Simulation studies on depth of interaction effect correction using a Monte Carlo computed system matrix for brain...

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Image Reconstruction

Background:

• Parallax errors in Positron Emission Tomography (PET) degrade image quality due to limited depth-of-interaction (DOI) information.
• Multi-layered detectors offer DOI information but increase system complexity and processing demands.

Purpose of the Study:

• To evaluate an inherited DOI effect correction method for single-layered PET detectors.
• To compare reconstruction quality against multi-layered detector systems.

Main Methods:

• Simulation studies using Monte Carlo computed system matrices for single-layered detectors.
• Comparison with Filtered Backprojection (FBP) reconstructions from multi-layered detector data.
• Investigation of multi-layered detector benefits within Maximum Likelihood Expectation Maximization (MLEM) iterations.

Main Results:

• Single-layered detectors with inherited DOI correction in MLEM outperformed FBP reconstructions from octuple-layered detectors.
• While multi-layered detectors improved MLEM results, the gains were not substantial enough to justify their complexity and cost.
• Inherited DOI correction for single-layered detectors proved sufficient for brain PET.

Conclusions:

• Detailed DOI information from multi-layered detectors is beneficial but not essential for brain PET.
• Single-layered detectors with Monte Carlo-based DOI effect correction provide adequate performance.
• This approach offers a practical alternative to complex multi-layered detector systems in brain PET.