Causal connectivity along the spinal cord using high-resolution 7T fMRI

项目摘要/摘要

PROJECT SUMMARY / ABSTRACT This proposal aims to (1) develop and validate high spatial and temporal resolution acquisition of functional magnetic resonance imaging (fMRI) in the human cervical spinal cord (C-spine) at ultra-high field (7 Tesla), and (2) non-invasively detect and characterize directional spinal cord networks using fMRI during both rest and task. These methods may find clinical application in central nervous system (CNS) diseases involving the spinal cord. Studying spinal cord function using fMRI has gained traction in the past decade. As of today, we know from fMRI functional connectivity (FC) studies that the left and right dorsal horns are connected (so are left/right ventral horns) within the same vertebral level, but no connection exists between dorsal-ventral horns or between levels. We do not understand why this is the case, and this puzzle has intrigued spinal cord researchers. FC only measures co-activation between spinal regions, and we argue that this traditional approach has not provided us with a comprehensive picture of spinal cord’s functional architecture. We propose to solve this conundrum by measuring directional effective connectivity (EC) among spinal regions instead of co-activation (FC). This choice is supported by the underlying anatomy. The ventral horns carry efferent motor signals from motor cortex down to higher vertebral levels and then to lower ones, while dorsal horns carry afferent somatosensory signals up from lower to higher levels and then to the postcentral gyrus. Corroborating this innate biology, we propose the existence of higher-to-lower EC in ventral and lower-to-higher EC in dorsal horns of the C-spine. Addressing this central question is critical for developing mechanistic models of healthy spinal cord function and subsequently its disruption in disorders of the spinal cord. We aim to take a step toward addressing this issue here. For the first time in the field of spinal cord imaging, we propose (Aim 2) to develop a template of EC in the healthy human C-spine and compare it with the FC template during rest. Furthermore, since resting state primes task responses, we propose to validate the existence of such resting-state EC patterns while participants engage in a simple bilateral finger tapping task (Aim 3). This aim will be achieved through dynamic EC modeling to capture EC patterns specifically during the motor task blocks. Participants will be scanned twice 4 weeks apart to ascertain test-retest reliability and reproducibility of the findings. These techniques require high temporal resolution to identify directional influence of one region over the other as well as capture fast dynamic EC transitions, but also require high spatial resolution to measure the narrow spinal gray matter. Thus, for the first time, we will also develop and validate sub-second and sub-millimeter fMRI acquisition protocols to image the C-spine at ultra-high field of 7T (Aim 1). In the future, we plan to utilize the developmental work from this proposal to study disorders involving the cord. The technology, techniques and mechanistic understanding developed thorough this project has potential for superior diagnosis, disease tracking and prediction of treatment response in CNS disorders such as multiple sclerosis, amyotrophic lateral sclerosis, chronic pain and spinal cord injury.