Automated removal of spurious intermediate cerebral blood flow volumes improves image quality among older patients: A clinical arterial spin labeling investigation. Issue 5 (15th April 2015)

• Record Type: Journal Article
• Authors: Shirzadi, Zahra; Crane, David E.; Robertson, Andrew D.; Maralani, Pejman J.; Aviv, Richard I.; Chappell, Michael A.; Goldstein, Benjamin I.; Black, Sandra E.; MacIntosh, Bradley J.
• Is Part Of: Journal of magnetic resonance imaging. Volume 42:Issue 5(2015)
• Page Start: 1377
• Page End: 1385
• Publication Date: 2015-04-15
• Subjects: Magnetic resonance imaging -- Periodicals; 616
• DOI: 10.1002/jmri.24918
• Languages: English
• ISSNs: 1053-1807

Abstract

<abstract abstract-type="main"> <title> <x xml:space="preserve">Abstract</x> </title> <sec id="jmri24918-sec-0001" sec-type="section"> <title>Purpose</title> <p>To evaluate the impact of rejecting intermediate cerebral blood flow (CBF) images that are adversely affected by head motion during an arterial spin labeling (ASL) acquisition.</p> </sec> <sec id="jmri24918-sec-0002" sec-type="section"> <title>Materials and Methods</title> <p>Eighty participants were recruited, representing a wide age range (14–90 years) and heterogeneous cerebrovascular health conditions including bipolar disorder, chronic stroke, and moderate to severe white matter hyperintensities of presumed vascular origin. Pseudocontinuous ASL and <italic>T</italic><sub>1</sub>‐weigthed anatomical images were acquired on a 3T scanner. ASL intermediate CBF images were included based on their contribution to the mean estimate, with the goal to maximize CBF detectability in gray matter (GM). Simulations were conducted to evaluate the performance of the proposed optimization procedure relative to other ASL postprocessing approaches. Clinical CBF images were also assessed visually by two experienced neuroradiologists.</p> </sec> <sec id="jmri24918-sec-0003" sec-type="section"> <title>Results</title> <p>Optimized CBF images (CBF<sub>opt</sub>) had significantly greater agreement with a synthetic ground truth CBF image and greater CBF detectability relative to the other ASL analysis methods (<italic>P</italic> < 0.05). Moreover, empirical CBF<sub>opt</sub> images showed a significantly improved signal‐to‐noise ratio relative to CBF images obtained from other postprocessing approaches (mean: 12.6%; range 1% to 56%; <italic>P</italic> < 0.001), and this improvement was age‐dependent (<italic>P</italic> = 0.03). Differences between CBF images from different analysis procedures were not perceptible by visual inspection, while there was a moderate agreement between the ratings (κ = 0.44, <italic>P</italic> < 0.001).</p> </sec> <sec id="jmri24918-sec-0004" sec-type="section"> <title>Conclusion</title> <p>This study developed an automated head motion threshold‐free procedure to improve the detection of CBF in GM. The improvement in CBF image quality was larger when considering older participants. J. Magn. Reson. Imaging 2015;42:1377–1385.</p> </sec> </abstract>