Superimposed Wavefront Imaging of Diffraction-enhanced X-rays: sparsity-aware CT reconstruction from limited-view...

Area of Science:

• Medical Imaging
• Computational Imaging
• X-ray Tomography

Background:

• Differential phase-contrast CT offers high-contrast imaging of biological tissues.
• The Superimposed Wavefront Imaging of Diffraction-enhanced X-rays (SWIDeX) technique generates high-contrast phase-contrast images.
• Reconstructing SWIDeX data typically requires a large number of projections.

Purpose of the Study:

• To develop and evaluate an algebraic reconstruction algorithm with total variation regularization (ART+TV) for SWIDeX.
• To significantly reduce the number of projections required for differential phase-contrast CT reconstruction using SWIDeX.
• To maintain image quality and structural integrity with fewer projections.

Main Methods:

• The proposed method utilizes ART+TV to reconstruct a Laplacian phase-contrast image from limited SWIDeX projections.
• Poisson's equation is solved for the Laplacian image to obtain the final tomographic image representing refractive index distribution.
• The technique leverages the sparsity of the Laplacian image for effective reconstruction.

Main Results:

• Simulations demonstrated that image quality was preserved with approximately 1/10th of the original projections.
• Actual X-ray experiments showed that biological tissue structures were maintained with as few as 1/30th of the original projections.
• The ART+TV approach effectively reduces projection requirements for SWIDeX-based CT.

Conclusions:

• SWIDeX is a powerful technique for high-contrast visualization of biological tissue internal structures.
• The developed ART+TV algorithm is highly effective for CT reconstruction from limited SWIDeX projection data.
• This method facilitates CT reconstruction with wide fields of view and high spatial resolution, even with reduced projection counts.