Quantitative analysis of a reconstruction method for fully three-dimensional PET

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Image Reconstruction

Background:

• Positron emission tomography (PET) systems with large area planar detectors offer a larger axial field-of-view (FOV) compared to ring systems.
• Traditional methods using a polar angle rejection criterion during backprojection limit FOV size and sensitivity trade-offs for space-invariant point spread functions.

Purpose of the Study:

• To introduce and evaluate a novel algorithm for list-mode PET data that overcomes the FOV-sensitivity trade-off.
• To compare the performance of the new subregion-based algorithm with the conventional backprojection-then-filter method.

Main Methods:

• A new algorithm by Defrise and co-workers, dividing the image into subregions, was applied to list-mode PET data.
• Simulated and real data from the MUP-PET camera were used for comparison.
• Signal-to-noise ratio (SNR) analysis was performed.

Main Results:

• The new subregion algorithm demonstrated potential improvements in signal-to-noise ratio by up to a factor of 1.4.
• Increased data usage, up to a factor of 2.5, was observed depending on the object's axial extent.
• Enhanced quantitation accuracy was also achieved.

Conclusions:

• The novel subregion-based algorithm effectively addresses the limitations of traditional PET image reconstruction, particularly concerning FOV and sensitivity.
• This method offers significant improvements in image quality and data utilization for PET imaging.
• The algorithm shows promise for advancing PET scanner performance and quantitative accuracy.