Continuous Arterial Spin Labeling (CASL)
What is CASL?
|
|
The technique we now know of as CASL (Continuous Arterial Spin Labeling) was the first true ASL method, described by Williams, Detre, and colleagues in the early 1990s. CASL used a constant, low-amplitude, continuous RF-pulse in conjunction with an imaging gradient to produce spin inversion of flowing blood proximal to the imaged slice. When these inverted spins flowed into the imaged slice they reduced its signal intensity slightly, detectable by a subtraction technique. Although now primarily of historical interest, the fundamental concepts embodied by CASL are needed to understand newer and more sophisticated ASL techniques.
The inversion pulses used in CASL (and all ASL techniques) give rise to an interesting (and problematic) MR phenomenon, magnetization transfer (MT). As described in a prior Q&A, MT occurs when an RF-pulse centered in the inversion slab "spills over" and indirectly lowers signal in the imaged slice. The MT effect takes place via off-resonance water-macromolecular interactions and is entirely separate from ASL signal changes caused by flow.
The MT phenomenon occurs in normal imaging, but usually is not noticeable because it affects tissue signal in adjacent slices by only a few percent. In ASL, however, where the change in signal between tagged and control images is less than 1%, unwanted MT effects may overwhelm detection of the desired flow-related changes.
All ASL techniques must therefore address how to cancel or diminish this MT effect. An illustration of the "MT Problem" and two solutions possible with the CASL method is given below.
The MT phenomenon occurs in normal imaging, but usually is not noticeable because it affects tissue signal in adjacent slices by only a few percent. In ASL, however, where the change in signal between tagged and control images is less than 1%, unwanted MT effects may overwhelm detection of the desired flow-related changes.
All ASL techniques must therefore address how to cancel or diminish this MT effect. An illustration of the "MT Problem" and two solutions possible with the CASL method is given below.
The MT problem and two CASL solutions.
In the top row the MT Problem is illustrated. A single inversion pulse applied off-resonance (at a different resonance frequency and location) reduces signal from the imaged slab. This confounds calculation of the desired signal change (Signalcontrol − Signalinversion) reflecting perfusion.
In the second row the original CASL solution is illustrated. Here the control image is obtained using a second inversion pulse identical to the first but applied on the opposite side of the imaged slice. The MT effects within the imaged slice are the same between Labeling and Control; they are eliminated when the two images are subtracted.
"CASL Solution #2" a modification of the original technique by Alsop and Detre (1998). Here, the inversion pulse played out during the Control sequence is split into two parts (sidebands) by amplitude modulation. Since the sidebands have half the amplitude of the original pulse and are very close together, the MT effect on the imaged slice remains unchanged and disappears when the sequences are subtracted. The split pulse results in no net inversion of flowing spins in the Control sequence since the pair acts as two 180° pulses = 360° (a double inversion) when placed side-by-side.
In the second row the original CASL solution is illustrated. Here the control image is obtained using a second inversion pulse identical to the first but applied on the opposite side of the imaged slice. The MT effects within the imaged slice are the same between Labeling and Control; they are eliminated when the two images are subtracted.
"CASL Solution #2" a modification of the original technique by Alsop and Detre (1998). Here, the inversion pulse played out during the Control sequence is split into two parts (sidebands) by amplitude modulation. Since the sidebands have half the amplitude of the original pulse and are very close together, the MT effect on the imaged slice remains unchanged and disappears when the sequences are subtracted. The split pulse results in no net inversion of flowing spins in the Control sequence since the pair acts as two 180° pulses = 360° (a double inversion) when placed side-by-side.
CASL techniques are now primarily of historical and didactic interest, having fallen out of favor by the late 1990s. The main problems with CASL were the high tissue energy deposition and transmitter duty-cycle requirements accompanying the use of continuous RF-pulses. Thus came PASL, a family of ASL methods using pulsed (rather than continuous) RF-excitation. PASL and its variants are discussed in the next Q&A.
References
Alsop DC, Detre JA. Multisection cerebral blood flow MR imaging with continuous arterial spin labeling. Radiology 1998; 208:410-416. (improvement of the CASL technique using an amplitude modulated inversion pulse to self-correct for off-resonance effects)
Williams DS, Detre JA, Leigh JS, Koretsky AP. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci USA 1992; 89:212-216. (first demonstration of ASL, using a single-slice continuous technique in a rat brain, later known as CASL)
Wong EC, Buxton RB, Frank LR. Implementation of quantitative perfusion imaging techniques for functional brain mapping using pulsed arterial spin labeling. NMR in Biomed 1997; 10:237-249. (comparison of CASL to pulsed methods)
Alsop DC, Detre JA. Multisection cerebral blood flow MR imaging with continuous arterial spin labeling. Radiology 1998; 208:410-416. (improvement of the CASL technique using an amplitude modulated inversion pulse to self-correct for off-resonance effects)
Williams DS, Detre JA, Leigh JS, Koretsky AP. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci USA 1992; 89:212-216. (first demonstration of ASL, using a single-slice continuous technique in a rat brain, later known as CASL)
Wong EC, Buxton RB, Frank LR. Implementation of quantitative perfusion imaging techniques for functional brain mapping using pulsed arterial spin labeling. NMR in Biomed 1997; 10:237-249. (comparison of CASL to pulsed methods)
Related Questions
Can you briefly explain the difference between the various ASL methods? Which is the best?
What is pCASL and how does it differ from CASL and PASL?
Can you briefly explain the difference between the various ASL methods? Which is the best?
What is pCASL and how does it differ from CASL and PASL?