An accurate probabilistic model with detector resolution and Doppler broadening correction in list-mode MLEM...

Area of Science:

• Medical imaging physics
• Nuclear medicine instrumentation
• Radiation detection technology

Background:

• Compton cameras are vital for medical applications and radioactive material detection.
• High-resolution imaging is challenging for Compton cameras with incident photon energies below 200 keV.
• Many medically relevant radionuclides emit photons in this low-energy range.

Purpose of the Study:

• To develop an improved probabilistic model for Compton camera image reconstruction.
• To enhance resolution for low-energy photons (<200 keV) by correcting for detector energy/spatial resolution and Doppler broadening.
• To integrate the model into the list-mode maximum likelihood expectation maximization (LM-MLEM) algorithm.

Main Methods:

• Theoretical analysis and quantitative calculation of factors affecting Compton camera angular resolution.
• Development and derivation of a novel probabilistic model incorporating Doppler broadening.
• Verification through Monte Carlo simulations (201Tl, 99mTc) and experiments using a Timepix3 detector-based Compton camera.

Main Results:

• Analysis confirmed the necessity of including Doppler broadening for accurate angular resolution.
• The proposed probabilistic model demonstrated improved imaging resolution in simulations and experiments.
• Effectiveness was particularly noted for low-energy photons and small scattering-absorption distances.

Conclusions:

• The developed probabilistic model significantly enhances Compton imaging resolution below 200 keV.
• The model's inclusion of Doppler broadening is critical for accurate reconstruction in this energy range.
• Validated through simulations and experiments, the algorithm offers improved performance for specific medical imaging scenarios.