Computed tomography carotid wall plaque characterization using a combination of discrete wavelet transform and...

Area of Science:

• Medical Imaging
• Computer-Aided Diagnosis
• Cardiovascular Disease

Background:

• Carotid artery stenosis due to plaque causes 30% of strokes.
• Accurate classification of plaque (symptomatic vs. asymptomatic) is crucial for identifying high-risk patients.
• Non-invasive methods are needed for early detection and risk stratification.

Purpose of the Study:

• To develop and evaluate a non-invasive computer-aided diagnostic technique for classifying carotid artery plaque.
• To differentiate between symptomatic and asymptomatic plaque using imaging features.
• To improve stroke risk prediction and guide treatment decisions.

Main Methods:

• Utilized Computed Tomography (CT) images of the carotid artery.
• Extracted Local Binary Pattern (LBP) and wavelet energy features.
• Employed supervised learning algorithms, specifically Support Vector Machine (SVM) with RBF kernel, for classification.

Main Results:

• The SVM classifier achieved 88% accuracy, 90.2% sensitivity, and 86.5% specificity.
• Identified significant features for classification and proposed an 'Atheromatic Index' for objective prediction.
• Demonstrated the potential of CT imaging for calculating stenosis percentage and plaque classification.

Conclusions:

• The proposed computer-aided technique accurately classifies carotid artery plaque, aiding in stroke risk assessment.
• The method enables objective and faster prediction of plaque type, supporting clinical decision-making.
• CT imaging combined with feature analysis offers a valuable tool for managing patients with carotid artery stenosis.