Improved Arterial Spin Labeling (ASL) Perfusion Imaging

Arterial Spin Labeling (ASL) [1] is a category of noninvasive MRI techniques for tissue perfusion measurements, which employ magnetically labeled arterial water as an endogenous tracer.

In any ASL experiment two sets of images are collected: 1) control image, which is basically a regular MRI image of the tissue; and 2) tag image, in which the magnetization of the inflowing arterial blood water is magnetically modified (tagged) below the imaging volume and is acquired after a delay allowing for tagged blood to flow into the target tissue. Employing appropriate pulse sequence parameters, the subtraction of control and tag images provides a signal that is directly proportional to tissue perfusion. 

Arterial Spin labeling with motion correction using pcASL and PROMO

(W. Zun, H. Jahanian)

ASL is susceptible to subject motion because it relies on the perfect cancellation of static tissue in the subtraction between control and tagged images. Prospective motion correction is a technique based on updating pulse sequences in real time such that the logical coordinate system remains fixed with respect to the subject during the entire scan. Compared to conventional motion correction based on retrospective techniques, prospective motion correction using a navigator has advantages such as compatibility with 3D segmented imaging and fully optimized background suppression, and less complicated image registration algorithm. We integrated PseudoContinuous ASL (pcASL) [2] and PROspective MOtion Correction (PROMO) [3] by inserting the navigator sequence in the idle time of ASL preparation without increasing repetition time of ASL sequence. We verified that the perturbation of ASL signal due to navigator was negligible and that ASL image artifacts from both controlled and random motions were removed significantly with PROMO [4]. Furthermore, our investigation suggested that motion correction is more critical in tagged images than control images because of tissue contrast being mostly contained in tagged images. Our next step is to evaluate the performance of pcASL-PROMO in acute stroke patients.

Optimization of 2D/3D velocity-selective arterial spin labeling

(W. Zun, H. Jahanian)

VS-ASL [5] is a promising method for measuring cerebral blood flow in the presence of slow or delayed flow. In this method, arterial blood is tagged based on its velocity instead of location and thereby in principle, arterial transit time is eliminated. In this project, we are improving VS-ASL sequence for clinical triage with both 2D and 3D image acquisitions. For 2D multislice VS-ASL, we maximized SNR efficiency using timing optimization, and achieved near-contiguous multislice imaging by removing stimulated echoes with variable spoilers and by sharpening the slice profile of spin-echo excitation with matched-phase RF design [6, 7]. Improved 2D VS-ASL showed reduced physiological noise and increased correlation with xenon-enhanced CT. On the other hand, 3D VS-ASL demonstrated higher SNR efficiency and improved background suppression compared to 2D VS-ASL. Optimization of 3D VS-ASL in patient population using motion correction is a work in progress. 

Cerebral hemodynamics and collaterals in acute ischemic stroke

(W. Zun, H. Jahanian)

In acute ischemic stroke, collateral flow may play a central role in sustaining tissue viability and reducing the risks of complications. However, there has been no study of the role of collateral flow in acute stroke patients to date. The overall goal of this project is to determine how to best incorporate collateral perfusion assessments with MRI into patient-selection algorithms for endovascular therapy. To this end, we are using multi-delay PCASL sequence to estimate arterial transit times and CBF on a voxel-by-voxel basis [8]. Acute stroke patients with collateral flow may present prolonged transit times but with intact CBF level in the transit time-corrected ASL measurement. While transit time will be compared with Tmax of bolus perfusion imaging, regional collateral scores based on transit time and CBF will be validated against digital subtraction angiography (DSA). Investigation of collaterals will provide a more comprehensive framework for predicting infarct growth and patient outcome in the work-up of acute stroke.

 


References: [1] Detre et al, MRM 1992. [2] Dai et al, MRM 2008. [3] White et al, MRM 2010. [4] Zun et al, MRM 2014. [5] Wong et al, MRM 2006. [6] Zun et al, MRM 2014. [7] Zun et al, JMRI 2014. [8] Dai et al, MRM 2012.

Greg Zaharchuk, MD., PhD.

Associate Professor of Radiology
Office: Lucas Center, PS-04
Phone: (650) 735-6172
Email: gregz@stanford.edu

Michael E. Moseley, PhD.

Professor, Radiology
Office: Lucas Center, Rm PS-062
Phone: (650) 723-8697
email: moseley@stanford.edu