Correlation analysis of stenotic aortic valve flow patterns using phase contrast MRI

Emily A. Waters, Shelton D. Caruthers, Samuel A. Wickline

Research output: Contribution to journalArticlepeer-review

18 Scopus citations

Abstract

Phase contrast MRI is an emerging tool for evaluating valvular pathology. However, the effects of variable image position and valvular pathology on velocity measurements have not been explored. We compare velocity maps with correlation methods, used in image processing to align images and quantify their similarity, to define these effects on calculations of valve orifice area. Quantitative flow images were acquired in four parallel planes (2 in aortic root, 2 in outflow tract), in patients (n = 22) with aortic stenosis. Velocity-time integrals (VTIs) were computed and cross-correlations were performed to quantitatively compare the shapes and relative positions of three-dimensional flow profiles between scans at various positions. Supravalvular VTIs correlated well with one another (R = 0.96), with comparable values. The two subvalvular VTIs exhibited a linear relationship (R = 0.93) but with a 23% difference in mean values. Cross-correlations between supravalvular levels were maximized at (0, 0) offset (indicating concentrically aligned jets) for 19/23 patients, with an average maximum value of 0.957 ± 0.028; the average for the remainder was 0.800 ± 0.037. For subvalvular levels, all cross-correlations were maximized at (0, 0) with average maximum 0.968 ± 0.160. The aortic VTI measurements were comparable, indicating relative insensitivity to the position of the imaging plane; in the LVOT, measurements were only somewhat position-dependent. We conclude that phase contrast MRI is a robust tool for the evaluation of aortic stenosis.

Original languageEnglish (US)
Pages (from-to)878-887
Number of pages10
JournalAnnals of Biomedical Engineering
Volume33
Issue number7
DOIs
StatePublished - Jul 2005
Externally publishedYes

Funding

The authors thank Mr. Todd Williams and Ms. Mary Watkins for data acquisition, as well as Ms. Katherine Lehr and Ms. Peggy Brown for patient recruitment. This work was supported in part by NIH grant HL-42950, a research grant from Philips Medical Systems, the Edith and Alan Wolff Charitable Fund, and the Florindon Foundation.

Keywords

  • Aorta
  • Aortic stenosis
  • Heart valve disease
  • Magnetic resonance imaging
  • Phase contrast

ASJC Scopus subject areas

  • Biomedical Engineering

Fingerprint

Dive into the research topics of 'Correlation analysis of stenotic aortic valve flow patterns using phase contrast MRI'. Together they form a unique fingerprint.

Cite this