High-resolution coded aperture optimization for super-resolved compressive x-ray cone-beam computed tomography.

Area of Science:

• Medical Imaging
• Computational Imaging
• X-ray Tomography

Background:

• Compressive X-ray cone-beam computed tomography (CBCT) uses coded apertures (CA) to improve image quality.
• Existing CA designs are often limited to fan-beam systems and computationally intensive optimization methods like singular value decomposition (SVD).
• Image resolution in current systems is constrained by detector pixel size.

Purpose of the Study:

• To develop a new method for designing CA patterns in compressive CBCT for super-resolution imaging.
• To achieve high-resolution images from low-resolution detectors in a single-shot CBCT system.
• To improve image quality and reduce computational complexity compared to existing CA optimization techniques.

Main Methods:

• A novel method for designing CA patterns in a compressive CBCT system utilizing a super-resolution configuration.
• Leveraging the Gershgorin theorem to minimize eigenvalue bounds and improve the condition of the system matrix.
• Simulations using medical datasets and Monte Carlo simulated projections.

Main Results:

• The proposed CA design successfully generates high-resolution images from lower-resolution detectors in a single-shot CBCT scenario.
• Image quality improved by up to 5 dB in peak signal-to-noise ratio compared to random CA patterns.
• Reconstructions from Monte Carlo simulations showed up to 3 dB improvement.
• The computational load was reduced by up to three orders of magnitude compared to SVD-based methods.

Conclusions:

• The developed CA design method enables high-resolution imaging in compressive CBCT with improved image quality and significantly reduced computational cost.
• This approach offers a practical and efficient solution for super-resolution imaging in CBCT systems.
• The findings suggest a promising direction for advancing medical imaging technologies through optimized coded aperture design.