High-Speed fMRI of Resting State Connectivity

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Functional connectomics using resting state fMRI is a rapidly expanding task-free approach, which is expected to have significant clinical impact. Recent studies have demonstrated more than 100 resting state networks (RSNs), which exhibit considerable temporal and spatial non-stationary. There is now considerable interest in characterizing resting state connectivity at much higher frequencies (up to 5 Hz) than detectable with traditional resting state fMRI. However, intra-scan non-stationary of connectivity and head movement artifacts compromise the sensitivity of detecting RSNs in single subjects. Advances in fMRI methodology to increase data acquisition rates have played an important role in improving BOLD sensitivity for task-based fMRI and for mapping the temporal dynamics of brain-states. Increasing the sampling rate improves segregation of resting state signal fluctuation and cardiac-related signal pulsation, which as our preliminary data show exhibits considerable regional differences of the pulsation waveform across cortical brain regions. However, high-speed fMRI suffers from reconstruction artifacts at high acceleration rates, and adding multi-echo acquisition to separate BOLD and non-BOLD signal sources has not been investigated. The long-term goal of this research is to develop a reliable and sensitive resting state fMRI method for presurgical mapping of eloquent brain regions in patients with neurological disorders. The objective of this proposal is to integrate simultaneous multi-slice (SMS) EPI and multi-slab echo-volumar imaging (EVI), which have emerged as the principal high-speed data acquisition approaches with multi-dimensional compressed sensing, to achieve unprecedented temporal and spatial resolution and BOLD sensitivity without sacrificing image quality. We will additionally integrate a comprehensive correction of motion artifacts. This approach will be validated in healthy controls and patients with brain tumors using time-dynamic statistical analysis methods to sensitively capture the temporal dynamics of resting state fluctuations. The rationale of this research is that resting state fMRI has the potential to replac task-based fMRI for mapping major brain networks, which would reduce costs and enable fMRI in patient groups that cannot be studied with task-based fMRI. While functional localization of resting state fMRI across groups of subjects has been demonstrated to be consistent with task-based fMRI, there is a need to improve reliability of resting state fMRI at the single subject leve. This research will establish ultra-high-speed fMRI as a sensitive and quantitative approach for mapping resting state connectivity in individual subjects in clinical and neuroscience research. Characterizing the temporal dynamics of resting state connectivity will enhance our understanding of intrinsic brain activity at rest to improve diagnostics and to guide therapy. This will ultimately lead to improved individualized treatment strategies and prognosis based upon patient specific functional brain mapping.