Speaker
Mr
Andrew Mather
(University of Liverpool)
Description
The SmartPET project will examine the possibility of using planer
Hyper Pure Germanium Detectors (HPGe) within a Positron Emission
Tomography (PET) system. This system is designed to image small
animals.
The image reconstruction algorithms take measured or simulated
projection data of an underlying radioactive source distribution and
attempt to produce an accurate 2D or 3D image of this distribution.
Two main groups of algorithm exist, analytical methods such as
Filtered Back Projection (FBP) and statistical methods such as
Maximum Likelihood Expectation Maximization (ML-EM). Each algorithm
has advantages and disadvantages; often the trade-off is between the
quality of the final image and the computational power needed to
produce it.
This presentation will provide an overview of the image
reconstruction process from collection of the projection data
(sinogram) to the use of this data in forming an image using either
FBP or ML-EM. Example reconstructed images will be shown for both
simulated and preliminary experimental data from the SmartPET system.
Primary author
Mr
Andrew Mather
(University of Liverpool)
