Low-dose x-ray phase-contrast and absorption CT using equally sloped tomography

Area of Science:

• Medical Imaging
• Computational Imaging
• Applied Mathematics

Background:

• Undersampled and noisy projections challenge accurate tomographic reconstruction in transmission CT.
• Conventional methods like filtered back projection struggle with low-dose and fast acquisition requirements.
• Equally Sloped Tomography (EST) shows promise for improved accuracy with undersampled data.

Purpose of the Study:

• To develop novel reconstruction algorithms for equally sloped tomography (EST) incorporating advanced regularization.
• To integrate mathematical constraints, such as nonlocal means total variational models, into EST.
• To evaluate the performance of the developed algorithm in X-ray absorption and phase-contrast CT using simulations and experiments.

Main Methods:

• Development of new reconstruction algorithms for EST with advanced regularization constraints.
• Incorporation of nonlocal means total variational models consistent with experimental projections.
• Evaluation using simulations and experimental data from the European Synchrotron Radiation Facility on phantoms.

Main Results:

• The developed algorithm can reduce the number of projections by 60-75% in parallel beam modalities.
• Achieved comparable or superior image quality to conventional reconstructions.
• Demonstrated effectiveness in both X-ray absorption and phase-contrast CT modalities.

Conclusions:

• The enhanced EST method offers accurate tomographic reconstruction from significantly undersampled and noisy projection data.
• This approach facilitates low-dose and fast X-ray CT imaging, crucial for biology and medicine.
• The technique holds significant potential for applications utilizing advanced synchrotron radiation sources.