ASL Acquisition & Reconstruction
Weiying Dai1

1Computer Science, State University of New York at Binghamton, NY, United States


The lecture will cover the key components of arterial spin labeling (ASL) data acquisition and reconstruction, including basic principles, different labeling approaches, background suppression techniques to improve the temporal stability of ASL signals, advanced ASL techniques, readout options, and image reconstruction.


Arterial spin labeling (ASL) is a completely noninvasive MRI technique to measure blood flow. It does not require injection of any exogenous contrast agent or radioactive tracer [1, 2]. It uses the radiofrequency (RF) pulses and gradients to label the naturally existing water spins (inversion or saturation) in the arterial blood. After the labeled spins flow into the image slices and maybe exchange with the tissue of interest, an labeled image is acquired, which represents the signals from static tissue and labeled spins. A control image without labeling is also acquired, which has the signals from static tissue and unlabeled blood. The subtraction of the two images will remove the signals from static tissue and thus reflects the amount of labeled spins, which is a proportional to the local tissue perfusion. Typically, the control-label pair images will be repeated over a time period of a few minutes to gain sufficient signal noise ratio (SNR). The averaged subtraction image is then fit into perfusion kinetic models to quantify the map of perfusion and arterial transit times. It can be used to evaluate the perfusion in the brain and other body organs (e.g. kidney). For the brain perfusion, it is frequently called cerebral blood flow (CBF).

Labeling Schemes

ASL labeling approaches can be grouped into three categories: (1) pulsed ASL (PASL), (2) (pseudo)continuous ASL ((p)CASL), and (3) velocity selective ASL (VSASL).

Pulsed ASL

In PASL, a short RF pulse (typically 10-20 ms) inverts arterial blood spins in a thick slab of tissue (typically 10-20 cm) proximal to the tissue of interest, and the labeled spins flow into the tissue of interest. PASL has a family of pulse sequences, such as flow alternating inversion recovery (FAIR) [3, 4], Signal Targeting with Alternating Radiofrequency (STAR) [5, 6], Transfer insensitive labeling technique (TILT) [7], proximal inversion with a control for off resonance effects (PICORE) [8], and quantitative imaging of perfusion using a single subtraction (QUIPSS) [9].

Continuous ASL

In CASL, a RF pulse (typically 1-3 s) inverts arterial blood spins as blood flows through a thin labeling plane. The inversion process is based on flow-driven adiabatic fast passage principle [10]. CASL was originally implemented for human use as a single-slice imaging [11], but was later extended to multi-slice imaging using double adiabatic inversion as the control condition [12]. In either implementation, one single, long continuous ASL pulse is applied.

Pseudocontinuous ASL

In PCASL [13], the long continuous labeling is replaced by a train of shaped RF pulses together with gradient pulses applied in a rapid session (e.g., every ms). The mean value of both RF and gradient pulses over time are similar to those used in CASL. It is worth noting that PCASL achieve continuous labeling of arterial spins flowing through the label plane although it uses the pulsed RF waveforms. PCASL provides superior labeling efficiency and is more compatible with modern RF transmission hardware because of its reduced duty cycle.

Velocity selective ASL

In velocity selective ASL [14, 15], a velocity selective pulse train saturates (or inverts) blood flowing above a chosen cutoff velocity. After a delay, a second velocity selective pulse train is applied with the same cutoff velocity, either within or prior to the imaging pulse sequence. Only signals from spins that have decelerated from above the cutoff velocity to below it are included in the image. This design mechanism below the cutoff velocity provides selectivity for arterial delivery, as blood on the venous side of the circulation generally accelerates with time. This saturation is typically spatially non-selective, and can even be within the imaging region. Velocity selective ASL does not suffer from long arterial transit times because the label is created much closer to the tissue compared to traditional ASL methods.

Background suppression

The ASL signal comes from inflowing blood and measured using the difference between label and control images, which includes the background signal from static tissue. The static tissue signal can be 100 times larger than the ASL signal. Subject motion, Physiological fluctuations and instrument instabilities may cause a small signal fluctuations in the background tissue signal but translate into large percentage of fluctuation to the ASL difference signal. Therefore, the signal-to-noise ratio (SNR) in ASL can be improved substantially if the signal from the static tissue can be reduced and the ASL difference signal can be left unaffected or not affected much. This can be achieved using background suppression (BS) technique. A typical BS module inserts saturation and inversion pulses in the ASL pulse sequence [16]. The timing of the saturation and inversion pulses needs to be optimized to enable the signal from the static tissue to pass near or through zero [13, 17, 18].

Advanced ASL Techniques

Flow territory mapping

Techniques have been developed to image the flow territory of single artery [19-21]. Only a single vessel got labeled either by carefully positioning of PASL labeling plane or by labeling a disc around the targeted vessel based on PCASL approach. The flow territory information is very useful for understanding collateral flow patterns and complicated vascular structures. Another way of flow territory mapping is to encode several vessels at the same time [22]. On the same labeling plane, some vessels are in a labeling condition, while other vessels are in a control condition. Sophisticated post-processing methods are needed to separate the flow territories from different vessels.

Time-encoded ASL

Arterial transit time (ATT) images from the labeling location to the image tissue can aid the optimization and quantification of ASL perfusion and may provide diagnostic information independent of perfusion. Typically, ATT measurement require several ASL measurements with different post-labeling delays and/or labeling durations, but these can be very time consuming. Hadamard encoded ASL [23-25] can speed up the ATT imaging and improve the signal to noise ratio (SNR). For Hadamard encoded scheme, each ASL period is divided into a few blocks, say 8 blocks. For each encoding, label or control is assigned to each block according to the 8-order Hadamard matrix. Each image from the Hadamard encoded acquisition is a mixture of label and control images from all the small blocks. At post-processing, the ASL image at each post-labeling delay can be separated by linear combination of the Hadamard encoded images. As can be seen, Hadamard encoded ASL can obtain the ASL images with 7 post-labeling delays in the same time as one measure the ASL image in a single post-labeling delay.


Several pulse sequences can be used as the ASL readout module after the label/control preparation sequence. 2D single-shot EPI is still commonly used ASL read out sequence for the brain [26] because of their availability on all modern MRI scanners and insensitivity to motion artifacts. However, 2D imaging results in poor BS for most slices, and longer scan time. Currently, segmented 3D sequences are the preferred methodology. Commonly used 3D segmented methods include 3D multiecho (RARE) stack of spirals [16, 27] and 3D GRASE [28, 29]. The 3D methods use a single excitation per TR and are thus optimal for BS. They provide nearly optimal SNR, and are relatively insensitive to field inhomogeneity.


All image ASL data are saved as raw echo intensities and needed to be reconstructed either with online or offline reconstruction [13, 19, 23, 30]. ASL label-control pairs are regridded into a matrix (if not acquired in a Cartesian grid), subtracted and averaged in complex k-space. The averaged ASL k-space image is Fourier transformed into complex valued image for each of the receiver coils. To improved SNR, a phased combination of different coils can be performed, rather than simple root mean square combination. Phase maps can be estimated from the low-frequency reference image (e.g. by smoothing). The final ASL difference image can be obtained by combining each coil image, weighted by the conjugate of the phase map. The ASL difference image can be taken to the next step for quantitative cerebral blood flow (CBF) maps using the kinetic models [31-33].


No acknowledgement found.


1. J.A. Detre, J.S. Leigh, D.S. Williams, and A.P. Koretsky, Perfusion imaging. Magn Reson Med, 1992. 23: p. 37-45.

2. D.S. Williams, J.A. Detre, J.S. Leigh, and A.P. Koretsky, Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci USA, 1992. 89: p. 212-16.

3. S.G. Kim, Quantification of relative cerebral blood flow change by flow-sensitive alternating inversion recovery (FAIR) technique: application to functional mapping. Magn Reson Med, 1995. 34(3): p. 293-301.

4. K.K. Kwong, D.A. Chesler, R.M. Weisskoff, K.M. Donahue, T.L. Davis, L. Ostergaard, T.A. Campbell, and B.R. Rosen, MR perfusion studies with T1-weighted echo planar imaging. Magn Reson Med, 1995. 34(6): p. 878-87.

5. R.R. Edelman, B. Siewert, M. Adamis, J. Gaa, G. Laub, and P. Wielopolski, Signal targeting with alternating radiofrequency (STAR) sequences: application to MR angiography. Magn Reson Med, 1994. 31(2): p. 233-8.

6. R.R. Edelman, B. Siewert, D.G. Darby, V. Thangaraj, A.C. Nobre, M.M. Mesulam, and S. Warach, Qualitative mapping of cerebral blood flow and functional localization with echo-planar MR imaging and signal targeting with alternating radio frequency. Radiology, 1994. 192(2): p. 513-20.

7. X. Golay, M. Stuber, K.P. Pruessmann, D. Meier, and P. Boesiger, Transfer insensitive labeling technique (TILT): application to multislice functional perfusion imaging. J Magn Reson Imaging, 1999. 9(3): p. 454-61.

8. E.C. Wong, R.B. Buxton, and L.R. Frank, Implementation of quantitative perfusion imaging techniques for functional brain mapping using pulsed arterial spin labeling. NMR Biomed, 1997. 10: p. 237-249.

9. E.C. Wong, R.B. Buxton, and L.R. Frank, Quantitative imaging of perfusion using a single subtraction (QUIPSS and QUIPSS II). Magn Reson Med, 1998. 39(5): p. 702-8.

10. W.T. Dixon, L.N. Du, D.D. Faul, M. Gado, and S. Rossnick, Projection angiograms of blood labeled by adiabatic fast passage. Magn Reson Med, 1986. 3(3): p. 454-62.

11. J.A. Detre, W. Zhang, D.A. Roberts, A.C. Silva, D.S. Williams, D.J. Grandis, A.P. Koretsky, and J.S. Leigh, Tissue specific perfusion imaging using arterial spin labeling. NMR Biomed, 1994. 7(1-2): p. 75-82.

12. D.C. Alsop and J.A. Detre, Multisection cerebral blood flow imaging with continous arterial spin labeling. Radiology, 1998. 208: p. 410-416.

13. W. Dai, D. Garcia, C. de Bazelaire, and D.C. Alsop, Continuous flow-driven inversion for arterial spin labeling using pulsed radio frequency and gradient fields. Magn Reson Med, 2008. 60(6): p. 1488-97.

14. E.C. Wong, M. Cronin, W.C. Wu, B. Inglis, L.R. Frank, and T.T. Liu, Velocity-selective arterial spin labeling. Magn Reson Med, 2006. 55(6): p. 1334-41.

15. G. Duhamel, C. de Bazelaire, and D.C. Alsop, Evaluation of systematic quantification errors in velocity-selective arterial spin labeling of the brain. Magn Reson Med, 2003. 50(1): p. 145-53.

16. F.Q. Ye, J.A. Frank, D.R. Weinberger, and A.C. McLaughlin, Noise reduction in 3D perfusion imaging by attenuating the static signal in arterial spin tagging (ASSIST). Magn Reson Med, 2000. 44(1): p. 92-100.

17. D.M. Garcia, C. de Bazelaire, and D.C. Alsop, Pseudo-continuous flow driven adiabatic inversion for arterial spin labeleing. Proceedings of the 13th ISMRM Annual Meeting, Miami, FL, USA, 2005: p. 9.

18. N. Maleki, W. Dai, and D.C. Alsop, Optimization of background suppression for arterial spin labeling perfusion imaging. Magma, 2012. 25(2): p. 127-33.

19. W. Dai, P.M. Robson, A. Shankaranarayanan, and D.C. Alsop, Modified pulsed continuous arterial spin labeling for labeling of a single artery. Magn Reson Med, 2010. 64(4): p. 975-82.

20. M. Helle, D.G. Norris, S. Rufer, K. Alfke, O. Jansen, and M.J. van Osch, Superselective pseudocontinuous arterial spin labeling. Magn Reson Med, 2010. 64(3): p. 777-86.

21. J. Hendrikse, J. van der Grond, H. Lu, P.C. van Zijl, and X. Golay, Flow territory mapping of the cerebral arteries with regional perfusion MRI. Stroke, 2004. 35(4): p. 882-7.

22. E.C. Wong, Vessel-encoded arterial spin-labeling using pseudocontinuous tagging. Magn Reson Med, 2007. 58(6): p. 1086-91.

23. W. Dai, A. Shankaranarayanan, and D.C. Alsop, Volumetric measurement of perfusion and arterial transit delay using hadamard encoded continuous arterial spin labeling. Magnetic resonance in medicine, 2013. 69(4): p. 1014-22.

24. M. Gunther, Highly efficient accelerated acquisition of perfusion inflow series by cycled arterial spin labeling. Proceedings of the 15th Annual Meeting of ISMRM, 2007: p. p380.

25. J.A. Wells, M.F. Lythgoe, D.G. Gadian, R.J. Ordidge, and D.L. Thomas, In vivo Hadamard encoded continuous arterial spin labeling (H-CASL). Magn Reson Med, 2010. 63(4): p. 1111-8.

26. G.H. Jahng, M.W. Weiner, and N. Schuff, Improved arterial spin labeling method: applications for measurements of cerebral blood flow in human brain at high magnetic field MRI. Med Phys, 2007. 34(11): p. 4519-25.

27. M. Vidorreta, Z. Wang, I. Rodriguez, M.A. Pastor, J.A. Detre, and M.A. Fernandez-Seara, Comparison of 2D and 3D single-shot ASL perfusion fMRI sequences. Neuroimage, 2013. 66: p. 662-71.

28. M.A. Fernandez-Seara, Z. Wang, J. Wang, H.Y. Rao, M. Guenther, D.A. Feinberg, and J.A. Detre, Continuous arterial spin labeling perfusion measurements using single shot 3D GRASE at 3 T. Magn Reson Med, 2005. 54(5): p. 1241-7.

29. M. Gunther, K. Oshio, and D.A. Feinberg, Single-shot 3D imaging techniques improve arterial spin labeling perfusion measurements. Magn Reson Med, 2005. 54(2): p. 491-8.

30. P.M. Robson, A.J. Madhuranthakam, W. Dai, I. Pedrosa, N.M. Rofsky, and D.C. Alsop, Strategies for reducing respiratory motion artifacts in renal perfusion imaging with arterial spin labeling. Magn Reson Med, 2009. 61(6): p. 1374-87.

31. D.C. Alsop and J.A. Detre, Reduced transit-time sensitivity in noninvasive magnetic resonance imaging of Human cerebral blood flow. J Cereb Blood Flow Metab, 1996. 16: p. 1236-49.

32. R.B. Buxton, L.R. Frank, E.C. Wong, B. Siewert, S. Warach, and R.R. Edelman, A general kinetic model for quantitative perfusion imaging with arterial spin labeling. Magn Reson Med, 1998. 40: p. 383-96.

33. J. Wang, D.C. Alsop, L. Li, J. Listerud, J.B. Gonzalez-At, M.D. Schnall, and J.A. Detre, Comparison of quantitative perfusion imaging using arterial spin labeling at 1.5 and 4.0 Tesla. Magn Reson Med, 2002. 48(2): p. 242-54.

Proc. Intl. Soc. Mag. Reson. Med. 26 (2018)