Multisite trial of MR flow measurement: Phantom and protocol design

Clinical Neurosciences, King's College London, London, England.
Journal of Magnetic Resonance Imaging (Impact Factor: 3.21). 05/2005; 21(5):620-31. DOI: 10.1002/jmri.20311
Source: PubMed


To describe a portable, easily assembled phantom with well-defined bore geometry together with a series of tests that will form the basis of a standardized quality assurance protocol in a multicenter trial of flow measurement by the MR phase mapping technique.
The phantom consists of silicone polymer layers containing parallel straight and stenosed flow channels in one layer and a U-bend in a second layer, separated by hermetically sealed agarose slabs. The phantom is constructed by casting low melting-point metal in an aluminum mold precisely milled to the desired geometry, and then using the low melting-point metal core as a negative around which the silicone is allowed to set. By melting out the metal, the flow channels are established. The milled aluminum mold is reusable, ensuring faithful reproduction of the flow geometry for all phantoms thus produced. The agarose layers provide additional loading and static background signal for background correction. With the use of the described phantom, one can evaluate flow measurement accuracy and repeatability, as well as the influence of several imaging geometry factors: slice offset, in-plane position, and slice-flow obliquity.
The new phantom is compact and portable, and is well suited for reassembly. We were able to demonstrate its facility in a battery of tests of interest in evaluating MR flow measurements.
The phantom is a robust standardized test object for use in a multicenter trial. Such a trial, to investigate the performance of MR flow measurement using the phantom and the tests we describe, has been initiated.

1 Follower
11 Reads
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Cardiac resynchronization therapy (CRT) has recently emerged as an effective treatment option for heart failure patients with dyssynchrony. Patients have traditionally been chosen for CRT based on a prolonged QRS interval. However, this selection method is far from ideal, as approximately 30% of those receiving CRT do not show any clinical improvement. Tissue Doppler imaging (TDI) suggests that one of the best predictors of response to CRT is the underlying level of mechanical dyssynchrony in the myocardial wall prior to CRT. As a result, there has been growing interest in direct imaging of the myocardial wall. Because myocardial contraction is a complex, three-dimensional movement, providing an accurate picture of myocardial wall motion can be challenging. Echocardiography initially emerged as the modality of choice, but the long list of limitations (limited echocardiographic windows, one direction of motion, poor reproducibility) has fostered interest in exploring the use of MR for myocardial wall imaging. Although MR presents some unique drawbacks (expensive equipment, longer imaging times), it is able to overcome many of the limitations of TDI. In particular, Phase Velocity Mapping (MR PVM) can provide a complete, three-directional description of motion throughout the entire myocardial wall at high spatial and temporal resolution. The overall goal of this project was to develop a patient-selection method for CRT based on myocardial wall velocities acquired with MR PVM. First the image acquisition and post-processing protocols for MR PVM imaging of myocardial tissue were developed. A myocardial motion phantom was used to verify the accuracy of, and optimize the acquisition parameters for, the developed MR PVM sequence. Excellent correlation was demonstrated between longitudinal myocardial velocity curves acquired with the optimized MR PVM sequence and Tissue Doppler velocities. A database describing the normal myocardial contraction pattern was constructed. A small group of dyssynchrony patients was compared to the normal database, and several areas of delayed contraction were identified in the patients. Furthermore, significantly higher levels of dyssynchrony were detected in the patients than the normal volunteers. Finally, a method for computing transmural, endocardial, and epicardial, radial strains and strain rates from MR PVM velocity data was developed Ph.D. Committee Chair: John Oshinski; Committee Member: Angel R. Leon; Committee Member: Oskar Skrinjar; Committee Member: Paul J. Benkeser; Committee Member: Robert Eisner
    Preview · Article ·
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper considers a near wall 3-D computational fluid dynamics model of the Karman Vortex in a cylindrical pipe, but in a medical imaging context, using a commercial finite volume fluids code (CFX 5.3, AEA Technology, Oxford, UK). Convergence was established using standard mesh invariance tests and the results were validated by comparison with data from published texts. However, the near wall conditions complicated comparison with Strouhal and Reynolds number data. Experimental validation was performed by comparing the computational results with fluorescent dye injection flow visualization. The dye patterns revealed a shedding periodicity of 0.84±0.07 seconds, as compared to 0.9 seconds obtained from the computational fluid dynamics. Comparison with a virtual flow visualization from the computed velocity dataset also demonstrated good agreement. Accurate description of the full flow field has implications for medical imaging since it offers a 'calibrated' standard by which the potency of flow-imaging systems may be judged.
    No preview · Conference Paper · Jan 2008
  • [Show abstract] [Hide abstract]
    ABSTRACT: A compact flow simulator for generating accurate pulsatile flow in a clinical magnetic resonance imaging (MRI) environment has been developed and integrated with a data acquisition (DAQ) system for recording scanner system activity and physiological waveforms. The flow simulator is relatively inexpensive, easy to construct and is controlled from a standard PC. The flow simulator is robust to repeated disassembly and reassembly (normalised cross-correlation 0.97) and generates accurate pulsatile flow (normalised cross-correlation 0.94). The DAQ system was used to monitor a standard MRI pulse sequence used in MR angiography and latent delays in the scanner gating subsystem.
    No preview · Article · May 2008 · Medical & Biological Engineering & Computing
Show more