Fluence adaptation for contrast-based dose optimization in x-ray phase-contrast imaging.

Area of Science:

• Medical Imaging Physics
• Radiological Sciences
• Computational Imaging

Background:

• X-ray phase-contrast imaging (XPCI) offers superior contrast for clinical and industrial applications, utilizing conventional X-ray sources.
• Grating-based imaging (GBI) and edge-illumination (EI) are key XPCI techniques, typically employing phase-stepping with constant X-ray fluence.
• Reducing patient radiation dose in XPCI is critical for widespread clinical adoption.

Purpose of the Study:

• To challenge the conventional constant fluence approach in XPCI phase-stepping acquisition.
• To propose and validate a novel fluence adaptation mechanism for dose reduction in XPCI.
• To achieve contrast-based dose optimization while maintaining or improving noise performance.

Main Methods:

• Derived an optimal adaptive fluence distribution based on analytic retrieval formulas and Gaussian noise models.
• The distribution is proportional to absolute weighting kernel functions and the root of original sample phase-stepping curves (PSCs).
• Proposed two practical adaptive fluence forms for GBI and EI systems, adaptable even without prior sample PSC knowledge.

Main Results:

• Simulations showed approximately 20% dose reduction for phase contrast using the proposed method compared to constant fluence.
• Experimental results confirmed better noise performance in low-dose scenarios, consistent with theoretical predictions.
• Identified effective parameter ranges for PSCs in GBI and EI, providing practical guidance for implementation.

Conclusions:

• A novel fluence adaptation mechanism for contrast-based dose optimization in XPCI (GBI and EI) has been successfully proposed.
• This method offers a new pathway for significant radiation dose reduction in XPCI.
• The approach has potential for extension to other XPCI systems and retrieval algorithms.