Suppressing multi-material and streak artifacts with an accelerated 3D iterative image reconstruction algorithm for...

Area of Science:

• Medical Imaging
• Computational Imaging
• Physics

Background:

• In-line X-ray phase-contrast computed tomography (CT) involves separate phase retrieval and CT reconstruction steps.
• Multi-material samples and streak artifacts pose significant challenges in conventional CT reconstruction.
• Existing methods often struggle to address these issues simultaneously and efficiently.

Purpose of the Study:

• To develop an accelerated 3D iterative algorithm for simultaneous phase retrieval and CT reconstruction.
• To overcome limitations associated with multi-material samples and streak artifacts in X-ray CT.
• To improve the accuracy and efficiency of phase-contrast CT imaging.

Main Methods:

• A novel accelerated 3D iterative image reconstruction algorithm is proposed.
• The algorithm merges phase retrieval and CT reconstruction into a single, unified step.
• It incorporates a data fidelity term in the raw projection domain and a 3D total variation regularization term in the image domain.
• A transport-of-intensity equation (TIE)-based phase retrieval method is used, updated alternately for different sample areas.

Main Results:

• The proposed algorithm effectively addresses both multi-material and streak artifacts.
• Simultaneous phase retrieval and reconstruction are achieved efficiently.
• Validation through simulation and experimental results confirms the algorithm's effectiveness.
• The integrated approach demonstrates superior performance compared to traditional methods.

Conclusions:

• The accelerated 3D iterative algorithm provides a robust solution for phase-contrast CT.
• It successfully tackles key artifacts in multi-material imaging.
• The unified approach enhances both the quality and efficiency of CT reconstruction.