Hierarchical Multi-resolution Mesh Networks for Brain Decoding

TL;DR

This paper introduces Hierarchical Multi-resolution Mesh Networks (HMMNs), a novel framework that leverages multi-scale fMRI data decomposition and ensemble learning to improve brain decoding accuracy by capturing diverse information across brain regions and resolutions.

Contribution

The paper presents a new hierarchical framework combining wavelet-based multi-resolution analysis, mesh network modeling, and fuzzy stacked generalization for enhanced brain decoding from fMRI signals.

Findings

HMMN effectively discriminates cognitive tasks using multi-resolution mesh network features.

Significant topological variations observed across different frequency subbands.

Classifier ensemble benefits from diverse information at multiple resolutions.

Abstract

We propose a new framework, called Hierarchical Multi-resolution Mesh Networks (HMMNs), which establishes a set of brain networks at multiple time resolutions of fMRI signal to represent the underlying cognitive process. The suggested framework, first, decomposes the fMRI signal into various frequency subbands using wavelet transforms. Then, a brain network, called mesh network, is formed at each subband by ensembling a set of local meshes. The locality around each anatomic region is defined with respect to a neighborhood system based on functional connectivity. The arc weights of a mesh are estimated by ridge regression formed among the average region time series. In the final step, the adjacency matrices of mesh networks obtained at different subbands are ensembled for brain decoding under a hierarchical learning architecture, called, fuzzy stacked generalization (FSG). Our results on Human Connectome Project task-fMRI dataset reflect that the suggested HMMN model can successfully discriminate tasks by extracting complementary information obtained from mesh arc weights of multiple subbands. We study the topological properties of the mesh networks at different resolutions using the network measures, namely, node degree, node strength, betweenness centrality and global efficiency; and investigate the connectivity of anatomic regions, during a cognitive task. We observe significant variations among the network topologies obtained for different subbands. We, also, analyze the diversity properties of classifier ensemble, trained by the mesh networks in multiple subbands and observe that the classifiers in the ensemble collaborate with each other to fuse the complementary information freed at each subband. We conclude that the fMRI data, recorded during a cognitive task, embed diverse information across the anatomic regions at each resolution.