Coronary stent imaging with dual-source CT: assessment of lumen visibility using different convolution kernels and...

Area of Science:

• Medical Imaging
• Radiology
• Cardiovascular Imaging

Background:

• Coronary computed tomography angiography (CCTA) requires specific image reconstructions for assessing coronary artery stents and atherosclerotic plaques.
• Current methods necessitate separate reconstructions using soft kernels for plaque analysis and edge-enhancing kernels for stent lumen evaluation.

Purpose of the Study:

• To evaluate a novel 2D filter tool designed for CCTA image postprocessing.
• The tool aims to convert soft kernel images to edge-enhanced images and vice versa, potentially eliminating the need for dual reconstructions.

Main Methods:

• Dual-source CT scans of vascular phantoms with 20 stents (3.0 mm diameter) were performed.
• Images were reconstructed with soft (B30f) and edge-enhancing (B46f) kernels and postprocessed with corresponding filters (F30, F46).
• Lumen visibility, intraluminal attenuation, and image noise were assessed by two readers and validated in vivo against invasive coronary angiography.

Main Results:

• The 2D filter (F30) reduced intraluminal attenuation and image noise compared to the soft kernel (B30f).
• Edge-enhancing kernels (B46f, F46) significantly improved visible stent diameter compared to soft kernels (B30f, F30).
• In vivo, B46f images showed superior lumen assessability and the best sensitivity (67%) and specificity (94%) for stenosis detection.

Conclusions:

• The evaluated 2D postprocessing filter effectively reduces image noise.
• However, the filter does not currently provide an alternative to dedicated edge-enhancing kernels for accurate stent lumen evaluation in CCTA.
• Further research and development are required to optimize this postprocessing technique for clinical application in stent assessment.