Arterial spin labeling (ASL) enables non-invasive, quantitative MRI measurements of tissue perfusion, and has a broad range of applications including functional brain imaging. ASL can concurrently measure perfusion and blood oxygenation level dependent (BOLD) signal changes, which proves useful for investigating the brain’s physiology in health and disease. However, ASL suffers from limited temporal resolution and has a lower signal-to-noise ratio (SNR) compared to conventional BOLD imaging. In this lecture, the functioning, advantages, disadvantages and application areas of ASL will be summarized. Furthermore, the acquisition approaches and imaging parameters that influence ASL’s SNR and temporal resolution will be reviewed.
The attendees will:
- Learn about different ASL techniques and their advantages and disadvantages regarding perfusion and BOLD imaging;
- Understand the relationship between different acquisition parameters and the perfusion and BOLD sensitivity (SNR & CNR);
- Learn potential strategies to improve the temporal and spatial resolution of ASL measurements;
- Understand the relationship between specific perfusion and BOLD imaging application areas and available ASL approaches; Learn about the analysis of (functional) ASL experiments.
Tissue perfusion is a fundamental physiological variable that refers to the delivery of blood to capillary beds. Perfusion is important because it determines the rate of delivery of oxygen and nutrients (glucose, etc.) to the tissue and the rate of clearance of waste products. Non-invasive quantitative perfusion measurements can be obtained using arterial spin labeling (ASL) [1], [2]. In the ASL approach, arterial blood water is used as an endogenous diffusible tracer. The tracer is created by inverting the magnetization of the blood in the arteries with radiofrequency (RF) pulses, before it flows into the target tissue. By subtracting the labeled images from control images, acquired without inverting the blood’s magnetization, the amount of blood that entered the tissue since the beginning of the labeling can be calculated. A time-series of interleaved labeled and control image pairs is usually acquired, enabling dynamic perfusion measurements. The lifetime of the endogenous label is governed by the longitudinal relaxation time (T1) of blood, which is on the order of 1–2 s [3]. The time between the image acquisition and the end of the labeling is called post-labeling delay (PLD). A fundamental trade-off is that a short PLD does not allow for complete delivery of the labeled blood water to the tissue, whereas a long PLD results in strong T1 decay and therefore reduced signal-to-noise ratio (SNR). The non-invasive nature of ASL and its ability to dynamically and quantitatively measure CBF make it an attractive approach for both neuroscience research and clinical applications [4], [5] and its capability to concurrently measure functional perfusion and BOLD changes is useful for investigating the brain's physiology in health and disease [6], [7].
Compared to BOLD fMRI, functional CBF measurements using ASL present some drawbacks [8], such as lower SNR of the perfusion-weighted signal due to the low microvascular density and reduced temporal resolution, because of the PLD necessary to allow the labeled blood to reach the imaging slab and the need to acquire pairs of label and control images. The ASL’s brain coverage is also typically smaller compared to BOLD imaging due to the increased power deposition and T1 decay of the labeled blood. To overcome these drawbacks, different approaches have been proposed. ASL’s low SNR can be compensated using a low spatial resolution (typically 2–4 mm voxels) and signal averaging across multiple repetitions. The perfusion temporal SNR (tSNR) is determined by the control image tSNR, labeling bolus length (amount of labeled blood), PLD and the degree of background suppression. Other imaging parameters, such as the voxel size, partial Fourier factor or the RF coil employed, influence the perfusion tSNR only indirectly through the control tSNR. The relationship between perfusion tSNR and control tSNR is complex as decreases in control tSNR also lead to reduction in perfusion tSNR, but the two become decoupled when background suppression is applied. Background suppression diminishes the static tissue signal and to a smaller extent the control tSNR. In contrast, perfusion tSNR is increased when background suppression is applied due to the reduction of physiological noise in the ASL time series. The control tSNR is influenced by a large number of factors including echo time (TE), voxel size, RF receive array, use of parallel imaging and magnetic field strength. Some of these parameters directly impact the perfusion tSNR like the field strength and TE. Other factors only modulate the perfusion tSNR through the control tSNR, like the receive array and voxel size. The TE is mainly determined by the readout approach employed. The readout options range from the commonly used 2D single-shot EPI, through 2D single-shot spiral to 3D segmented spin-echo (SE) approaches. Other readout techniques include 2D simultaneous multi-slice (SMS) EPI techniques [9], [10], [11], and 2D turboFLASH [12], [13]. The 3D SE approaches with their increased SNR and optimal combination with background suppression have become commonly used [14]. The PLD is mostly determined by the research question and physiology and therefore offers a relatively limited opportunity to increase the ASL (t)SNR. The amount of labeled blood is determined by the bolus length and depends on the labeling technique employed. Pulsed, continuous, pseudo-continuous and velocity-selective approaches have been proposed.
Higher CBF sensitivity can be achieved by using background suppression and short TE irrespective of the readout approach chosen. Unfortunately, exactly they also cause a decreased BOLD sensitivity. Therefore, multi-echo EPI or spiral readouts offer a good compromise between CBF and BOLD sensitivity when their concurrent measurement is required. The labelling approach (duration) and PLD also have a major influence on the ASL’s SNR. Longer labeled bolus increases ASL’s SNR, while longer PLD decreases it. Often a compromise for their combined duration needs to be made to avoid unnecessary increase in TR and thereby decrease of temporal resolution. This is just an example of the recurring situation where the researcher needs to trade SNR for temporal resolution and/or artefact reduction. Another case of this tradeoff is segmentation in 3D SE acquisitions. Recently, advanced parallel imaging strategies that enable substantial acquisition acceleration with minor SNR penalties, irrespective of the readout, have been proposed and these will become more widely used in the future. Both clinical and functional ASL will largely benefit from these technologies as they allow rapid, accurate and reliable measurements of perfusion. Finally, ultra-high magnetic fields may further increase CBF sensitivity, once the remaining challenges related to power deposition and field inhomogeneities are tackled by developments in hardware and parallel transmission technologies [15].
1. Detre JA, Leigh JS, Williams DS, Koretsky AP. Perfusion imaging. Magn Reson Med. 1992;23(1):37-45.
2. Williams DS, Detre JA, Leigh JS, Koretsky AP. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci U S A. 1992;89(1):212-6.
3. Zhang X, Petersen ET, Ghariq E, De Vis JB, Webb AG, Teeuwisse WM et al. In vivo blood T(1) measurements at 1.5 T, 3 T, and 7 T. Magn Reson Med. 2013;70(4):1082-6. doi:10.1002/mrm.24550.
4. Detre JA, Alsop DC, Vives LR, Maccotta L, Teener JW, Raps EC. Noninvasive MRI evaluation of cerebral blood flow in cerebrovascular disease. Neurology. 1998;50(3):633-41.
5. Detre JA, Wang J. Technical aspects and utility of fMRI using BOLD and ASL. Clin Neurophysiol. 2002;113(5):621-34.
6. Bulte DP, Kelly M, Germuska M, Xie J, Chappell MA, Okell TW et al. Quantitative measurement of cerebral physiology using respiratory-calibrated MRI. Neuroimage. 2012;60(1):582-91. doi:10.1016/j.neuroimage.2011.12.017.
7. Buxton RB, Griffeth VE, Simon AB, Moradi F, Shmuel A. Variability of the coupling of blood flow and oxygen metabolism responses in the brain: a problem for interpreting BOLD studies but potentially a new window on the underlying neural activity. Front Neurosci. 2014;8:139. doi:10.3389/fnins.2014.00139.
8. Alsop DC, Detre JA, Golay X, Gunther M, Hendrikse J, Hernandez-Garcia L et al. Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: A consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia. Magn Reson Med. 2015;73(1):102-16. doi:10.1002/mrm.25197.
9. Feinberg DA, Beckett A, Chen L. Arterial spin labeling with simultaneous multi-slice echo planar imaging. Magn Reson Med. 2013;70(6):1500-6. doi:10.1002/mrm.24994.
10. Kim T, Shin W, Zhao T, Beall EB, Lowe MJ, Bae KT. Whole brain perfusion measurements using arterial spin labeling with multiband acquisition. Magn Reson Med. 2013;70(6):1653-61. doi:10.1002/mrm.24880.
11. Ivanov D, Poser BA, Huber L, Pfeuffer J, Uludag K. Optimization of simultaneous multislice EPI for concurrent functional perfusion and BOLD signal measurements at 7T. Magn Reson Med. 2017;78(1):121-9. doi:10.1002/mrm.26351.
12. Zuo Z, Wang R, Zhuo Y, Xue R, St Lawrence KS, Wang DJ. Turbo-FLASH based arterial spin labeled perfusion MRI at 7 T. PloS one. 2013;8(6):e66612. doi:10.1371/journal.pone.0066612.
13. Wang Y, Moeller S, Li X, Vu AT, Krasileva K, Ugurbil K et al. Simultaneous multi-slice Turbo-FLASH imaging with CAIPIRINHA for whole brain distortion-free pseudo-continuous arterial spin labeling at 3 and 7T. Neuroimage. 2015;113:279-88. doi:10.1016/j.neuroimage.2015.03.060.
14. Vidorreta M, Wang Z, Rodriguez I, Pastor MA, Detre JA, Fernandez-Seara MA. Comparison of 2D and 3D single-shot ASL perfusion fMRI sequences. Neuroimage. 2013;66:662-71. doi:10.1016/j.neuroimage.2012.10.087.
15. Ivanov D, Gardumi A, Haast RAM, Pfeuffer J, Poser BA, Uludag K. Comparison of 3T and 7T ASL techniques for concurrent functional perfusion and BOLD studies. Neuroimage. 2017;156:363-76. doi:10.1016/j.neuroimage.2017.05.038.