Two-Steps Coronary Artery Segmentation Algorithm Based on Improved Level Set Model in Combination with Weighted...

Area of Science:

• Medical Imaging
• Image Processing
• Cardiovascular Technology

Background:

• Coronary artery segmentation from angiography images is challenging due to complex structures, blurred images, and low contrast.
• Existing segmentation methods struggle with the uneven distribution of contrast agents, leading to difficulties in accurate vessel extraction.

Purpose of the Study:

• To propose a novel two-step segmentation algorithm for precise coronary artery image analysis.
• To overcome limitations of current methods in segmenting blurred and low-contrast coronary angiography images.

Main Methods:

• A two-step approach combining Hessian matrix analysis and level set evolution.
• Initial extraction of potential blood vessels using Hessian matrix eigenvalues and geometric features.
• Incorporation of novel regularization and area constraints within a local data energy fitting functional for level set evolution.

Main Results:

• The proposed algorithm successfully segments coronary artery images with improved precision.
• Experimental results demonstrate superior performance compared to existing segmentation algorithms.
• The method effectively handles image degradation issues like blurriness and low contrast.

Conclusions:

• The developed two-step algorithm offers a robust solution for coronary artery segmentation.
• This approach enhances the accuracy and reliability of analyzing coronary angiography images.
• The findings contribute to advancements in cardiovascular imaging analysis and diagnostics.