Atherosclerotic carotid plaque segmentation

Area of Science:

• Medical Imaging
• Cardiovascular Disease Research
• Biomedical Engineering

Background:

• Atherosclerosis, a leading cause of heart attack and stroke, necessitates accurate plaque characterization.
• Ultrasound imaging is crucial for visualizing carotid artery plaque, but manual segmentation can be time-consuming and subjective.

Purpose of the Study:

• To develop and validate a computerized method for automated segmentation of atherosclerotic carotid plaque from ultrasound images.
• To assess the accuracy and reproducibility of the automated method compared to expert manual delineations.

Main Methods:

• Utilized blood flow imaging to establish an initial plaque contour.
• Applied despeckle filtering and active contour models (snakes) for precise boundary fitting.
• Validated the method on 35 longitudinal ultrasound images of carotid arteries.

Main Results:

• The computerized method achieved high accuracy, with a true positive fraction (TPF) of 86.44% and a true negative fraction (TNF) of 84.03%.
• False positive fraction (FPF) was 7% and false negative fraction (FNF) was 8.5%, indicating robust performance.
• The automated segmentation required minimal to no manual correction in most cases, demonstrating excellent reproducibility.

Conclusions:

• The developed computerized method provides a reliable and efficient tool for atherosclerotic carotid plaque segmentation.
• This automated approach has the potential to enhance the diagnosis and monitoring of cardiovascular diseases.
• Further integration into clinical workflows could improve efficiency and consistency in plaque analysis.