FBP initialized few-view CT reconstruction algorithm using similar prior image constraint.

Area of Science:

• Medical Imaging
• Computational Imaging
• Radiology

Background:

• Few-view CT reconstruction is crucial for reducing radiation dose.
• Filtered back projection (FBP) in few-view CT often results in severe artifacts.
• Utilizing prior information can enhance image quality, but structural differences pose challenges.

Purpose of the Study:

• To develop a novel technique for few-view CT reconstruction that minimizes artifacts and preserves image details.
• To leverage similar prior images effectively despite potential structural variations.
• To improve the accuracy and robustness of low-dose CT reconstruction.

Main Methods:

• Proposed an FBP initialized similar prior image constrained technique (FISPICT).
• FISPICT utilizes FBP distribution information from few projection views.
• Guides the search for relevant distribution regions within prior images to approximate the current scan.

Main Results:

• FISPICT effectively suppresses structure differences and streak artifacts.
• Simulation experiments demonstrate superior performance and detail preservation compared to existing methods.
• The algorithm exhibits high robustness even when the prior image differs significantly from the current scan.

Conclusions:

• FISPICT offers a promising approach for high-quality few-view CT reconstruction with reduced radiation dose.
• The method successfully integrates prior image information while mitigating distortion risks.
• This technique enhances the reliability and applicability of low-dose CT imaging.