PSF reconstruction for Compton-based prompt gamma imaging.

Area of Science:

• Medical Physics
• Nuclear Imaging
• Radiation Oncology

Background:

• Compton-based prompt gamma (PG) imaging is crucial for in vivo range verification in proton therapy.
• Image quality in PG imaging is degraded by Compton camera limitations like spatial and energy resolution, and spatially variant resolution loss.

Purpose of the Study:

• To investigate the performance of the list-mode ordered subset expectation maximization algorithm with a shift-variant point spread function (LM-OSEM-SV-PSF) model.
• To evaluate the ability of SV-PSF modeled PG images to accurately reproduce the distal falloff of the proton beam.

Main Methods:

• The study utilized simulation data of point sources to estimate SV-PSF parameters.
• Simulated PGs were generated in a water phantom irradiated with a proton beam.
• The LM-OSEM-SV-PSF algorithm was compared against the standard LM-OSEM algorithm.

Main Results:

• The LM-OSEM-SV-PSF algorithm significantly improved PG image quality and distal falloff estimation compared to LM-OSEM.
• Accurate reconstruction of 4.44 and 5.25 MeV PG emissions was achieved with LM-OSEM-SV-PSF.
• Limited improvement was observed for 2.31 and 6.13 MeV PG emissions.

Conclusions:

• The LM-OSEM-SV-PSF algorithm enhances PG imaging for proton therapy range verification.
• While effective for certain energies, alternative methods like LM-OSEM with Richardson-Lucy deconvolution offer comparable visual quality with reduced computation time.