphs for disease prediction

Cross-GCN: Enhancing Graph Convolutional Network with K-Order Feature Interactions

Dual-channel deep graph convolutional neural networks

Exploiting Node-Feature Bipartite Graph in Graph Convolutional Networks.

Multi-Behavior Enhanced Heterogeneous Graph Convolutional Networks Recommendation Algorithm based on Feature-Interaction

DAGCN: Dual Attention Graph Convolutional Networks

Dual Graph Complementary Network

Bilinear Graph Neural Network with Neighbor Interactions

HDGCN: Dual-Channel Graph Convolutional Network with Higher-Order Information for Robust Feature Learning

Cross feature enhanced graph convolutional network for aspect-based sentiment analysis

2 Related Work Following the success of deep neural networks for image classification

Multigraph Fusion for Dynamic Graph Convolutional Network

DualGCN: Representation Learning for Property Graphs Via Dual-level Convolutional Networks

Adaptive Multi-Channel Contrastive Graph Convolutional Network with Graph and Feature Fusion

A Graph Convolutional Neural Network Model Based on Fused Multi-Subgraph As Input and Fused Feature Information As Output

Unified GCNs: Towards Connecting GCNs with CNNs

Factorizable Graph Convolutional Networks.

Dual Graph Convolutional Network for Semantic Segmentation.

Multi-scale Graph Convolutional Networks with Self-Attention

CatGCN: Graph Convolutional Networks with Categorical Node Features