High-Field Functional MRI from the Perspective of Single Vessels in Rats and Humans

Abstract

Functional MRI (fMRI) has been employed to map brain activity and connectivity based on the neurovascular coupled hemodynamic signal. However, in most cases of fMRI studies, the cerebral vascular hemodynamic signal has been imaged in a spatially smoothed manner due to the limit of spatial resolution. There is a need to improve the spatiotemporal resolution of fMRI to map dynamic signal from individual venule or individual arteriole directly. Here, the thesis aims to provide a vascular-specific view of hemodynamic response during active state or resting state. To better characterize the temporal features of task-related fMRI signal from different vascular compartments, we implemented a line-scanning method to acquire vessel-specific blood-oxygen-level-dependent (BOLD) / cerebral-blood-volume (CBV) fMRI signal at 100-ms temporal resolution with sensory or optogenetic stimulation. Furthermore, we extended the line-scanning method with multi-echo scheme to provide vessel-specific fMRI with the higher contrast-to-noise ratio (CNR), which allowed us to directly map the distinct evoked hemodynamic signal from arterioles and venules at different echo time (TE) from 3 ms to 30 ms. The line-scanning fMRI methods acquire single k-space line per TR under a reshuffled k space acquisition scheme which has the limitation of sampling the fMRI signal in real-time for resting-state fMRI studies. To overcome this, we implemented a balanced Steady-state free precession (SSFP) to map task-related and resting-state fMRI (rsfMRI) with high spatial resolution in anesthetized rats. We reveal venule-dominated functional connectivity for BOLD fMRI and arteriole-dominated functional connectivity for CBV fMRI. The BOLD signal from individual venules and CBV signal from individual arterioles show correlations at an ultra-slow frequency (< 0.1 Hz), which are correlated with the intracellular calcium signal measured in neighboring neurons. In complementary data from awake human subjects, the BOLD signal is spatially correlated among sulcus veins and specified intracortical veins of the visual cortex at similar ultra-slow rhythms. This work provides a high-resolution fMRI approach to resolve brain activation and functional connectivity at the level of single vessels, which opened a new avenue to investigate brian functional connectivity at the scale of vessels.