Journal of Radiation Protection and Research

  • 제46권4호
  • /
  • Pages.204-212
  • /
  • 2021
  • /
  • 2508-1888(pISSN)
  • /
  • 2466-2461(eISSN)
대한방사선방어학회 (The Korean Association for Radiation Protection)
권호 표지
DOI QR코드

DOI QR Code

Comparison of Machine Learning-Based Radioisotope Identifiers for Plastic Scintillation Detector

  • Jeon, Byoungil (Artificial Intelligence Application & Strategy Team, Korea Atomic Energy Research Institute) ;
  • Kim, Jongyul (Neutron Science Division, Korea Atomic Energy Research Institute) ;
  • Yu, Yonggyun (Artificial Intelligence Application & Strategy Team, Korea Atomic Energy Research Institute) ;
  • Moon, Myungkook (Neutron Science Division, Korea Atomic Energy Research Institute)
  • 투고 : 2021.05.04
  • 심사 : 2021.06.17
  • 발행 : 2021.12.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Background: Identification of radioisotopes for plastic scintillation detectors is challenging because their spectra have poor energy resolutions and lack photo peaks. To overcome this weakness, many researchers have conducted radioisotope identification studies using machine learning algorithms; however, the effect of data normalization on radioisotope identification has not been addressed yet. Furthermore, studies on machine learning-based radioisotope identifiers for plastic scintillation detectors are limited. Materials and Methods: In this study, machine learning-based radioisotope identifiers were implemented, and their performances according to data normalization methods were compared. Eight classes of radioisotopes consisting of combinations of 22Na, 60Co, and 137Cs, and the background, were defined. The training set was generated by the random sampling technique based on probabilistic density functions acquired by experiments and simulations, and test set was acquired by experiments. Support vector machine (SVM), artificial neural network (ANN), and convolutional neural network (CNN) were implemented as radioisotope identifiers with six data normalization methods, and trained using the generated training set. Results and Discussion: The implemented identifiers were evaluated by test sets acquired by experiments with and without gain shifts to confirm the robustness of the identifiers against the gain shift effect. Among the three machine learning-based radioisotope identifiers, prediction accuracy followed the order SVM > ANN > CNN, while the training time followed the order SVM > ANN > CNN. Conclusion: The prediction accuracy for the combined test sets was highest with the SVM. The CNN exhibited a minimum variation in prediction accuracy for each class, even though it had the lowest prediction accuracy for the combined test sets among three identifiers. The SVM exhibited the highest prediction accuracy for the combined test sets, and its training time was the shortest among three identifiers.

키워드

과제정보

This work was supported by the Korea Atomic Energy Research Institute (No. 79501-19), Ministry of Oceans and Fisheries (KIMST) (No. 20200611).

참고문헌

  1. Stromswold DC, Darkoch JW, Ely JH, Hansen RR, Kouzes RT, Milbrath BD, et al. Field tests of a NaI(Tl)-based vehicle portal monitor at border crossings. Proceedings of IEEE Symposium Conference Record Nuclear Science; 2004 Oct 16-22; Rome, Italy. p. 196-200.
  2. Kouzes RT, Siciliano ER. The response of radiation portal monitors to medical radionuclides at border crossings. Radiat Meas. 2006;41(5):499-512. https://doi.org/10.1016/j.radmeas.2005.10.005
  3. Ludlum Measurements Inc. Radiation portal monitors (Model 4525 Series) [Internet]. Sweetwater, TX: Ludlum Measurements Inc.; c2021 [cited 2021 Aug 28]. Available from: https://ludlums.com/products/all-products/product/model-4525-series.
  4. Thermo Fisher Scientific Inc. Radiation portal monitors [Internet]. Waltham, MA: Thermo Fisher Scientific Inc.; c2021 [cited 2021 Aug 28]. Available from: https://www.thermofisher.com/order/catalog/product/4254982?SID=srch-srp-4254982.
  5. Pacific Northwest National Laboratory. Radiation detectors at U.S. ports of entry now operate more effectively, efficiently [Internet]. Richland, WA: Pacific Northwest National Laboratory; 2016 [cited 2021 Aug 29]. Available from: https://www.pnnl.gov/news/release.aspx?id=4245.
  6. Anderson KK, Jarman KD, Mann ML, Pfund DM, Runkle RC. Discriminating nuclear threats from benign sources in gammaray spectra using a spectral comparison ratio method. J Radioanal Nucl Chem. 2008;276(3):713-718. https://doi.org/10.1007/s10967-008-0622-x
  7. Ely J, Kouzes R, Schweppe J, Siciliano E, Strachan D, Weier D. The use of energy windowing to discriminate SNM from NORM in radiation portal monitors. Nucl Instrum Methods Phys Res A. 2006;560(2):373-387. https://doi.org/10.1016/j.nima.2006.01.053
  8. Hevener R, Yim MS, Baird K. Investigation of energy windowing algorithms for effective cargo screening with radiation portal monitors. Radiat Meas. 2013;58:113-120. https://doi.org/10.1016/j.radmeas.2013.08.004
  9. Siciliano ER, Ely JH, Kouzes RT, Milbrath BD, Schweppe JE, Stromswold DC. Comparison of PVT and NaI(Tl) scintillators for vehicle portal monitor applications. Nucl Instrum Methods Phys Res A. 2005;550(3):647-674. https://doi.org/10.1016/j.nima.2005.05.056
  10. Shin WG, Lee HC, Choi CI, Park CS, Kim HS, Min CH. A Monte Carlo study of an energy-weighted algorithm for radionuclide analysis with a plastic scintillation detector. Appl Radiat Isot. 2015;101:53-59. https://doi.org/10.1016/j.apradiso.2015.03.014
  11. Lee HC, Shin WG, Park HJ, Yoo DH, Choi CI, Park CS, et al. Validation of energy-weighted algorithm for radiation portal monitor using plastic scintillator. Appl Radiat Isot. 2016;107:160-164. https://doi.org/10.1016/j.apradiso.2015.10.019
  12. Paff MG, Di Fulvio A, Clarke SD, Pozzi SA. Radionuclide identification algorithm for organic scintillator-based radiation portal monitor. Nucl Instrum Methods Phys Res A. 2017;849:41-48. https://doi.org/10.1016/j.nima.2017.01.009
  13. Russ WR. Library correlation nuclide identification algorithm. Nucl Instrum Methods Phys Res A. 2007;579(1):288-291. https://doi.org/10.1016/j.nima.2007.04.062
  14. Min E, Ko M, Lee H, Kim Y, Joung J, Joo SK, et al. Identification of radionuclides for the spectroscopic radiation portal monitor for pedestrian screening under a low signal-to-noise ratio condition. Nucl Instrum Methods Phys Res A. 2014;758:62-68. https://doi.org/10.1016/j.nima.2014.05.021
  15. Hague EJ, Kamuda M, Ford WP, Moore ET, Turk J. A comparison of adaptive and template matching techniques for radio-isotope identification. In: Algorithms, technologies, and applications for multispectral and hyperspectral imagery XXV. Bellingham, WA: International Society for Optics and Photonics; 2019.
  16. Runkle RC, Tardiff MF, Anderson KK, Carlson DK, Smith LE. Analysis of spectroscopic radiation portal monitor data using principal components analysis. IEEE Trans Nucl Sci. 2006;53(3): 1418-1423. https://doi.org/10.1109/TNS.2006.874883
  17. Boardman D, Flynn A. A gamma-ray identification algorithm based on Fisher linear discriminant analysis. IEEE Trans Nucl Sci. 2012;60(1):270-277. https://doi.org/10.1109/TNS.2012.2226472
  18. Sullivan CJ, Stinnett J. Validation of a Bayesian-based isotope identification algorithm. Nucl Instrum Methods Phys Res A. 2015;784:298-305. https://doi.org/10.1016/j.nima.2014.11.113
  19. Ruch ML, Paff M, Sagadevan A, Clarke SD, Pozzi SA. Radionuclide identification by an EJ309 organic scintillator-based pedestrian radiation portal monitor using a least squares algorithm. Proceedings of the 55th Annual Meeting of Nuclear Materials Management; 2014 Jul 20-24; Atlanta, GA. p. 22-24.
  20. Kim Y, Kim M, Lim KT, Kim J, Cho G. Inverse calibration matrix algorithm for radiation detection portal monitors. Radiat Phys Chem. 2019;155:127-132. https://doi.org/10.1016/j.radphyschem.2018.07.022
  21. Medhat ME. Artificial intelligence methods applied for quantitative analysis of natural radioactive sources. Ann Nucl Energy. 2012;45:73-79. https://doi.org/10.1016/j.anucene.2012.02.013
  22. Sheinfeld M, Levinson S, Orion I. Highly accurate prediction of specific activity using deep learning. Appl Radiat Isot. 2017;130: 115-120. https://doi.org/10.1016/j.apradiso.2017.09.023
  23. Hata H, Yokoyama K, Ishimori Y, Ohara Y, Tanaka Y, Sugitsue N. Application of support vector machine to rapid classification of uranium waste drums using low-resolution γ-ray spectra. Appl Radiat Isot. 2015;104:143-146. https://doi.org/10.1016/j.apradiso.2015.06.030
  24. Abdel-Aal RE, Al-Haddad MN. Determination of radioisotopes in gamma-ray spectroscopy using abductive machine learning. Nucl Instrum Methods Phys Res A. 1997;391(2):275-288. https://doi.org/10.1016/S0168-9002(97)00391-4
  25. Bobin C, Bichler O, Lourenco V, Thiam C, Thevenin M. Real-time radionuclide identification in γ-emitter mixtures based on spiking neural network. Appl Radiat Isot. 2016;109:405-409. https://doi.org/10.1016/j.apradiso.2015.12.029
  26. Kamuda M, Stinnett J, Sullivan CJ. Automated isotope identification algorithm using artificial neural networks. IEEE Trans Nucl Sci. 2017;64(7):1858-1864. https://doi.org/10.1109/TNS.2017.2693152
  27. He J, Tang X, Gong P, Wang P, Wen L, Huang X, et al. Rapid radionuclide identification algorithm based on the discrete cosine transform and BP neural network. Annals of Nuclear Energy. 2018;112:1-8. https://doi.org/10.1016/j.anucene.2017.09.032
  28. Kamuda M, Sullivan CJ. An automated isotope identification and quantification algorithm for isotope mixtures in low-resolution gamma-ray spectra. Radiat Phys Chem. 2019;155:281-286. https://doi.org/10.1016/j.radphyschem.2018.06.017
  29. Murray SJ, Schmitz J, Balkir S, Hoffman MW. A low complexity radioisotope identification system using an integrated multichannel analyzer and embedded neural network. Proceedings of 2019 IEEE International Symposium on Circuits and Systems (ISCAS); 2019 May 26-29; Sapporo, Japan. p. 1-5.
  30. Kamuda M, Zhao J, Huff K. A comparison of machine learning methods for automated gamma-ray spectroscopy. Nucl Instrum Methods Phys Res A. 2020;954:161385. https://doi.org/10.1016/j.nima.2018.10.063
  31. Kangas LJ, Keller PE, Siciliano ER, Kouzes RT, Ely JH. The use of artificial neural networks in PVT-based radiation portal monitors. Nucl Instrum Methods Phys Res A. 2008;587(2-3):398-412. https://doi.org/10.1016/j.nima.2008.01.065
  32. Kim J, Park K, Cho G. Multi-radioisotope identification algorithm using an artificial neural network for plastic gamma spectra. Appl Radiat Isot. 2019;147:83-90. https://doi.org/10.1016/j.apradiso.2019.01.005
  33. Jeon B, Lee Y, Moon M, Kim J, Cho G. Reconstruction of Compton edges in plastic gamma spectra using deep autoencoder. Sensors (Basel). 2020;20(10):2895. https://doi.org/10.3390/s20102895
  34. Jeon B, Kim J, Lee E, Moon M, Cho G. Pseudo-gamma spectroscopy based on plastic scintillation detectors using multitask learning. Sensors (Basel). 2021;21(3):684. https://doi.org/10.3390/s21030684
  35. Werner CJ, Bull JS, Solomon CJ, Brown FB, McKinney GW, Rising ME, et al. MCNP version 6.2 Release notes (No. LA-UR-18-20808). Los Alamos, NM; Los Alamos National Laboratory; 2018.
  36. McConn RJ, Gesh CJ, Pagh RT, Rucker RA, Williams R. Compendium of material composition data for radiation transport modeling (No. PNNL-15870). Richland, WA; Pacific Northwest National Laboratory; 2011.
  37. Jeon B, Kim J, Moon M, Cho G. Parametric optimization for energy calibration and gamma response function of plastic scintillation detectors using a genetic algorithm. Nucl Instrum Methods Phys Res A. 2019;930:8-14. https://doi.org/10.1016/j.nima.2019.03.003
  38. Chang CC, Lin CJ. LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol. 2011;2(3):1-27. https://doi.org/10.1145/1961189.1961199
  39. Abadi M, Agarwal A, Barham P, Bervdo E, Chen Z, Citro C, et al. TensorFlow: large-scale machine learning on heterogeneous distributed systems [Internet]. Ithaca, NY: arXiv.org; 2016 [cited 2021 Aug 30]. Available from: https://arxiv.org/abs/1603.04467.
  40. Snoek J, Larochelle H, Adams RP. Practical Bayesian optimization of machine learning algorithms. Adv Neural Inf Process Syst. 2012;25:2960-2968.
  41. Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. Proceedings of the 3rd International Conference on Learning Representations (ICLR); 2015 May 7-9; San Diego, CA.
  42. Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, et al. Going deeper with convolutions. Proceedings of the IEEE conference on computer vision and pattern recognition; 2015 Jun 7-12; Boston, MA. p. 1-9.
  43. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition; 2016 Jun 27-30; Las Vegas, NV. p. 770-778.
  44. Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convolutional networks. Proceedings of the IEEE conference on computer vision and pattern recognition; 2017 Jul 21-26; Honolulu, HI. p. 2261-2269.