Design of optimal wire mesh collimator for gamma camera in mapping of cancer images

One of the factors leading to a high mortality rate fo r breast cancer patient is the late detection of the cancer cells. An early detection of breast cancer is vital as it increases the rate of survival among patients with breast cancer. Thus, that has become the main motivation of th is project. Several types of modalities have been identified fo r use in breast cancer detection such as scintimammography, mammography, ultrasound and magnetic resonance imaging. Scintimammography is a functional type of imaging that is capable of reducing false detection per cancer ratio in other types of imaging. Gamma camera is one of the nuclear imaging devices used for scintimammography. One of the important structures in gamma camera is a collimator. The aim of this study was to propose an optimal configuration of a wire mesh collimator as a replacement for the conventional muhihole collimator, specifically for breast cancer detection. Monte Carlo N-Particle code version 5 is used to simulate a semi-compressed breast phantom imaging with gamma camera where the energy of the radio-tracer used is 140 keY and the acceptance window of the detector is between 126 and 154 keY. Our evaluation was based on several parameters that give impact on the output image which indicates by the sensitivity, tumor contrast, spatial resolution and signal-to-noise ratio. These parameters were compared with a conventional muhihole collimator, which is our benchmark. The results were further processed by using an optimal parameter selection of Metz filter. At the end of this study, we managed to propose an optimal configuration of a wire mesh collimator for breast cancer imaging. Its percentage of weight was reduced to 44.09% from the weight of multihole collimator. At the same time, the sensitivity had also increased, while the tumor contrast, spatial resolution and signal-lo-noise ratio were preserved.

Text FK 2012 111R(UPM).pdf Download (952kB)

Actions (login required)

View Item