Radionuclide identification method for NaI low-count gamma-ray spectra using artificial neural network
2
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
An artificial neural network (ANN) that identifies radionuclides from low-count gamma spectra of a NaIscintillator is proposed. The ANN was trained and tested using simulated spectra. 14 target nuclides wereconsidered corresponding to the requisite radionuclide library of a radionuclide identification devicementioned in IEC 62327-2017. The network shows an average identification accuracy of 98.63% on thevalidation dataset, with the gross counts in each spectrum Nc ¼ 100~10000 and the signal to noise ratioSNR ¼ 0.05e1. Most of the false predictions come from nuclides with low branching ratio and/or similardecay energies. If the Nc>1000 and SNR>0.3, which is defined as the minimum identifiable condition, theaveraged identification accuracy is 99.87%. Even when the source and the detector are covered with leadbricks and the response function of the detector thus varies, the ANN which was trained using nonshieldingspectra still shows high accuracy as long as the minimum identifiable condition is satisfied. Among all the considered nuclides, only the identification accuracy of 235U is seriously affected by theshielding. Identification of other nuclides shows high accuracy even the shielding condition is changed,which indicates that the ANN has good generalization performance.
Abstract
An artificial neural network (ANN) that identifies radionuclides from low-count gamma spectra of a NaIscintillator is proposed. The ANN was trained and tested using simulated spectra. 14 target nuclides wereconsidered corresponding to the requisite radionuclide library of a radionuclide identification devicementioned in IEC 62327-2017. The network shows an average identification accuracy of 98.63% on thevalidation dataset, with the gross counts in each spectrum Nc ¼ 100~10000 and the signal to noise ratioSNR ¼ 0.05e1. Most of the false predictions come from nuclides with low branching ratio and/or similardecay energies. If the Nc>1000 and SNR>0.3, which is defined as the minimum identifiable condition, theaveraged identification accuracy is 99.87%. Even when the source and the detector are covered with leadbricks and the response function of the detector thus varies, the ANN which was trained using nonshieldingspectra still shows high accuracy as long as the minimum identifiable condition is satisfied. Among all the considered nuclides, only the identification accuracy of 235U is seriously affected by theshielding. Identification of other nuclides shows high accuracy even the shielding condition is changed,which indicates that the ANN has good generalization performance.
참고문헌 (24)
- M. Alamaniotis, "Kernel-based machine learning for background estimation of NaI low-count gamma-ray spectra", IEEE Trans. Nucl. Sci, 60(3), pp.2209~2221 (2013)
- M. Alamaniotis, "Intelligent analysis of low-count scintillation spectra using support vector regression and fuzzy logic", Nucl. Technol, 191(1), pp.41~57 (2015)
- Benchmarking Algorithm for Radio Nuclide Identification (BARNI) Literature Review (No. LLNL-SR-767242), "M. Monterial", Lawrence Livermore National Lab.(LLNL), ~ (2019)
- J. Kim, "Quantitative analysis of NaI(Tl)gamma-ray spectrometry using an artificial neural network", Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip, 944, pp.162549~ (2019)
- J. Stinnett, "Uncertainty analysis of wavelet-based feature extraction for isotope identification on NaI gamma-ray spectra", IEEE Trans. Nucl. Sci, 64(7), pp.1670~1676 (2017)
- J. M. Blackadar, "Automatic isotope identifiers and their features", IEEE Sensor. J, 5(4), pp.589~592 (2005)
- S. Dragovic, "Artificial neural network data analysis for classification of soils based on their radionuclide content", Russ. J. Phys. Chem. A, 81(9), pp.1477~1481 (2007)
- H. Sahiner, "Gamma spectral analysis by artificial neural network coupled with Monte Carlo simulations", Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip, 953, pp.163062~ (2020)
- P. E. Keller, "Nuclear spectral analysis via artificial neural networks for waste handling", IEEE Trans. Nucl. Sci, 42(4), pp.709~715 (1995)
- C. Zhang, "Identification of SNM based on low-resolution gamma-ray characteristics and neural network", Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip, 927, pp.155~160 (2019)
- S. Dragovic, "Artificial neural network data analysis for classification of soils based on their radionuclide content", Russ. J. Phys. Chem. A, 81(9), pp.1477~1481 (2007)
- L. J. Kangas, "The use of artificial neural networks in PVT-based radiation portal monitors", Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip, 587(2-3), pp.398~412 (2008)
- J. He, "Rapid radionuclide identification algorithm based on the discrete cosine transform and BP neural network", Ann. Nucl. Energy, 112, pp.1~8 (2018)
- C. Bobin, "Real-time radionuclide identification in g-emitter mixtures based on spiking neural network", Appl. Radiat. Isot, 109, pp.405~409 (2016)
- M. Kamuda, "Automated isotope identification algorithm using artificial neural networks", IEEE Trans. Nucl. Sci, 64(7), pp.1858~1864 (2017)
- S.M. Galib, "Applications of machine learning in nuclear imaging and radiation detection", ~ (2019)
- N. Srivastava, "Dropout : a simple way to prevent neural networks from overfitting", J. Mach. Learn. Res, 15(1), pp.1929~1958 (2014)
- D.P. Kingma, "Adam: a method for stochastic optimization", ICLR 2015 : International Conference on Learning Representations 2015 (2015)
- K. Hu, "Retinal vessel segmentation of color fundus images using multiscale convolutional neural network with an improved cross-entropy loss function", Neurocomputing, 309, pp.179~191 (2018)
- Y. Wu, "CAD-based Monte Carlo program for integrated simulation of nuclear system SuperMC", Ann. Nucl. Energy, 82, pp.161~168 (2015)
- "Gamma Library" (2020)[URL]
- J. Kim, "Multi-radioisotope identification algorithm using an artificial neural network for plastic gamma spectra", Appl. Radiat. Isot, 147, pp.83~90 (2019)
- S. Jhung, "A neural network approach for identification of gamma-ray spectrum obtained from silicon photomultipliers", Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip, 954, pp.161704~ (2020)
- C. Kim, "Radiation resistance and temperature dependence of Ce : GPS scintillation crystal", Radiat. Phys. Chem, 183, pp.109396~ (2021)