Development of motion correction algorithms for functional MRI data using a custom simultaneously excited multi-slice MRI acquisition with prospectively injected motion - Project Summary Head motion and its residual artifact after motion correction are one of the most dominant non-neuronal sources of artifact in fMRI data. Currently, the most commonly employed fMRI motion correction algorithms assume that head motion is synchronized to the volume acquisition. However, typical fMRI data is acquired as a stack of 2 dimensional multi- slice EPI images, acquired at different time points throughout the volume repetition time. This inter-volume motion assumption is not corrected by these correction techniques. In a previous project (R03EB012968), we modified the Prospective Acquisition CorrEcted (PACE) EPI sequence to inject intra-volume (slice-wise) motion during EPI acquisition. This modified sequence is referred to as SIMulated PACE (SIMPACE). Using SIMPACE, we scanned in-situ cadaver brains, free from physiologic noise or head motion and generated intra- volume motion injected SIMPACE dataset. Using these data, we developed the SLice-Oriented MOtion COrrection (SLOMOCO) method. SLOMOCO is the only publicly available method to correct intra-volume motion for a conventional fMRI dataset. The current project will extend the SLOMOCO project in order to 1) collect simultaneously excited multi-slice (SMS) SIMPACE data with various pattern of inter-/intra- volume head motion using ex-vivo brain phantom, 2) improve the slicewise motion correction method and its pipeline and 3) propose intra-motion index to represent quality of (rs-)fMRI analysis output. The generated SMS SIMPACE data will be made available to the public. The improved SLOMOCO software will be shared in a neuro-community. We will validate the new SLOMOCO on HCP resting state fMRI data and present the new motion index to present quality of fMRI data, which will be used as the motion outlier for the censoring volume or the exclusion of the study. Successful outcome of the proposed project will provide a validated, gold-standard metric for assessment of residual motion artifact in fMRI data.
