Coronary artery segmentation in X-ray angiograms using gabor filters and differential evolution

Area of Science:

• Medical Imaging
• Computer-Aided Diagnosis
• Biomedical Engineering

Background:

• X-ray angiograms are crucial for diagnosing coronary artery disease.
• Challenges include non-uniform illumination and low contrast, hindering accurate vessel segmentation.
• Existing enhancement methods have limitations in performance and computational cost.

Purpose of the Study:

• To develop a novel method for enhancing coronary blood vessels in X-ray angiograms.
• To optimize Gabor filter parameters for improved vessel detection and reduced computational expense.
• To enhance the accuracy of coronary artery segmentation for computer-aided diagnosis.

Main Methods:

• Proposed a novel blood vessel enhancement technique using Gabor filters.
• Employed Differential Evolution (DE) optimization strategy to tune Gabor filter parameters.
• Utilized the area under the receiver operating characteristics curve (Az) as the objective function for optimization.

Main Results:

• Achieved an Az of 0.9388 on a training set (40 images) and 0.9538 on a test set (40 images).
• Outperformed six state-of-the-art vessel detection methods in experimental results.
• Attained a vessel segmentation accuracy of 0.9423 on the test set.

Conclusions:

• The proposed Gabor filter-based method with DE optimization significantly enhances coronary artery segmentation.
• The method demonstrates superior performance and accuracy compared to existing techniques.
• It is highly suitable for clinical decision support due to its efficiency and accuracy.