Regularized iterative integration combined with non-linear diffusion filtering for phase-contrast x-ray computed...

Area of Science:

• Medical Imaging
• X-ray Computed Tomography
• Image Processing

Background:

• Phase-contrast X-ray computed tomography offers advantages over conventional absorption-contrast imaging.
• Clinical implementation is hindered by noise and resolution limitations.

Purpose of the Study:

• To investigate noise-reducing, resolution-preserving analytical reconstruction methods for differential phase-contrast imaging.
• To improve the quality of phase-contrast imaging for potential clinical use.

Main Methods:

• Applied the non-linear Perona-Malik filter to phase-contrast data before or after filtered backprojection.
• Replaced the Hilbert kernel with regularized iterative integration followed by ramp filtered backprojection.

Main Results:

• Successfully revealed relevant sample features by combining the Perona-Malik filter with the integration algorithm.
• Quantitatively confirmed improvements through significantly increased structural similarity indices and contrast-to-noise ratios.
• Demonstrated that phase-contrast imaging can be performed at considerably lower doses.

Conclusions:

• The developed methods effectively reduce noise while preserving resolution in differential phase-contrast imaging.
• This approach enhances image quality and allows for dose reduction, facilitating clinical implementation.
• Improved phase-contrast imaging holds promise for advancing medical diagnostics.