Performance of the Fourier rebinning algorithm for PET with large acceptance angles

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Image Reconstruction

Background:

• The Fourier rebinning (FORE) technique is a novel method for 3D Positron Emission Tomography (PET) reconstruction.
• Evaluating reconstruction techniques across varying axial acceptance angles is crucial for optimizing image quality in 3D PET.

Purpose of the Study:

• To investigate the performance of the Fourier rebinning (FORE) technique in 3D PET reconstruction.
• To compare the FORE technique with the 3D-Region-of-Interest (3D-RP) technique for large axial acceptance angles.
• To assess spatial resolution, contrast, and noise characteristics of the FORE technique.

Main Methods:

• Spatial resolution, contrast, and noise figures of merit were used to evaluate FORE performance.
• Reconstruction performance was compared between FORE (with multislice 2D reconstruction) and 3D-RP for large acceptance angles (+/-26.25 degrees).

Main Results:

• FORE does not impact transverse resolution.
• Axial resolution with FORE deteriorates more rapidly than with 3D-RP at large radii and increasing acceptance angles.
• Filtering demonstrates superior noise suppression in FORE reconstructions compared to 3D-RP, particularly at the axial field of view edges.

Conclusions:

• The FORE technique is a practical option for 3D PET applications due to its good performance and efficient reconstruction time.
• FORE offers advantages in noise reduction, especially for peripheral regions within the axial field of view.
• While axial resolution may be a consideration at extreme parameters, FORE's overall efficiency and noise handling are beneficial.