A beta-gamma position-sensitive detector based on rod plastic scintillators and artificial neural networks

Abstract

Mixed fields of gamma and beta radiations may exist under different conditions such as radiation dosimetry, PET medical imaging, medical therapy treatments, space radiation detection, etc. In these circumstances, it is important to determine the dose rate, intensity, incident position of the beta particles and gamma radiation separately. In this work, a radiation detector was developed, which could monitor the incident position and type of beta particles and gamma rays in the mixed radiation fields. The detector consisted of an ø 5.5 × 5 cm rod plastic scintillator coupled to a 2′′ photomultiplier at one end. 137Cs and 147Pm were employed as gamma and beta sources, respectively. Two sources were placed in different positions in front of the detector and the related spectra were recorded using a multi-channel analyzer. An artificial neural network consisting of a three-layer perceptron was developed and trained using the recorded mixed gamma and beta spectra to predict their type and position.