Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Bright blood cine images acquired using magnetic resonance imaging contain simple contrast that is tractable to automated analysis, which can be used to derive a measure of arterial compliance that is known to correlate with disease severity. The purpose of this work was to evaluate whether automated methods could be used reliably on a clinically relevant population, and to assess the precision of these measurements so that it could be compared with expert manual assessment. In this paper we apply an algorithm similar to that used by Krug et al., and the exact processing steps are described in detail to allowing easy reproduction of our methods. Phantoms of different sizes have been assessed and the MRI measurements are found to correlate well (r = 0.9998) with physical measurement. Reproducibility assessment was performed on 33 CAD subjects in three anatomical locations along the aorta. Six normal volunteers and ten patients with more severe aortic plaques were investigated to assess reproducibility and sensitivity to pathological changes, respectively. The performance was also assessed on carotid vessels in 40 patients with known arterial plaques. In the human aorta the method is found to be robust (failing in only 7% of cases, all due to clear errors with image acquisition), and to be quantifiably consistent with expert clinical measurement, but showing smaller errors than that approach [<1.21% (5.62 mm(2)) manual vs. <0.58% (2.71 mm(2)) automated, for the aortic area] and with reduced bias, and operated correctly in advanced disease. We have proved over a large number of subjects the superiority of this automated method for evaluating dynamic area changes over the Gold-standard manual approach.

Original publication

DOI

10.1007/s10554-009-9495-5

Type

Journal article

Journal

Int J Cardiovasc Imaging

Publication Date

12/2009

Volume

25

Pages

797 - 808

Keywords

Algorithms, Aorta, Aortic Diseases, Automation, Laboratory, Carotid Arteries, Carotid Artery Diseases, Compliance, Humans, Image Interpretation, Computer-Assisted, Magnetic Resonance Angiography, Magnetic Resonance Imaging, Cine, Models, Cardiovascular, Phantoms, Imaging, Predictive Value of Tests, Reproducibility of Results