System and method for separating gamma spectrum of internal whole body contamination test equipment (WBC) for generating training data for artificial neural network (ANN)

Abstract

The present invention relates to a system for separating gamma spectra of internal whole-body contamination testing equipment for generating artificial neural network training data, and more particularly to a system for separating a spectrum of a single nuclide from a spectrum of a mixture of various nuclides to generate training data for an artificial neural network. The present invention includes an efficiency calculation unit for calculating the efficiency of each of A, B, and C channels of the internal whole-body contamination testing equipment, a counting contribution calculation unit for calculating a channel-by-channel counting rate contribution by multiplying a mixed radiation source by the gamma-ray emission rate of each nuclide for generating artificial neural network training data, and a normalization unit for subtracting the channel-by-channel counting rate contribution from the channel-by-channel counting rate measured by the nuclide and normalizing the result to obtain a normalized nuclide spectrum, thereby enabling the separation of a spectrum of a single nuclide from a spectrum of a mixture of various nuclides and the generation of a synthetic spectrum for artificial intelligence training.

Images