Analytical Reconstruction of Human-Scale Dark-Field CT.

Area of Science:

• Medical Imaging
• Physics
• Biomedical Engineering

Background:

• Grating-based X-ray dark-field imaging enhances sensitivity to lung structures like alveoli.
• Human-scale dark-field CT shows clinical potential for lung disease diagnosis.
• Positional dependence of dark-field signals during CT scans causes reconstruction inaccuracies.

Purpose of the Study:

• To develop an accurate dark-field CT reconstruction method.
• To address the challenge of rotational variance in dark-field CT.
• To eliminate artefacts caused by positional signal dependence.

Main Methods:

• Modeling dark-field CT as a weighted Radon transform.
• Applying an analytical inversion formula to the weighted Radon transform model.
• Validating the method using simulations and experiments with an anthropomorphic chest phantom.

Main Results:

• Achieved artifact-free dark-field CT reconstruction.
• Successfully eliminated positional dependence of the dark-field signal.
• Demonstrated the method's applicability to cone-beam geometry.

Conclusions:

• The weighted Radon transform approach provides accurate dark-field CT reconstruction.
• This method overcomes limitations of existing calibration techniques.
• The approach is extendable and validated for clinical lung imaging applications.