Atherosclerotic carotid plaque texture features variability in ultrasound videos

Ultrasound video of the common carotid artery (CCA) combined with texture feature analysis has the potential in differentiating between asymptomatic and symptomatic atherosclerotic carotid plaque and is used widely for the assessment of cardiovascular disease. Texture features were extracted from the segmented atherosclerotic carotid plaque for each video frame, from 10 ultrasound videos of the CCA, acquired from 5 asymptomatic and 5 symptomatic subjects. All videos were resolution normalized, despeckled and the plaque was automatically segmented by snakes. The cardiac cycle in each video was identified and the M-mode was generated, thus identifying systolic and diastolic states. From the segmented plaque, different texture features were extracted and studied throughout the cardiac cycle. It is shown that: (i) the plaque gray-scale median (GSM) for the asymptomatic group was significantly higher when compared to the symptomatic group and the same applies for the %CV and the %MAE; (ii) similar to this trend was also the case for features angular second moment, contrast, entropy, and difference variance; (iii) there is plaque GSM variability per frame throughout the cardiac cycle, and this variability differs between the systolic and diastolic states. It is anticipated that the proposed method may help the physician in the assessment of cardiovascular video analysis. However, exhaustive evaluation has to be carried out with more videos and texture features