[2J01-08] Dose Assessment

[2J03] Radiation source distribution estimation by machine learning using gamma ray energy spectrum

*Yuka Kumada¹, Taiyo Sato¹, Yuto Kondo¹, Tsugiko Takase¹, Masaharu Matsumoto¹, Katsuhiko Yamaguchi¹ (1. Fukushima Univ.)

Keywords:Machine Learning, Radiation source distribution estimation, PHITS, γ-ray energy spectrum

We developed a technique for estimating radiation source distribution on a two-dimensional plane using machine learning. Spectral data from MC simulations for radiation transport calculations (PHITS) are used to learn a neural network(NN). At the same time, we developed another machine learning that transforms the shape of the energy spectra between measurement and simulation. By inputting the converted measured spectral data into the system for radiation source distribution estimation, highly accurate distribution estimation results were obtained.