Elsevier

NeuroImage

Volume 66, 1 February 2013, Pages 662-671
NeuroImage

Comparison of 2D and 3D single-shot ASL perfusion fMRI sequences

https://doi.org/10.1016/j.neuroimage.2012.10.087Get rights and content

Abstract

Arterial spin labeling (ASL) can be implemented by combining different labeling schemes and readout sequences. In this study, the performance of 2D and 3D single-shot pulsed-continuous ASL (pCASL) sequences was assessed in a group of young healthy volunteers undergoing a baseline perfusion and a functional study with a sensory-motor activation paradigm. The evaluated sequences were 2D echo-planar imaging (2D EPI), 3D single-shot fast spin-echo with in-plane spiral readout (3D FSE spiral), and 3D single-shot gradient-and-spin-echo (3D GRASE). The 3D sequences were implemented with and without the addition of an optimized background suppression (BS) scheme. Labeling efficiency, signal-to-noise ratio (SNR), and gray matter (GM) to white matter (WM) contrast ratio were assessed in baseline perfusion measurements. 3D acquisitions without BS yielded 2-fold increments in spatial SNR, but no change in temporal SNR. The addition of BS to the 3D sequences yielded a 3-fold temporal SNR increase compared to the unsuppressed sequences. 2D EPI provided better GM-to-WM contrast ratio than the 3D sequences. The analysis of functional data at the subject level showed a 3-fold increase in statistical power for the BS 3D sequences, although the improvement was attenuated at the group level. 3D without BS did not increase the maximum t-values, however, it yielded larger activation clusters than 2D. These results demonstrate that BS 3D single-shot imaging sequences improve the performance of pCASL in baseline and activation studies, particularly for individual subject analyses where the improvement in temporal SNR translates into markedly enhanced power for task activation detection.

Highlights

► 3D pCASL sequences increased perfusion spatial SNR by a factor of 2. ► The addition of BS to 3D readouts increased perfusion temporal SNR by a factor of 3. ► 2D EPI showed higher GM-to-WM CBF contrast ratio. ► Statistical power to detect functional activation was higher in 3D BS sequences. ► 3D BS single-shot sequences appear as the preferable readout choice in pCASL fMRI.

Introduction

Arterial spin labeling (ASL) (Detre et al., 1992, Williams et al., 1992) utilizes magnetically labeled arterial blood water as an endogenous tracer, enabling repeatable non-invasive quantification of cerebral blood flow (CBF) in units of mL/min/100 g of tissue.

ASL MRI has a broad range of applications in basic and clinical neuroscience, including the evaluation of brain perfusion in disease states and its use as a means of monitoring regional brain function based on the tight coupling between regional CBF and neural activity (Raichle, 1998). Experimental evidence suggests that ASL fMRI could provide several advantages over Blood Oxygenation Level Dependent (BOLD) contrast as a biomarker of regional neural activity, though for most studies of evoked activity BOLD is still much more sensitive. One such advantage is its superior low-frequency sensitivity, which allows ASL to be used to monitor brain function over much longer intervals than BOLD (Aguirre et al., 2002, Wang et al., 2003). CBF changes have also been shown to provide superior spatial (Silva et al., 2000) and temporal (Huppert et al., 2006) resolution for detecting task effects, but these benefits have not been widely realized in practice due to the relatively poor sensitivity of most ASL methods.

ASL can be considered as a form of magnetization preparation. There are two main approaches to ASL, known as continuous and pulsed ASL (CASL and PASL, respectively). CASL methods generally provide higher SNR, but are more difficult to implement since Radio-Frequency (RF) hardware found in most current clinical MRI systems is not designed to give prolonged RF pulses (Detre et al., 2012). Pulsed-continuous ASL (pCASL) (Dai et al., 2008, Wu et al., 2007) overcomes this difficulty using a train of pulsed RF pulses for labeling, achieving both higher efficiency and compatibility with standard MRI hardware.

The effects of ASL are measured by comparison with control labeling, and this difference can be modeled with additional assumptions or measurements to derive CBF (Detre et al., 1992, Williams et al., 1992). At 3 T, pCASL can label approximately 1% of brain water.

The ASL contrast can be sampled with any imaging sequence, though high sensitivity sequences are most desirable to optimally measure the subtle effects of ASL. Much of the existing literature on ASL MRI is based on the use of echo-planar imaging (EPI) due to its high sensitivity and rapid acquisition of volumetric data that allows label and control images to be interleaved every few seconds. EPI also adds little additional RF deposition to ASL MRI, though this is generally not a limitation at field strengths up to 3 T.

However, some drawbacks of 2D multi-slice acquisitions (such as the slice dependence of the obtained perfusion SNR, due to the different slice acquisition times) have encouraged the use of 3D readouts (Duhamel and Alsop, 2004, Fernandez-Seara et al., 2005, Gai et al., 2011, Gunther et al., 2005, Nielsen and Hernandez-Garcia, in press). In addition, background suppression (BS) of static tissue signal (Garcia et al., 2005, Ye et al., 2000), which is optimally combined with 3D acquisitions (Fernandez-Seara et al., 2008, Gunther et al., 2005), has been shown to improve the sensitivity of ASL by reducing physiological noise (Wu et al., 2009), further increasing the interest in the utilization of 3D single-shot readout sequences such as 3D gradient-and-spin-echo (GRASE) (Gunther et al., 2005) and 3D fast spin-echo (FSE) with in-plane spiral trajectory (Dai et al., 2008).

The purpose of this work was to compare the standard 2D EPI to the 3D pCASL sequences with and without BS by evaluating their performance in baseline perfusion measurements and functional activation studies. The three imaging sequences mentioned above were selected due to their prevalence in the literature: 2D EPI, 3D GRASE and 3D FSE spiral. Single-shot versions of these sequences were chosen to allow for evoked activation studies. Both 3D sequences were implemented and tested with and without an identical optimized BS scheme. All five sequence variants were compared in a group of ten young healthy volunteers, during a baseline perfusion measurement and a functional study with a sensory-motor activation paradigm.

Section snippets

Subjects

A total of 10 young healthy volunteers (5 females; mean age ± standard deviation (SD) = 29 ± 3 years) participated in the study, after signing a written informed consent.

Pulsed-continuous arterial spin labeling sequences

Five pCASL sequence variants with different readout schemes were subjected to testing: a standard 2D EPI sequence, a 3D single-shot FSE with in-plane spiral readout (3D FSE spiral), and a homologous 3D single-shot GRASE with Cartesian readout (3D GRASE). Both 3D sequences were tested with and without the addition of an optimized BS

PC velocity MRI results

The whole-brain mean CBF estimated for the ten subjects participating in the study from the PC velocity MRI data was 53.5 ± 7.9 mL/min/100 g (mean ± SD across subjects). This result agrees well with values reported for this age range (for a review, see (Lassen, 1985)).

Baseline perfusion results

Representative perfusion and CBF maps from one subject acquired with the five tested sequences are displayed in Fig. 3. Table 1 summarizes the main parameters characterizing the performance of the five sequences. Unsuppressed 3D

Discussion

The effect of readout choice in pCASL perfusion MRI sequences was assessed in this work. To this end, five sequence variants were compared in a group of ten young healthy volunteers, both during a baseline perfusion measurement and during a functional study with a sensory-motor activation paradigm.

Implementations of 2D EPI, 3D GRASE and 3D FSE spiral, with the same repetition time, resolution and identical pCASL pulse were compared. In 2D readouts each slice has a different acquisition time,

Acknowledgments

We are thankful to Dr. David Alsop for his help with the implementation of the 3D FSE spiral sequence and BS optimization. We also thank Drs. Josef Pfeuffer and Tiejun Zhao from Siemens for their help with the spiral readout sequence and online reconstruction programming. We thank Drs. David Feinberg and Matthias Guenther for sharing the original 3D GRASE source code.

This work was supported by the Spanish Ministry of Science and Innovation (grants SAF2008-00678 and RYC-2010-07161) and the

References (47)

  • R.B. Buxton et al.

    A general kinetic model for quantitative perfusion imaging with arterial spin labeling

    Magn. Reson. Med.

    (1998)
  • Y. Chen et al.

    Comparison of arterial transit times estimated using arterial spin labeling

    MAGMA

    (2012)
  • J.Y. Cheng et al.

    Fast concomitant gradient field and field inhomogeneity correction for spiral cardiac imaging

    Magn. Reson. Med.

    (2011)
  • Z.H. Cho et al.

    Reduction of susceptibility artifact in gradient-echo imaging

    Magn. Reson. Med.

    (1992)
  • W. Dai et al.

    Continuous flow-driven inversion for arterial spin labeling using pulsed radio frequency and gradient fields

    Magn. Reson. Med.

    (2008)
  • J.A. Detre et al.

    Perfusion imaging

    Magn. Reson. Med.

    (1992)
  • J.A. Detre et al.

    Applications of arterial spin labeled MRI in the brain

    J. Magn. Reson. Imaging

    (2012)
  • G. Duhamel et al.

    Single-shot susceptibility insensitive whole brain 3D fMRI with ASL

  • M.A. Fernandez-Seara et al.

    Continuous arterial spin labeling perfusion measurements using single shot 3D GRASE at 3 T

    Magn. Reson. Med.

    (2005)
  • M.A. Fernandez-Seara et al.

    Minimizing acquisition time of arterial spin labeling at 3 T

    Magn. Reson. Med.

    (2008)
  • M. Frigo et al.

    The design and implementation of FFTW3

    Proc. IEEE

    (2005)
  • N.D. Gai et al.

    Whole-brain cerebral blood flow mapping using 3D echo planar imaging and pulsed arterial tagging

    J. Magn. Reson. Imaging

    (2011)
  • D.M. Garcia et al.

    Efficiency of inversion pulses for background suppressed arterial spin labeling

    Magn. Reson. Med.

    (2005)
  • Cited by (135)

    View all citing articles on Scopus
    1

    Both authors contributed equally to this work.

    View full text