Histogram-based selective deblurring to improve computed tomography imaging of calcifications.

Area of Science:

• Medical Imaging
• Image Processing
• Radiology

Background:

• Computed tomography (CT) imaging struggles with image blur when visualizing small, high-density structures like calcifications.
• This blur complicates accurate assessment and quantification of these structures.

Purpose of the Study:

• To develop and evaluate a deblurring method for calcifications in CT images.
• To improve visualization and quantification of calcifications while minimizing artifacts.

Main Methods:

• A novel histogram-based selective deblurring (HiSD) method was developed.
• HiSD combines original low-intensity image data with deconvolved high-intensity data.
• The method was validated using phantom and in vitro atherosclerotic plaque CT images against micro-CT references.

Main Results:

• HiSD effectively reduced calcification blur in both transverse and longitudinal directions.
• The method avoided noise amplification and edge-ringing artifacts in surrounding tissues.
• Restored images showed significantly improved calcification area and volume measurements, reducing overestimation by 32% and 83% respectively.

Conclusions:

• HiSD significantly enhances the visualization of small, high-density structures in CT imaging.
• The technique improves the quantitative accuracy of calcification measurements in vitro.
• HiSD offers a promising solution for overcoming limitations in CT imaging of calcifications.