PatchDSA: improving digital subtraction angiography with patch-based phase-matching in natural breathing scenarios.

Area of Science:

• Medical Imaging
• Radiology
• Image Processing

Background:

• Digital subtraction angiography (DSA) visualizes blood vessels by subtracting mask images from contrast images.
• Misalignment between mask and contrast images causes motion artifacts, obscuring structures and leading to misidentification.

Purpose of the Study:

• To develop a novel method for generating high-quality abdominal DSA images by addressing motion artifacts.
• To improve the accuracy and clarity of DSA imaging under natural breathing conditions.

Main Methods:

• Proposed a patch-based phase-matching technique for aligning mask and contrast images.
• Divided images into patches, selected structurally similar mask patches, and refined alignment at a subpixel level.
• Evaluated the method on 20 abdominal angiogram cases, comparing it to an existing technique.

Main Results:

• The proposed method effectively reduced motion artifacts in abdominal DSA images.
• Outperformed the existing phase matching-based method in all evaluated cases.
• Successfully identified optimal mask images on a patch-by-patch basis, suppressing physiological motion artifacts.

Conclusions:

• The novel patch-based phase-matching technique generates higher-quality abdominal DSA images.
• This method significantly suppresses artifacts from peristalsis and cardiac pulsation.
• Enhances diagnostic accuracy by improving the visualization of vascular structures and organ movements.