Combining graph neural networks and ROI-based convolutional neural networks to infer individualized graphs for Alzheimer's prediction.
Kevin MuellerAnke Meyer-BaeseGordon ErlebacherPublished in: Medical Imaging: Biomedical Applications in Molecular, Structural, and Functional Imaging (2023)
Keyphrases
- convolutional neural networks
- neural network
- graph representation
- graph theory
- graph structure
- graph matching
- graph databases
- graph model
- directed graph
- graph theoretic
- adjacency matrix
- graph structures
- weighted graph
- labeled graphs
- graph construction
- graph classification
- graph mining
- graph theoretical
- prediction model
- series parallel
- graph search
- graph clustering
- graph partitioning
- bipartite graph
- prediction accuracy
- pattern recognition
- graph data
- graph isomorphism
- subgraph isomorphism
- spanning tree
- graph properties
- region of interest
- reachability queries
- undirected graph
- structural pattern recognition
- dynamic graph
- graph transformation
- finding the shortest path
- random graphs
- inexact graph matching
- minimum spanning tree
- maximum common subgraph
- connected graphs
- graph patterns
- graph representations
- convolutional network
- graph kernels
- maximum clique
- directed acyclic
- maximum cardinality
- evolving graphs
- topological information
- bounded treewidth
- multiresolution
- connected dominating set
- attributed relational graph
- small world
- dense subgraphs
- shortest path
- back propagation
- structured data
- maximal cliques
- graph layout
- query graph
- medical images
- massive graphs
- attributed graphs
- planar graphs
- neighborhood graph
- web graph
- edge weights