Use of multi‐velocity encoding 4D flow MRI to improve quantification of flow patterns in the aorta. Issue 2 (30th June 2015)

• Record Type: Journal Article
• Authors: Callaghan, Fraser M.; Kozor, Rebecca; Sherrah, Andrew G.; Vallely, Michael; Celermajer, David; Figtree, Gemma A.; Grieve, Stuart M.
• Is Part Of: Journal of magnetic resonance imaging. Volume 43:Issue 2(2016)
• Page Start: 352
• Page End: 363
• Publication Date: 2015-06-30
• Subjects: 4D‐flow imaging -- magnetic resonance imaging -- cardiovascular magnetic resonance -- flow measurement; Magnetic resonance imaging -- Periodicals; 616
• DOI: 10.1002/jmri.24991
• Languages: English
• ISSNs: 1053-1807

Abstract

Abstract : Purpose: To show that the use of a multi‐velocity encoding (VENC) 4D‐flow approach offers significant improvements in the characterization of complex flow in the aorta. Four‐dimensional flow magnetic resonance imaging (MRI) (4D‐flow) can be used to measure complex flow patterns and dynamics in the heart and major vessels. The quality of the information derived from these measures is dependent on the accuracy of the vector field, which is limited by the vector‐to‐noise ratio. Materials and Methods: A 4D‐flow protocol involving three different VENC values of 150, 60, and 20 cm/s was performed on six control subjects and nine patients with type‐B chronic aortic dissection at 3T MRI. Data were processed using a single VENC value (150 cm/s) or using a fused dataset that selected the lowest appropriate VENC for each voxel. Performance was analyzed by measuring spatial vector angular correlation, magnitude correlation, temporal vector conservation, and "real‐world" streamline tracing performance. Results: The multi‐VENC approach provided a 31% improvement in spatial and 53% improvement in temporal precision of velocity vector measurements during the mid‐late diastolic period, where 99% of the flow vectors in the normal aorta are below 20 cm/s. In low‐flow conditions this resulted in practical improvements of greater than 50% in pathline tracking and streamline tracing quantified by streamline curvature measurements. Conclusion: A multi‐VENC 4D‐flow approach provides accurate vector data across normal physiological velocities observed in the aorta, dramatically improving outputs such as pathline tracking, streamline estimation, and further advanced analyses.J. Magn. Reson. Imaging 2016;43:352–363.