| Summary: | Beta-emitting radioisotopes such as 90y & 32p do not emit gamma radiation, and so their detection during radionuclide therapies relies on the bremsstrahlung photons released when electrons interact in tissue. The aim of this project was to optimise acquisition parameters (energy window and collimator) for imaging the complex spectra, which are continuous up to relatively high energies, and are of low intensity. Experimental work and theoretical explorations used a combination of list-mode acquisition on an ADAC Forte gamma camera and EGSnrc Monte Carlo simulations. Initially, the camera's energy linearity was investigated and appropriate settings selected for the wide energy range. Photon kernels were developed to approximate beta sources in simulations and were shown to increase speeds significantly. Acquisitions and simulations were then made of a range of phantoms containing 90y or 32p, with low, meditirn and high energy collimation. The data were binned into narrow energy windows, and the resulting images were assessed for vari a tion in qual ity with energy and collima tor, and with parameters such as depth and source-to-background ra tio. Medium energy collimation was found to offer the best compromise between sensitivity and spatial resolution. Image contrast was highest in energy windows near 100 keY, but signal-to-noise ratios (SNR) were highest at lower energies. Wide windows showed improved SNR without significant loss of spatial resolution or contrast. A setting of 60 to 170 keY was selected for the clinic, allowing for practical limits on the camera's window width. The Monte Carlo simulations demonstrated which photon interactions led to these results, for example at which energies septal penetration began to dominate, and how much of the blurring at low energies was due to characteristic x-rays produced in the collimator. The results of this work have been used for initial investigations into the optimisation of analogue imaging with gamma-emitting radioisotopes, for example 111In for 9OY.
|
|---|
