Image reconstruction from data over two orthogonal arcs of limited-angular ranges

Area of Science:

• Medical Imaging
• Computed Tomography
• Image Reconstruction

Background:

• Computed tomography (CT) scanning over limited-angular ranges (LARs) is practical for reducing imaging dose and time.
• Nonstandard CT scan designs are also of interest.

Purpose of the Study:

• Investigate image reconstruction for two nonoverlapping arcs of LARs.
• Demonstrate that two arcs may allow more accurate image reconstruction than a single arc.

Main Methods:

• Considered a two-orthogonal-arc configuration with nonoverlapping LARs.
• Generated data from a chest phantom using two-orthogonal-arc configurations.
• Reconstructed images using the directional-total-variation (DTV) algorithm.
• Compared results with a single-arc configuration and evaluated accuracy using nRMSE.

Main Results:

• The two-orthogonal-arc configuration generally yielded more accurate image reconstruction than the single-arc counterpart.
• Increasing total angular range enhanced reconstruction accuracy with the DTV algorithm.
• Specific two-orthogonal-arc configurations significantly outperformed others, showing nRMSE lower by over an order of magnitude.

Conclusions:

• Appropriately designed two-orthogonal-arc configurations can improve CT image reconstruction accuracy with reduced angular coverage.
• This approach offers insights for innovative CT scan designs to lower scan time and radiation dose.
• Potential applications include C-arm CT to avoid scan collisions.