PROJECT SUMMARY
The goal of this project is to further develop and optimize the next generation arterial spin labeling (ASL)
technologies for quantitative mapping of microvascular perfusion at the level of cortical layers and columns on
the first FDA approved ultrahigh magnetic (UHF) system, the 7T Terra. Blood oxygen level dependent (BOLD)
fMRI is the most widely used non-invasive imaging modality for studying the dynamics of macroscopic brain
networks and mesoscopic brain circuits. However, BOLD contrast is susceptible to contaminations of pial veins
on the cortical surface that significantly confounds laminar fMRI. Cerebral blood flow (CBF) or microvascular
perfusion measured by ASL is a key parameter for in vivo assessment of neurovascular function. The ASL
signal is localized close to the site of neural activation and offers the unique capability for quantitative CBF
measurements both at rest and during task activation. We pioneered laminar perfusion imaging using 3D inner-
volume GRASE (Gradient and Spin Echo) ASL at 7T with a spatial resolution of ~1mm3. However, a sub-
millimeter spatial resolution, and ideally at the level of ~0.1mm3 (or ~0.5x0.5x0.5mm3), is required for reliable
differentiation of neural activities across cortical layers and columns, as well as for comparison with the state-
of-the-art BOLD and CBV fMRI. We will take advantage of a few latest technical breakthroughs in our lab: 1)
Cutting-edge ASL pulse sequences with optimized spin labeling strategies for laminar perfusion imaging at 7T;
2) A novel k-t CAIPIRANHA scheme in conjunction with a total-generalized-variation (TGV) regularized
algorithm for robust under-sampling patterns and constrained reconstruction; and 3) A novel denoising
technique termed k-space weighted image average (KWIA) invented by our group that is able to reduce the
thermal noise by 50% and double the signal-to-noise ratio (SNR) of dynamic MRI without significantly affecting
the spatial and temporal resolution. We will then apply the advanced ASL methods to precisely measure
perfusion, arterial transit time (ATT) and T1 of brain tissue across cortical layers during resting state, as well as
to precisely measure task induced perfusion signal changes across cortical layers and columns using
sensorimotor stimulation and working memory tasks. As an exploratory aim, we will further develop an
innovative pulse sequence for concurrent measurements of T2w BOLD, CBF and CBV contrasts at 7T for
mesoscopic imaging of metabolic activities. The developed ASL technologies and research findings will be
highly valuable to both basic and clinical neuroscientific research. We will also evaluate the developed next-
generation ASL pulse sequences and post-processing algorithms at 3T, and disseminate these technologies to
other sites with 3 and/or 7T MR systems to facilitate the widespread adoption of our technologies by the
neuroscientific community.