Automatic 2-D/3-D Vessel Enhancement in Multiple Modality Images Using a Weighted Symmetry Filter

Area of Science:

• Medical Imaging
• Image Processing
• Vascular Biology

Background:

• Automated detection of vascular structures is crucial for diagnosing and treating vascular pathologies.
• Challenges include poor contrast, varied backgrounds, anatomical differences, and image noise.
• Existing methods struggle with these complex imaging conditions.

Purpose of the Study:

• To propose a novel 2-D/3-D symmetry filter for enhancing vascular structures.
• To address limitations of current automated vascular detection techniques.
• To improve the accuracy and robustness of vessel enhancement across different imaging modalities.

Main Methods:

• Developed a novel 2-D/3-D symmetry filter incorporating local phase features via a quadrature filter.
• Utilized a weighted geometric mean of blurred and shifted quadrature filter responses.
• Designed for enhanced tolerance to irregular vessel appearances and imaging artifacts.

Main Results:

• The filter demonstrated a strong response to vascular features under typical imaging conditions.
• Evaluated on six 2-D, one 3-D, and one 3-D synthetic dataset.
• Showcased superior performance compared to existing state-of-the-art methods.

Conclusions:

• The proposed 2-D/3-D symmetry filter effectively enhances vascular structures in medical images.
• This method offers a robust solution for automated vascular detection despite imaging challenges.
• The filter shows significant potential for improving the diagnosis and understanding of vascular diseases.