A reconstruction algorithm for coherent scatter computed tomography based on filtered back-projection

Area of Science:

• Medical Imaging
• X-ray Physics
• Computational Imaging

Background:

• Coherent scatter computed tomography (CSCT) is an advanced imaging method.
• Current reconstruction methods, like algebraic reconstruction techniques (ART), can be slow.
• There is a need for faster and more flexible reconstruction algorithms in CSCT.

Purpose of the Study:

• To introduce a novel filtered back-projection (FBP) algorithm for CSCT.
• To evaluate the performance of the proposed 3D FBP algorithm using simulated and experimental data.
• To compare the proposed algorithm with existing ART methods in terms of speed and image quality.

Main Methods:

• Developed a 3D filtered back-projection algorithm utilizing curved back-projection lines.
• Applied the algorithm to 2D coherent scatter projection data.
• Tested the algorithm with simulated data and data from a multi-line CT scanner demonstrator.

Main Results:

• The modified 3D FBP algorithm achieved comparable image quality to ART.
• The new algorithm is approximately two orders of magnitude faster than ART.
• Enabled subfield-of-view reconstruction for selective imaging of the coherent-scatter form factor.

Conclusions:

• The modified 3D FBP algorithm is a powerful and efficient reconstruction technique for CSCT.
• This method significantly improves the speed and flexibility of CSCT systems.
• Coherent scatter CT has the potential to become a competitive imaging modality for medical and industrial applications.