Robust Sample Information Retrieval in Dark-Field Computed Tomography With a Vibrating Talbot-Lau Interferometer.

Area of Science:

• Medical Imaging
• Physics
• Biomedical Engineering

Background:

• X-ray computed tomography (CT) provides essential 3D diagnostic information.
• Grating-based dark-field X-ray imaging offers microstructural insights beyond conventional CT.
• Clinical application of dark-field CT is limited by interferometer sensitivity to vibrations.

Purpose of the Study:

• To develop a processing pipeline for clinical dark-field CT prototypes.
• To mitigate vibration artifacts in human-sized dark-field CT imaging.
• To enable large field-of-view dark-field CT for clinical use.

Main Methods:

• A novel processing and reconstruction pipeline was developed for a clinical dark-field CT prototype.
• Spatially and temporally variable vibrations were identified using air reference scans.
• Correlations from reference scans were applied to sample scans to mitigate vibration effects.
• The method allows for the separation of fluctuation and sample information without requiring sample-free detector areas.

Main Results:

• The processing pipeline effectively mitigates vibration artifacts in dark-field CT.
• Reconstructions of human-scale objects, including an anthropomorphic thorax phantom, were successfully achieved.
• The technique allows for the reconstruction of samples of arbitrary dimensions without artifacts.
• The method eliminates the need for sample-free detector areas, simplifying the process.

Conclusions:

• The proposed processing pipeline enables robust dark-field CT imaging for human-scale objects.
• This advancement overcomes previous limitations in clinical dark-field CT applications.
• The technique holds significant potential for improving medical diagnosis through enhanced microstructural imaging.