Spatially Variant Positron Range Modeling Derived from CT for PET Image Reconstruction

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Physics

Background:

• Positron range limits spatial resolution in Positron Emission Tomography (PET) scanners.
• This effect is significant for higher energy isotopes (e.g., Nitrogen-13, Rubidium-82) and at tissue boundaries.
• Current methods may not fully account for this spatial variation.

Purpose of the Study:

• To develop and validate a method for correcting positron range effects in PET imaging.
• To model the positron range effect as shift-variant, anisotropic kernels derived from CT scans.
• To improve the spatial resolution of PET images, particularly in challenging imaging scenarios.

Main Methods:

• Utilized CT scans to derive a positron range effect map.
• Modeled the effect using shift-variant, anisotropic kernels.
• Applied a positron range compensation map within a modified Ordered Subset Expectation Maximization (OSEM) algorithm.

Main Results:

• Successfully modeled the positron range effect across tissue boundaries.
• Demonstrated the application of the compensation map in a modified OSEM algorithm.
• Validated the method using both simulated and measured PET data.

Conclusions:

• The proposed method effectively compensates for positron range effects in PET imaging.
• This approach enhances spatial resolution, especially for higher energy isotopes and at tissue interfaces.
• The technique offers a practical solution for improving PET image quality in clinical settings.