Fast iterative reconstruction for helical pinhole SPECT imaging.

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Nuclear Medicine

Background:

• Pinhole SPECT is crucial for small animal imaging in molecular biology and pharmaceutical development.
• Whole-body imaging of lab animals is in high demand.
• Helical trajectories reduce acquisition time and avoid gaps but increase computational complexity for iterative reconstruction.

Purpose of the Study:

• To develop a fast iterative image reconstruction method for helical pinhole SPECT.
• To address the computational challenges posed by helical trajectories in SPECT imaging.
• To improve the efficiency of whole-body small animal imaging.

Main Methods:

• A novel voxel-driven (VD) system model was proposed.
• The VD model integrates geometric symmetries of the helical trajectory.
• The model facilitates faster system matrix calculation and reduces storage requirements.
• Parallel coding was utilized to enhance forward/backward projection efficiency.

Main Results:

• The VD model adequately models helical pinhole SPECT scanners.
• Manageable storage size for the system matrix was achieved.
• Clinically acceptable computation loading for image reconstruction was demonstrated.
• Phantom studies validated the model's effectiveness.

Conclusions:

• The proposed VD system model enables fast and efficient iterative reconstruction for helical pinhole SPECT.
• This approach effectively manages computational complexity and storage demands.
• It supports routine whole-body imaging of small animals, advancing preclinical research.