Methods to detect, characterize, and remove motion artifact in resting state fMRI

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Biophysics

Background:

• Resting-state functional connectivity (RSFC) fMRI is crucial for understanding brain function.
• Head motion is a significant confound in RSFC studies, systematically altering connectivity measures.
• The precise impact of motion on signal intensity and correlations requires detailed examination.

Purpose of the Study:

• To investigate how head motion affects signal intensity and RSFC correlations.
• To evaluate the effectiveness of different artifact removal strategies, including global signal regression and censoring.
• To identify reliable metrics for assessing data quality in the presence of motion.

Main Methods:

• Analysis of motion-induced signal changes in fMRI data, characterizing their waveform, spatial extent, and temporal persistence.
• Assessment of RSFC correlations during and after motion, examining distance-dependent effects.
• Comparison of motion-correction techniques: motion-based regressors, global signal regression, and volume censoring.

Main Results:

• Motion-induced signal changes are complex, widespread, and can persist long after motion stops.
• These signal changes increase RSFC correlations in a distance-dependent manner.
• Global signal regression effectively reduces motion artifacts, while motion-based regressors are less effective. Volume censoring significantly reduces group differences due to motion.

Conclusions:

• Head motion introduces systematic biases in RSFC, which are not always corrected by standard methods.
• Global signal regression and volume censoring are effective strategies for mitigating motion artifacts.
• Careful artifact removal is essential for accurate interpretation of RSFC findings and data quality assessment.