Assessing motion-associated tSNR of brainstem and spinal cord fMRI in a cohort with chronic hemiparetic stroke

Motivation: Motor-task fMRI is a critical modality for studying post-stroke neuronal changes in the brainstem and spinal cord, but data quality is reduced in these regions. Goal(s): We aimed to anticipate the degree of head-motion confounds before a motor-task fMRI scan and assess how this head motion affects fMRI data quality. Approach: Six post-stroke individuals performed a hand-grasp task during a head motion-capture session, then during cortical-brainstem and spinal-cord fMRI scans. Results: Head motion in the lab was positively correlated with head and spinal-cord motion during fMRI. Head and spinal-cord motion were correlated with decreased tSNR in the cortex, brainstem, and spinal cord. Impact: Head motion outside of the scanner is linked with head and spinal-cord fMRI motion and tSNR. With lab-based motion capture systems, we can anticipate motion and tSNR impacts in motor-task fMRI, useful as screening in clinical populations with increased movement.