Coded aperture optimization using Monte Carlo simulations

Coded apertures using Uniformly Redundant Arrays (URA) have been unsuccessfully evaluated for two-dimensional and three-dimensional imaging in Nuclear Medicine. The images reconstructed from coded projections contain artifacts and suffer from poor spatial resolution in the longitudinal direction. We...

Full description

Saved in:
Bibliographic Details
Published in:Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment Accelerators, spectrometers, detectors and associated equipment, 2010-04, Vol.616 (1), p.75-80
Main Authors: Martineau, A., Rocchisani, J.M., Moretti, J.L.
Format: Article
Language:English
Subjects:
Algorithms
Coded aperture
Computer simulation
GATE
Mathematical models
Monte Carlo
Monte Carlo methods
Projection
Reconstruction
Spatial resolution
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Coded apertures using Uniformly Redundant Arrays (URA) have been unsuccessfully evaluated for two-dimensional and three-dimensional imaging in Nuclear Medicine. The images reconstructed from coded projections contain artifacts and suffer from poor spatial resolution in the longitudinal direction. We introduce a Maximum-Likelihood Expectation–Maximization (MLEM) algorithm for three-dimensional coded aperture imaging which uses a projection matrix calculated by Monte Carlo simulations. The aim of the algorithm is to reduce artifacts and improve the three-dimensional spatial resolution in the reconstructed images. Firstly, we present the validation of GATE (Geant4 Application for Emission Tomography) for Monte Carlo simulations of a coded mask installed on a clinical gamma camera. The coded mask modelling was validated by comparison between experimental and simulated data in terms of energy spectra, sensitivity and spatial resolution. In the second part of the study, we use the validated model to calculate the projection matrix with Monte Carlo simulations. A three-dimensional thyroid phantom study was performed to compare the performance of the three-dimensional MLEM reconstruction with conventional correlation method. The results indicate that the artifacts are reduced and three-dimensional spatial resolution is improved with the Monte Carlo-based MLEM reconstruction.
ISSN:0168-9002
1872-9576
DOI:10.1016/j.nima.2010.02.261