Evaluating the Influence of Spatial Resampling for Motion Correction in Resting-State Functional MRI

Area of Science:

• Neuroimaging
• Magnetic Resonance Imaging
• Brain Activity Analysis

Background:

• Head motion is a significant challenge in resting-state functional MRI (fMRI).
• Image realignment corrects motion but spatial resampling can introduce spurious variance.
• This variance may lead to errors in the Blood-Oxygen-Level-Dependent (BOLD) signal amplitude.

Purpose of the Study:

• To characterize variance introduced by spatial resampling in fMRI.
• To identify brain regions susceptible to spatial resampling artifacts.
• To compare the efficacy of different motion regression methods in mitigating these artifacts.

Main Methods:

• Two simulation experiments were conducted using fMRI data.
• Simulations involved estimated motion parameters and different motion types.
• Four popular motion regression approaches were evaluated.

Main Results:

• Spurious variance was most pronounced in peripheral cortical regions.
• The correlation between spurious variance and head motion varied across the brain and motion types.
• Friston 24 and Voxel-specific 12 models demonstrated superior performance in reducing spurious variance.

Conclusions:

• Spatial resampling is a source of artifact in fMRI analysis.
• Understanding these artifacts is crucial for accurate interpretation of motion-BOLD relationships.
• Friston 24 and Voxel-specific 12 regression models are recommended for mitigating resampling-induced variance.