Time Varying Velocity Labeling Efficiency Correction in Arterial Spin Labeling
Adam Bush1, Yaqoing Chia2, Julie Coloigner2, Thomas Coates3, Natasha Lepore4, and John Wood1

1Biomedical Engineering/ Cardiology, University of Southern California/ Children's Hospital Los Angeles, Los Angeles, CA, United States, 2Cardiology, Children's Hospital Los Angeles, Los Angeles, CA, United States, 3Hematology, Children's Hospital Los Angeles, Los Angeles, CA, United States, 4Children's Hospital Los Angeles, Los Angeles, CA, United States

Synopsis

In this study we propose a time varying correction technique for pseudo continuous arterial spin labeling efficiency. We compare this correction method to other correction methods and phase contrast MRI.

Introduction

Pseudo Continuous Arterial Spin Labeling (PCASL) has emerged as the recommended labeling approach for clinical quantification of cerebral blood flow (CBF) using ASL[1]. Unfortunately, problems regarding accurate quantification using PCASL limit its clinical use. One of the largest problems is that of labeling efficiency (LE). Labeling efficiency represents the inversion rate of spins as they flow across the labeling plane and is highly dependent on blood velocity and off resonance in the labeled vessel. Inefficient labeling will lead to low signal to noise and underestimations of CBF. Several groups have developed strategies to address this problem however few of these approaches are widely adopted [3,4]. Further, the effect of pulsatile blood flow on LE has yet to be studied in simulation. As result we modelled LE in laminar and pulsatile flow using Bloch simulations, measured global CBF using phase contrast (PC) and compared different LE correction strategies in healthy controls and patients with sickle cell disease (SCD).

Methods

This study was performed under an IRB approved protocol with informed consent/assent. Imaging was performed on a Philips 3T Achieva scanner with 8 channel head coil. Study participants consisted of 8 SCD and 9 healthy controls. A single, orthogonal ungated and gated PC scan was used to extract the average velocity and the velocity over 30 cardiac cycles, respectively, in the cerebral feeding vessels. Repetition time 12.3 ms, echo time 7.5 ms, field of view 260 mm, thickness 5 mm, signal averages 10, acquisition matrix 204 x 201, reconstruction matrix 448 x 448, bandwidth 244 Hz/pixel, and velocity encoding gradient of 200 cm/s. Parameters for PCASL efficiency Bloch simulations were as follows: the unbalanced gradient PCASL consisted of 50microsecond Hanning-shaped RF pulses, with maximum gradient amplitude of 10G/cm and mean gradient of 1 G/cm over the pulse interval of 1 ms. Ungated average velocity was used to calculate LE. Cardiac cycle velocities were used to simulate magnetization evolution throughout the cardiac cycle on a voxelwise basis. Results were flow weighted and an aggregate time varying velocity (TVV) LE was calculated. A multishot, GRASE PCASL sequence was used in controls and patients FOV 220, thickness 10mm, voxelsize 3.44x3.4, label duration 1.6 seconds and labeling delay 2.0seconds.

Results

Figure 3 shows PC and ASL CBF quantification with respect to one another. Figure 4 displays the Bland Altman statistics for CBF with 1) LE equal to .85; 2)LE corrected with average velocity; 3)LE using a TVV correction. The mean difference between 1) and PC MRI is 15.6 ml/100g/min,which is highest compared to -3.6 and -0.9 ml/100g/min. The static LE efficiency underestimated the CBF measured by PC in our cohort (p<0.05). There was no statistical difference between PC and either ASL velocity correction method (p>0.05). Average velocity and TVV correction methods were statistically different from one another, reflecting a bias of 2.9 ml/100/min. The precision of the average velocity and TVV method is increased compared to the static method as observed in the lower variance.

Discussion

In this study we demonstrate it is possible to correct for labeling inefficiencies using PC MRI and Bloch simulations. These data suggest that the precision and accuracy of ASL CBF quantification can be significantly improved with the use of LE velocity correction. In this paper, we propose two separate techniques that can be used for LE correction. Though TVV correction is a more rigorous correction method, it failed to improve on the gains made with an average velocity correction. Given its increased scan time and computational requirements, we suggest an average velocity method be utilized. We are not the first group to utilize a velocity correction for LE [4, 6]. However, we are the first to numerically demonstrate the effects of cardiac pulsatility on PCASL labeling efficiency and measure it in situ using a recommended ASL sequence [1]. We realize PC MRI is not the gold standard to measure CBF and may have its own bias. Despite this, limitation we conclusively demonstrate the importance of velocity correction. This correction becomes increasingly important in patients anemia induced hyperemia, like those with SCD. This work is compulsory if PCASL is ever to leave the fully be adopted as a clinical techinque.

Acknowledgements

Special thanks to Jon Chia at Philips Healthcare for technical support and pulse sequence implementation on our Philips 3T Achieva scanner.

This work was supported by the CTSI Translational Science, CTSI Neuropsychology core, and National Heart Lung and Blood Institute 1U01HL117718-01.

References

[1] David C et. al., “Recommended implementation of arterialspin-labeled perfusion mri for clinical applications:A consensus of the ismrm perfusion study group and the european consortium for asl in dementia,” MagneticResonance in Medicine, vol. 73, no. 1, pp. 102–116, 2015.

[2] Wen-Chau W et. al., “A theoretical and experimentalinvestigation of the tagging efficiency of pseudocontinuousarterial spin labeling,” vol. 58, no. 5, pp. 1020–1027, 2007.

[3] Youngkyoo J et. al., “Multiphase pseudocontinuousarterial spin labeling (mp-pcasl) for robust quantificationof cerebral blood flow,” Magnetic Resonance inMedicine, vol. 64, no. 3, pp. 799–810, 2010.

[4] Sina A et. al., “Estimation of labeling efficiency inpseudocontinuous arterial spin labeling,” Magnetic resonancein medicine, vol. 63, no. 3, pp. 765–771, 2010.

[5] Alperin N et. al., “Comparison between total cbf valuesmeasured by asl and phase contrast over increased rangeof cbf values,” .

[6] Adam A et. al., “Comparison between total cbf valuesmeasured by asl and phase contrast over increased rangeof cbf values,” 2015.

[7] Gevers S et. al., “Arterial spin labeling measurement ofcerebral perfusion in children with sickle cell disease,”Journal of Magnetic Resonance Imaging, vol. 35, no. 4,pp. 779–787, 2012.

Figures

Spatial velocity profile at different times throughout the cardiac cycle measured via gated phase contrast A-F.

Representative plot of average spatial velocity across the cardiac cycle super imposed on top of time varying labeling efficiency calculation using Bloch simulation.

Cerebral blood flow as measured by phase contrast plotted against cerebral blood flow as measured by ASL using separate labeling efficiency correction methods.

Bland Altman plot of percent difference in labeling efficiency correction techniques vs phase contrast.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
2881