A sophisticated estimation of scatter component in energy spectra using an artificial neural network in radionuclide imaging

The authors present a novel method for estimating primary photons using an artificial neural network in radionuclide imaging. The neural network for Tc-99m has three layers, one input layer with five units, one hidden layer with five units, and one output layer with two units. As input values to the input units, count ratios were used which were the ratios of the counts acquired by narrow windows to the total count acquired by a broad window with the energy range from 125 to 154 keV. The outputs were a scatter count ratio and a primary count ratio. Using the primary count ratio and the total count, the primary count of the pixel was calculated directly. The neural network was trained with a backpropagation algorithm using calculated true energy spectra obtained by a Monte Carlo method. The simulation showed that accurate estimation of primary photons was accomplished within an error ratio of about 3% for primary photons