Anatomically-adapted graph wavelets for improved group-level fMRI activation mapping

Area of Science:

• Neuroimaging
• Graph Theory
• Signal Processing

Background:

• Functional magnetic resonance imaging (fMRI) is crucial for mapping brain activity.
• Conventional statistical parametric mapping (SPM) has limitations in group-level analysis.
• Existing wavelet-based SPM (WSPM) offers an alternative but can be further enhanced.

Purpose of the Study:

• To introduce a novel graph-based framework for fMRI brain activation mapping.
• To develop an advanced multi-resolutional spatial transformation for fMRI data using spectral graph wavelet transform (SGWT).
• To improve upon existing methods for group-level fMRI analysis.

Main Methods:

• Utilizing spectral graph wavelet transform (SGWT) for spatial transformation of fMRI data.
• Constructing brain graphs with subgraphs for cerebral and cerebellar gray matter (GM) connectivity.
• Employing template GM representations to address inter-subject variability.
• Developing GM-based spatial transformation using tailored graph wavelets.

Main Results:

• The proposed graph-based framework demonstrates superior type-I error control compared to SPM and classical WSPM.
• Evaluated using real and semi-synthetic multi-subject fMRI data.
• Achieved higher detection sensitivity in real data analysis.
• Showcased capability in capturing subtle, connected patterns of brain activity.

Conclusions:

• The novel graph-based framework offers enhanced performance for fMRI brain activation mapping.
• The approach provides improved statistical control and sensitivity for group-level analyses.
• This method advances the analysis of complex, interconnected brain activity patterns.