Identification of Ultrasonic Echolucent Carotid Plaques Using Discrete Fréchet Distance Between Bimodal Gamma...

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Cardiovascular Research

Background:

• Echolucent carotid plaques are linked to acute cardiovascular and cerebrovascular events (ACCEs).
• Accurate identification of plaque echogenicity is crucial for risk stratification.
• Current noninvasive methods may require enhancement for precise plaque characterization.

Purpose of the Study:

• To develop and evaluate a computer-aided method for identifying echolucent carotid plaques.
• To assess the feasibility of using discrete Fréchet distance features (DFDFs) for plaque classification.
• To enhance the noninvasive ultrasonic assessment of ACCE risk in patients with atherosclerosis.

Main Methods:

• A dataset of 315 carotid plaque ultrasound images (echolucent, echo-rich, intermediate) was analyzed.
• A bimodal gamma distribution modeled pixel statistics, and DFDFs were extracted.
• Linear discriminant analysis identified most discriminative features (MDFs) for k-nearest-neighbor classification.

Main Results:

• The developed method achieved a 77.46% classification accuracy for the three plaque types.
• The area under the receiver operating characteristics curve for identifying echolucent plaques was 0.831.
• The results demonstrate the potential of DFDFs in differentiating plaque echogenicity.

Conclusions:

• The computer-aided method based on DFDFs shows feasibility for identifying echolucent carotid plaques.
• This technique may improve the noninvasive ultrasonic prediction of ACCEs in patients with atherosclerotic plaques.
• Further validation could integrate this method into routine clinical practice for enhanced cardiovascular risk assessment.