Accurate helical cone-beam CT reconstruction with redundant data

Area of Science:

• Medical Imaging
• Computed Tomography
• Image Reconstruction

Background:

• Helical cone-beam computed tomography (CT) is crucial for medical imaging.
• Reconstruction algorithms face challenges with data acquired at maximum pitch.
• Efficiently utilizing all measured data remains an ongoing research area.

Purpose of the Study:

• To introduce a new, theoretically exact, and stable image reconstruction algorithm for helical cone-beam CT.
• To develop an algorithm optimized for data collected at or near maximum pitch.
• To enhance reconstruction efficiency by operating volume-wise.

Main Methods:

• The algorithm employs differentiated backprojection and finite-support Hilbert transform inversion.
• Operations are applied to M-lines, with data from three M-line sets averaged for comprehensive data utilization.
• The approach is volume-wise, differing from previous voxel-wise methods for improved efficiency.

Main Results:

• Preliminary results demonstrate numerical stability of the algorithm.
• The method effectively utilizes redundant data for enhanced reconstruction.
• The algorithm successfully processes data acquired with an angular flying focal spot.

Conclusions:

• The presented algorithm provides an exact and stable reconstruction for helical cone-beam CT.
• The volume-wise approach significantly improves reconstruction efficiency.
• The algorithm shows promise for handling challenging data acquisition scenarios, including high pitch and flying focal spot techniques.