The intima-media thickness (IMT) of the common carotid artery (CCA) is widely used as an early indicator of the development of cardiovascular disease (CVD). It was proposed but not thoroughly investigated that the media layer (ML) thickness (MLT), its composition, and its texture may be indicative of cardiovascular risk and for differentiating between patients with high and low risk. In this study, we investigate an automated method for segmenting the ML and the intima layer (IL) and measurement of the MLT and the intima layer thickness (ILT) in ultrasound images of the CCA. The snakes segmentation method was used and was evaluated on 100 longitudinal ultrasound images acquired from asymptomatic subjects, against manual segmentation performed by a neurovascular expert. The mean standard deviation (sd) for the first and second sets of manual and the automated IMT, MLT, and ILT measurements were 0.71 0.17 mm, 0.72 0.17 mm, 0.67 0.12 mm; 0.25 0.12 mm, 0.27 0.14 mm, 0.25 0.11 mm; and 0.43 0.10 mm, 0.44 0.13 mm, and 0.42 0.10 mm, respectively. There was overall no significant difference between the manual and the automated IMC, ML, and IL segmentation measurements. Therefore, the automated segmentation method proposed in this study may be used successfully in the measurement of the MLT and ILT complementing the manual measurements. MLT was also shown to increase with age (for both the manual and the automated measurements). Future research will incorporate the extraction of texture features from the segmented ML and IL bands, which may indicate the risk of future cardiovascular events. However, more work is needed for validating the proposed technique in a larger sample of subjects.
|Number of pages||12|
|Journal||IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control|
|Publication status||Published - May 2009|