DOI QR코드

DOI QR Code

Explainable radionuclide identification algorithm based on the convolutional neural network and class activation mapping

  • 투고 : 2022.06.09
  • 심사 : 2022.08.11
  • 발행 : 2022.12.25

초록

Radionuclide identification is an important part of the nuclear material identification system. The development of artificial intelligence and machine learning has made nuclide identification rapid and automatic. However, many methods directly use existing deep learning models to analyze the gamma-ray spectrum, which lacks interpretability for researchers. This study proposes an explainable radionuclide identification algorithm based on the convolutional neural network and class activation mapping. This method shows the area of interest of the neural network on the gamma-ray spectrum by generating a class activation map. We analyzed the class activation map of the gamma-ray spectrum of different types, different gross counts, and different signal-to-noise ratios. The results show that the convolutional neural network attempted to learn the relationship between the input gamma-ray spectrum and the nuclide type, and could identify the nuclide based on the photoelectric peak and Compton edge. Furthermore, the results explain why the neural network could identify gamma-ray spectra with low counts and low signal-to-noise ratios. Thus, the findings improve researchers' confidence in the ability of neural networks to identify nuclides and promote the application of artificial intelligence methods in the field of nuclide identification.

키워드

참고문헌

  1. M. Gomez-Fernandez, K. Higley, A. Tokuhiro, K. Welter, W.-K. Wong, H. Yang, Status of research and development of learning-based approaches in nuclear science and engineering: a review, Nucl. Eng. Des. 359 (2020), 110479, https://doi.org/10.1016/j.nucengdes.2019.110479.
  2. B.T. Koo, H.C. Lee, K. Bae, Y. Kim, J. Jung, C.S. Park, H.-S. Kim, C.H. Min, Development of a radionuclide identification algorithm based on a convolutional neural network for radiation portal monitoring system, Radiat. Phys. Chem. 180 (2021), 109300, https://doi.org/10.1016/j.radphyschem.2020.109300.
  3. J.T. Mihalczo, J. Mattingly, J. Neal, J. Mullens, NMIS plus gamma spectroscopy for attributes of HEU, PU and HE detection, Nucl. Instrum. Methods Phys. Res. Sect. B: Beam Interact. Mater. Atoms 213 (2004) 378-384, https://doi.org/10.1016/S0168-583X(03)01651-3.
  4. M.J. Aitkenhead, M. Owen, D.M. Chambers, Use of artificial neural networks in measuring characteristics of shielded plutonium for arms control, J. Anal. At. Spectrom. 27 (2012) 432-439, https://doi.org/10.1039/C2JA10230G.
  5. C. Zhang, G. Hu, F. Luo, Y. Xiang, G. Ding, C. Chu, J. Zeng, R. Ze, Q. Xiang, Identification of SNM based on low-resolution gamma-ray characteristics and neural network, Nucl. Instrum. Methods in Phys. Res. Sect. A:Accel. Spectrom. Detect. Assoc. Equip. 927 (2019) 155-160, https://doi.org/10.1016/j.nima.2019.02.023.
  6. M. Kamuda, J. Zhao, K. Huff, A comparison of machine learning methods for automated gamma-ray spectroscopy, Nucl. Instrum. Methods in Phys. Res. Sect. A:Accel. Spectrom. Detect. Assoc. Equip. 954 (2020), 161385, https://doi.org/10.1016/j.nima.2018.10.063.
  7. D. Liang, P. Gong, X. Tang, P. Wang, L. Gao, Z. Wang, R. Zhang, Rapid nuclide identification algorithm based on convolutional neural network, Annals of Nuclear Energy 133 (2019) 483-490, https://doi.org/10.1016/j.anucene.2019.05.051.
  8. S.M. Galib, P.K. Bhowmik, A.V. Avachat, H.K. Lee, A comparative study of machine learning methods for automated identification of radioisotopes using NaI gamma-ray spectra, Nucl. Eng. Technol. 53 (2021) 4072-4079, https://doi.org/10.1016/j.net.2021.06.020.
  9. S. Wu, X. Tang, P. Gong, P. Wang, D. Liang, Y. Li, C. Zhou, X. Zhu, Peak-searching method for low count rate g spectrum under short-time measurement based on a generative adversarial network, Nucl. Instrum. Methods in Phys. Res. Sect. A:Accel. Spectrom. Detect. Assoc. Equip. 1002 (2021), 165262, https://doi.org/10.1016/j.nima.2021.165262.
  10. S. Qi, S. Wang, Y. Chen, K. Zhang, X. Ai, J. Li, H. Fan, H. Zhao, Radionuclide identification method for NaI low-count gamma-ray spectra using artificial neural network, Nucl. Eng. Technol. 54 (2022) 269-274, https://doi.org/10.1016/j.net.2021.07.025.
  11. C. Li, S. Liu, C. Wang, X. Jiang, X. Sun, M. Li, L. Wei, A new radionuclide identification method for low-count energy spectra with multiple radionuclides, Applied Radiation and Isotopes 185 (2022), 110219, https://doi.org/10.1016/j.apradiso.2022.110219.
  12. I. Rahwan, M. Cebrian, N. Obradovich, J. Bongard, J.-F. Bonnefon, C. Breazeal, J.W. Crandall, N.A. Christakis, I.D. Couzin, M.O. Jackson, N.R. Jennings, E. Kamar, I.M. Kloumann, H. Larochelle, D. Lazer, R. McElreath, A. Mislove, D.C. Parkes, A. 'Sandy' Pentland, M.E. Roberts, A. Shariff, J.B. Tenenbaum, M. Wellman, Machine behaviour, Nature 568 (2019) 477-486, https://doi.org/10.1038/s41586-019-1138-y.
  13. W. Samek, T. Wiegand, K.-R. Muller, Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models, 2017, p. 10, https://doi.org/10.48550/arXiv.1708.08296.
  14. G. Daniel, F. Ceraudo, O. Limousin, D. Maier, A. Meuris, Automatic and real-time identification of radionuclides in gamma-ray spectra: a new method based on convolutional neural network trained with synthetic data set, IEEE Trans. Nucl. Sci. 67 (2020) 644-653, https://doi.org/10.1109/TNS.2020.2969703.
  15. M. Gomez-Fernandez, W.-K. Wong, A. Tokuhiro, K. Welter, A.M. Alhawsawi, H. Yang, K. Higley, Isotope identification using deep learning: an explanation, Nucl. Instrum. Methods in Phys. Res. Sect. A:Accel. Spectrom. Detect. Assoc. Equip. 988 (2021), 164925, https://doi.org/10.1016/j.nima.2020.164925.
  16. J. Ryu, C. Park, J. Park, N. Cho, J. Park, G. Cho, Development of neural network model with explainable AI for measuring uranium enrichment, IEEE Trans. Nucl. Sci. 68 (2021) 2670-2681, https://doi.org/10.1109/TNS.2021.3116090.
  17. S. Agostinelli, J. Allison, K. Amako, J. Apostolakis, H. Araujo, P. Arce, Geant4-a simulation toolkit, Nucl. Instrum. Methods in Phys. Res. Sect. A:Accel. Spectrom. Detect. Assoc. Equip. 506 (2003) 250-303, https://doi.org/10.1016/S0168-9002(03)01368-8.
  18. IAEA, Technical and Functional Specifications for Border Monitoring Equipment, INTERNATIONAL ATOMIC ENERGY AGENCY, Vienna, 2006. https://www.iaea.org/publications/7400/technical-and-functional-specifications-for-border-monitoring-equipment.
  19. A. Krizhevsky, Learning Multiple Layers of Features from Tiny Images, undefined, 2009. https://www.semanticscholar.org/paper/Learning-MultipleLayers-of-Features-from-Tiny-Krizhevsky/5d90f06bb70a0a3dced62413346235c02b1aa086. (Accessed 8 August 2022). accessed.
  20. J. Snell, K. Swersky, R.S. Zemel, Prototypical Networks for Few-Shot Learning, 2017, https://doi.org/10.48550/arXiv.1703.05175 arXiv.
  21. M. Ren, E. Triantafillou, S. Ravi, J. Snell, K. Swersky, J.B. Tenenbaum, H. Larochelle, R.S. Zemel, Meta-Learning for Semi-supervised Few-Shot Classification, 2018, https://doi.org/10.48550/arXiv.1803.00676 arXiv.
  22. A. Krizhevsky, I. Sutskever, G.E. Hinton, ImageNet classification with deep convolutional neural networks, Commun. ACM. 60 (2017) 84-90, https://doi.org/10.1145/3065386.
  23. A. Karpathy, G. Toderici, S. Shetty, T. Leung, R. Sukthankar, L. Fei-Fei, Largescale Video Classification with Convolutional Neural Networks, 2014, p. 8.
  24. X. Zeng, W. Ouyang, B. Yang, J. Yan, X. Wang, Gated Bi-directional CNN for object detection, in: B. Leibe, J. Matas, N. Sebe, M. Welling (Eds.), Computer Vision - ECCV 2016, Springer International Publishing, Cham, 2016, pp. 354-369, https://doi.org/10.1007/978-3-319-46478-7_22.
  25. Martin Abadi, Ashish Agarwal, Barham Paul, Eugene Brevdo, Zhifeng Chen, Citro Craig, Greg S. Corrado, Andy Davis, TensorFlow, Large-Scale Machine Learning on Heterogeneous Systems, 2015. http://tensorflow.org/.
  26. C. Seifert, A. Aamir, A. Balagopalan, D. Jain, A. Sharma, S. Grottel, S. Gumhold, Visualizations of deep neural networks in computer vision: a survey, in: T. Cerquitelli, D. Quercia, F. Pasquale (Eds.), Transparent Data Mining for Big and Small Data, Springer International Publishing, Cham, 2017, pp. 123-144, https://doi.org/10.1007/978-3-319-54024-5_6.
  27. D.V. Carvalho, E.M. Pereira, J.S. Cardoso, Machine learning interpretability: a survey on methods and metrics, Electronics 8 (2019) 832, https://doi.org/10.3390/electronics8080832.
  28. W. Samek, G. Montavon, S. Lapuschkin, C.J. Anders, K.-R. Muller, Toward Interpretable Machine Learning: Transparent Deep Neural Networks and beyond, CoRR. Abs/2003, 2020, 07631. https://arxiv.org/abs/2003.07631.
  29. B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, A. Torralba, Learning deep features for discriminative localization, in: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Las Vegas, NV, USA, 2016, pp. 2921-2929, https://doi.org/10.1109/CVPR.2016.319.
  30. People's Republic of China national metrology specification, Calibration specification for hand-held radiation monitors for detection and identification of radionuclides (JJF 1687-2018), General Administration of Quality Supervision, Inspection and Quarantine, Beijing, China, 2018 (in Chinese).