In vivo evaluation of carotid artery stenosis: Magnetic Resonance 3.0T vs Ultrasound Imaging

Quantitative characterization of carotid atherosclerosis and classication of plaques is crucial in the diagnosis and treatment planning. The degree of carotid stenosis is, up to now, considered one of the most important features for determining the risk of brain stroke. Carotid ultrasonography (US) has been shown to be a useful predictor of incident cardiovascular events. Magnetic resonance (MR) imaging is an alternative approach that can also be used to identify carotid plaque. Carotid MR imaging can accurately depict plaque components, such as the lipid core, and it can be used to identify and monitor vulnerable plaque. The aim of this study was the comparison of segmentation techniques of US images to characterize plaque morphology and composition with the 3T MR, using US image as the gold standard. This analysis was conducted on 22 patients with pathology of the carotid arteries showed an on US examination of Sovraotic Trunch. From each patient, a varying number of images has been taken to form the nal dataset. All patients underwent to a MR examination. The US data were obtained as longitudinal cross-sections using a Philips iU22 ultrasound scanner with an L9-3 probe and included B-Mode (i.e. greyscale) and Colour Doppler image sequences. Then all patients underwent to a MR examination on a 3T MR system with Sense Head coil. The same plaque was evaluated by US and MR examination, relatively to size, consistency of plaques, intima-media thickness. Concordance Correlation Coecient (CCC) was calculated on three extracted parameters to evaluate the consistency of two methods. The subjects presented clinical and vascular risk factors. Two subjects that have hypoechoic plaques not highlighted by MR were excluded. F-test no highlighted signicant variance dierences between the two methodologies (p > 0:05). The values of three plaque parameters obtained by automatic segmentation were highly signicantly correlated with those obtained from manual segmentation (r1 = 0:78, r2 = 0:84, r3 = 0:89, with p < 0:001). Our results showed a very high comparison between US and MRI examinations. From the results obtained, there were no signicant dierences between the two techniques. The minimal dierence is, probably, related to the fact that the US and MRI numerical data were obtained by the operator in a total manual modality. However, a limitation of our study is that hypoechoic plaques evaluated with US methodical are dicult to detect by MRI. The obtained results could to argue that MR examination is the most promising objective method.