Texture Feature Variability in Ultrasound Video of the Atherosclerotic Carotid Plaque

Abstract

The objective of this paper was to investigate texture feature variability in ultrasound video of the carotid artery during the cardiac cycle in an attempt to define new discriminatory biomarkers of the vulnerable plaque. More specifically, in this paper, 120 longitudinal ultrasound videos, acquired from 40 normal (N) subjects from the common carotid artery and 40 asymptomatic (A) and 40 symptomatic (S) subjects from the proximal internal carotid artery were investigated. The videos were intensity normalized and despeckled, and the intima-media complex (IMC) (from the N subjects) and the atherosclerotic carotid plaques (from the A and S subjects) were segmented from each video, in order to extract the M-mode image, and the texture features associated with cardiac states of systole and diastole. The main results of this paper can be summarized as follows: 1) texture features varied significantly throughout the cardiac cycle with significant differences identified between the cardiac systolic and cardiac diastolic states; 2) gray scale median was significantly higher at cardiac systole versus diastole for the N, A, and S groups investigated; 3) plaque texture features extracted during the cardiac cycle at the systolic and diastolic states were statistically significantly different between A and S subjects (and can thus be used to discriminate between A and S subjects successfully). The combination of systolic and diastolic features yields better performance than those alone. It is anticipated that the proposed system may aid the physician in clinical practice in classifying between N, A, and S subjects using texture features extracted from ultrasound videos of IMC and carotid artery plaque. However, further evaluation has to be carried out with more videos and additional features.

Keywords: Ultrasound video; cardiovascular disease; carotid plaque; texture analysis; texture variability.

FIGURE 1.

Illustration of the steps followed for segmenting the plague, computing the M-mode image and the state diagram from a video of an asymptomatic male subject at the age of 69 with 65% stenosis. a) frame of the B-mode ultrasound video (where the original image before normalisation and despeckle filtering is shown), and segmentation of the plaque at the far wall of the left carotid artery. b) frame of the video after normalisation and despeckle filtering with 4 perpendicular rays selected. c) Segmented M-mode image at the far and near wall boundaries prescribed by the upper and lower white curves produced for the fourth ray in Fig. 1b. For each of the four rays an M-Mode image was produced. The M-mode image shows the dynamic variations of the near and far wall boundaries. d) Four different diameter change curves were produced from the subtraction of the near and far wall boundaries from the corresponding images produced in c). The average of the four was computed and displayed in Fig. 1d). The diagram shows cardiac systolic and cardiac diastolic video frames from 0–2000 (100 frames per second=20 seconds). The contraction and distension frames are given in the first 10 sec. Contraction frames: 72–113, 174–208, 242–304, 367–393, 461–487, 515–568, 639–659, 706–749, 793–834, 892–929. Distension frames: 114–173, 209–241, 305–366, 394–460, 488–514, 569–638, 660–705, 750–792, 835–891, 930–980. Maximum carotid diameter 9.18 mm at frame 173, Minimum carotid diameter: 7.46 mm at frame 639.

FIGURE 2.

Texture feature variability through the cardiac cycle with systolic (s) and diastolic states (d) for the texture features GSM (SF), periodicity (SFM) and entropy (SGLDMr) for a) an normal (N) (male, age 38), b) an asymptomatic (A) (male, age 61) and c) a symptomatic (S) subject (male, age 63). The features for the N subject were extracted form he IMC and for the A and S subjects were extracted from the segmented plaque area (ROIs) as shown in Fig. 1a and Fig. 1b.

FIGURE 3.

Box plots for the texture feature GSM-SF for the normal (N), asymptomatic (A) and symptomatic (S) videos for cardiac systolic (s) and cardiac diastolic (d) states extracted from the corresponding ROIs investigated in this study. Inter-quartile (±IQR) values are shown above the box plots. Straight lines connect the nearest observation with 1.5 of the IQR of the lower and upper quartiles.