Software correction of scatter coincidence in positron CT

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Computational Physics

Background:

• Scatter coincidence is a significant source of error in positron emission tomography (PET) quantitation.
• Accurate scatter correction is crucial for reliable quantitative analysis of PET images.

Purpose of the Study:

• To develop and validate a software-based method for correcting scatter coincidence in PET.
• To improve the quantitative accuracy of PET imaging by mitigating scatter effects.

Main Methods:

• A novel integral transform method was employed to estimate scatter projections from true projections.
• Kernels for the integral transform were derived using Klein-Nishina's formula for various line source positions.
• The method was validated through computer simulations and phantom experiments using Positologica software.

Main Results:

• Scatter coincidence effects were found to be substantial and non-negligible in PET image quantitation.
• The proposed software correction method significantly reduced the impact of scatter coincidence.
• Quantitative accuracy of PET images was demonstrably improved post-correction.

Conclusions:

• The developed software correction effectively addresses scatter coincidence in PET.
• This method offers a viable solution for enhancing the accuracy of quantitative PET imaging.
• The findings underscore the importance of scatter correction for reliable PET data interpretation.