A feature-based approach to combine functional MRI, structural MRI and EEG brain imaging data.

Area of Science:

• Neuroimaging
• Biomedical Data Analysis
• Computational Neuroscience

Background:

• Multimodal brain imaging data (structural MRI, functional MRI, EEG) are common but often analyzed separately.
• Existing methods lack integrated approaches for analyzing diverse neuroimaging data simultaneously.
• This limits the comprehensive understanding of brain structure-function relationships.

Purpose of the Study:

• To develop and evaluate a novel methodology for joint independent component analysis (jICA) across multiple neuroimaging modalities.
• To enable the integrated analysis of structural MRI, functional MRI, and EEG data.
• To improve the detection of group differences using fused neuroimaging data.

Main Methods:

• Proposed a novel joint independent component analysis (jICA) framework for multimodal data fusion.
• Integrated structural MRI, functional MRI (task-based), and EEG data.
• Utilized Kullback-Leibler divergence for evaluating data combination utility and a joint histogram for visualization.

Main Results:

• Demonstrated the method's effectiveness on a dataset of schizophrenia patients and healthy controls.
• Showcased improved group discrimination by combining different neuroimaging data types.
• Identified specific data combinations that are most informative for distinguishing between groups.

Conclusions:

• Joint analysis of multimodal neuroimaging data offers superior insights compared to separate analyses.
• The proposed jICA methodology enhances the ability to differentiate between clinical groups, such as schizophrenia patients and controls.
• This approach holds promise for advancing diagnostic and research capabilities in neuroscience.