Sub-voxel position identification in Cadmium Zinc Telluride detectors for Low Dose Molecular Breast Imaging

Lead Research Organisation: University of Liverpool
Department Name: Physics

Abstract

Breast cancer is the most common form of the disease in the UK, directly affecting 1 in 8
women. Cancer treatment has developed significantly over the last few decades such that there
are currently incredibly sophisticated techniques and technologies available with an ever
growing success rate for cancer sufferers. However, a major limiting factor to the success of
such treatments is the stage at which the cancer is diagnosed. The earlier the disease is
diagnosed, the greater the probability of treatments working to eradicate the disease. Hence,
developments of sophisticated diagnostic techniques are also required.
Approximately half of women of screening age have what are considered 'mammographically
dense breasts'. Although this is a normal and common characteristic, it limits the diagnostic
abilities of the screening methods available which are used to detect cancer. The abilities of all
common diagnostic techniques are limited when conducted on breasts that are considered to be
mammographically dense. Women with mammographically dense breasts have a higher risk of
developing breast cancer and a lower chance of having it detected. Hence, alternative methods
are required to improve the diagnostic performance when screening for breast cancer.
Molecular Breast Imaging is a branch of diagnostic Nuclear Medicine which utilises the
radioisotope Tech-99m to obtain functional images and identify lesions within the breast. The
radioisotope is injected into the patient where it moves to the breast region and is preferentially
up-taken by cancer cells. With a half-life of 6 hours, the isotope decays within the patient,
emitting a characteristic 141 keV gamma ray as it does so. The decay products leave the body,
interacting with the material in their path as they do so. A gamma camera situated outside the
patient detects the emitted gammas in order to process the events and reconstruct the image of
radionuclide uptake in the breast, which is indicative of cancerous tissue.
The main advantage of MBI is the ability of the technique to detect breast cancer at an early
stage. However, as with any imaging technique, although MBI offers good advantages over
other modalities for women with dense breasts, it is not without its downfalls. The main
disadvantage associated with MBI is the dose delivered to the patient as a result of the source
administered, which is significantly higher than the typical dose delivered in conventional
mammography. More recent studies have sought to develop the methodology in order to reduce
the activity of the source used in MBI, in turn reducing the dose delivered to the patient, without
compromising the sensitivity or specificity of the system. (1) Cadmium Zinc Telluride is a
desirable detector material for use in MBI primarily due to its good position resolution. This
property makes the detector highly sensitive to 141 keV gamma rays and therefore allows for an
isotope of lower activity to be administered to patients without compromising the image quality.
The ultimate project aim is to develop a CZT detector system optimised for application in MBI.
The project seeks to achieve the best possible position resolution of the detector system such
that a low activity source can be administered to the patient, hence reducing the dose delivered
and, in turn, the radiation risks associated with the imaging modality. Firstly, it is necessary to
quantify charge sharing effects in CZT detectors to enable the magnitude of the problem to be
understood and, ultimately, solved. A Geant4 based simulation study will aid the development of
algorithms which will aim to maximise the obtainable position resolution of the system.
Throughout the thesis, how different pixel geometries and collimator designs affect the position
resolution of the system will be investigated and analysed accordingly in order to determine the
optimal properties of the final detector system. The final system will b

Publications

10 25 50