Application of the ML-EM algorithm in the suppression of Compton background of gamma-ray spectroscopy system

In the present study, the Compton background of the measured gamma-ray spectrum was suppressed by using the deconvolution method with applying the maximum likelihood fitting by expectation maximization (MLEM) algorithm. The MATLAB program for deconvolution of the measured spectrum were written based on the applying of ML-EM algorithm in conjunction with a matrix of detector response functions within the energy range 0-1.9 MeV. The Compton Suppression Factor (SF) ratio of the deconvolved spectrum is increased by approximately 3-9 times relative to the measured spectrum of ¹³⁷Cs and ⁶⁰Co, respectively. Additionally, the photopeak areas of deconvolved spectra were considerably raised to compare with ones of measured spectra owing to accumulating the counting numbers of the background into appropriate photopeaks. (author)