Removing the effects of task-related motion using independent-component analysis

Area of Science:

• Neuroimaging
• Signal Processing
• Biomedical Engineering

Background:

• Task-related motion is a significant source of noise in functional magnetic resonance imaging (fMRI) time series.
• Motion artifacts often persist even after standard spatial realignment procedures.
• Residual motion effects can confound the interpretation of fMRI results.

Purpose of the Study:

• To develop and validate a novel method for removing persistent task-related motion effects in fMRI data.
• To improve the accuracy and reliability of fMRI analyses by reducing motion-induced noise.
• To offer an alternative to conventional regression-based methods for motion artifact correction.

Main Methods:

• Functional magnetic resonance imaging (fMRI) data preprocessing using Independent Component Analysis (ICA).
• Automatic classification and rejection of Independent Components (ICs) associated with residual task-related motion.
• Reconstruction of fMRI time series excluding identified motion-related ICs.
• Utilized signal intensity and variance changes linked to tasks for IC classification.

Main Results:

• The proposed ICA-based method effectively removed task-related motion effects.
• The method demonstrated superior performance compared to conventional voxel-wise regression-based approaches.
• Verified effectiveness in an fMRI experiment with controlled head motion.

Conclusions:

• The ICA-based approach offers a robust solution for mitigating persistent task-related motion artifacts in fMRI.
• This technique enhances the quality of fMRI data, leading to more reliable scientific findings.
• The method provides a valuable tool for neuroimaging researchers dealing with motion-corrupted datasets.