PARAMETERIZATION OF THE SCATTER RESPONSE FUNCTION IN SPECT IMAGING USING MONTE-CARLO SIMULATION

In single photon emission computed tomography (SPECT), scattered radiation results in a loss of image contrast, spatial resolution and quantitative accuracy. In order to compensate accurately for these effects, the scatter response function (SRF) must be known. However, due to the incoherent nature of scatter radiation and the complex dependence of the SRF on many physical and imaging parameters, including primary photon energy, energy resolution and energy window of the detection system, the distribution and configuration of the radioactive sources, and the composition and geometry of the scattering medium, it is difficult to derive an analytic expression for the SRF. We have generated SRFs using the Monte Carlo simulation method and investigated the characteristics of the scattered radiation by fitting the SRFs with fitting functions. The parameters of the fitting functions were studied as a function of source position in a water filled cylindrical phantom with circular cross section. The parameterization of the SRF provides insight into the spatial distribution of detected scattered radiation in SPECT and is useful in developing methods to compensate for its effects in order to improve both the quality and quantitative accuracy of SPECT images. © 1990 IEEE