An automated technique for carotid far wall classification using grayscale features and wall thickness variability

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Cardiovascular Diagnostics

Background:

• Carotid artery atherosclerosis is a major risk factor for stroke.
• Accurate differentiation of symptomatic and asymptomatic carotid artery disease is crucial for risk stratification.
• Current diagnostic methods may require further refinement for optimal patient management.

Purpose of the Study:

• To evaluate a computer-aided diagnostic (CAD) system for classifying carotid B-mode ultrasonographic images.
• To differentiate between symptomatic and asymptomatic carotid artery plaque.
• To assess the feasibility and accuracy of automated image analysis for cardiovascular risk assessment.

Main Methods:

• Development of the Atheromatic system for automated intima-media thickness (IMT) and intima-media thickness variability (IMTVpoly) calculation.
• Extraction of nonlinear features using higher-order spectral analysis.
• Application of a multiclassifier system for symptomatic/asymptomatic (Sym/Asym) labeling of carotid artery images.

Main Results:

• A support vector machine classifier achieved 99.1% accuracy in differentiating Sym/Asym.
• Key discriminating features included IMT, IMTVpoly, and bispectral entropies at specific angles.
• The system effectively utilized both geometric and spectral image characteristics.

Conclusions:

• Computer-aided classification of carotid artery images into symptomatic and asymptomatic categories is feasible and highly accurate.
• This method holds potential for early atherosclerosis detection.
• The system can assist in identifying patients at higher cardiovascular risk, enabling timely intervention.