Study of OSEM with different subsets in grating-based X-ray differential phase-contrast imaging

Area of Science:

• Medical Imaging
• Physics
• Computational Science

Background:

• X-ray phase contrast computed tomography (XPCT) offers higher contrast than absorption-based imaging.
• Grating interferometry is a key technique for XPCT.
• A novel "single-step" method simplifies data acquisition and reduces sample dose compared to phase-stepping approaches.

Purpose of the Study:

• To evaluate the performance and applicability of the Ordered Subsets Expectation Maximization (OSEM) algorithm for single-step XPCT image reconstruction.
• To compare OSEM's image quality and convergence speed with varying subset numbers.
• To assess OSEM's suitability for clinical XPCT applications.

Main Methods:

• Application of the OSEM iterative image reconstruction algorithm to experimental data from single-step XPCT.
• Characterization of OSEM performance using computer simulations and experimental results.
• Comparison of OSEM with the traditional filtered back projection (FBP) algorithm, particularly under noisy conditions.

Main Results:

• OSEM reliably reconstructs images for single-step XPCT.
• OSEM demonstrates superior image quality, higher spatial resolution, and lower noise compared to FBP, especially in noisy acquisitions.
• The algorithm's convergence speed and image quality were characterized based on the number of subsets used.

Conclusions:

• The OSEM algorithm is a reliable and effective method for reconstructing images in single-step X-ray phase contrast computed tomography.
• OSEM provides improved image quality over traditional methods, making it highly compatible with future clinical X-ray phase contrast imaging.
• This advancement holds significant promise for enhancing diagnostic capabilities in medical imaging.