Inferring Brain Dynamics via Multimodal Joint Graph Representation EEG-fMRI

TL;DR

This paper introduces a multimodal joint graph representation approach to analyze EEG and fMRI data simultaneously, revealing brain dynamics and functional changes through nonlinear fusion and deep learning techniques.

Contribution

It presents a novel graph-based framework for integrating EEG and fMRI data, enabling improved analysis of brain components and neuroplasticity.

Findings

Enhanced detection of brain component correlations

Effective nonlinear fusion of EEG and fMRI streams

Potential for improved neuroplasticity diagnosis

Abstract

Recent studies have shown that multi-modeling methods can provide new insights into the analysis of brain components that are not possible when each modality is acquired separately. The joint representations of different modalities is a robust model to analyze simultaneously acquired electroencephalography and functional magnetic resonance imaging (EEG-fMRI). Advances in precision instruments have given us the ability to observe the spatiotemporal neural dynamics of the human brain through non-invasive neuroimaging techniques such as EEG & fMRI. Nonlinear fusion methods of streams can extract effective brain components in different dimensions of temporal and spatial. Graph-based analyzes, which have many similarities to brain structure, can overcome the complexities of brain mapping analysis. Throughout, we outline the correlations of several different media in time shifts from one source with graph-based and deep learning methods. Determining overlaps can provide a new perspective for diagnosing functional changes in neuroplasticity studies.