DOI QR코드

DOI QR Code

Comparison of Machine Learning Classification Models for the Development of Simulators for General X-ray Examination Education

일반엑스선검사 교육용 시뮬레이터 개발을 위한 기계학습 분류모델 비교

  • Lee, In-Ja (Department of Radiological Technology, Dongnam Health University) ;
  • Park, Chae-Yeon (Department of Radiological Technology, Dongnam Health University) ;
  • Lee, Jun-Ho (Business Support Team, Korea Medical Device Development Fund)
  • 이인자 (동남보건대학교 방사선학과) ;
  • 박채연 (동남보건대학교 방사선학과) ;
  • 이준호 (범부처전주기의료기기연구개발사업단 사업화지원팀)
  • Received : 2021.12.23
  • Accepted : 2022.03.04
  • Published : 2022.04.30

Abstract

In this study, the applicability of machine learning for the development of a simulator for general X-ray examination education is evaluated. To this end, k-nearest neighbor(kNN), support vector machine(SVM) and neural network(NN) classification models are analyzed to present the most suitable model by analyzing the results. Image data was obtained by taking 100 photos each corresponding to Posterior anterior(PA), Posterior anterior oblique(Obl), Lateral(Lat), Fan lateral(Fan lat). 70% of the acquired 400 image data were used as training sets for learning machine learning models and 30% were used as test sets for evaluation. and prediction model was constructed for right-handed PA, Obl, Lat, Fan lat image classification. Based on the data set, after constructing the classification model using the kNN, SVM, and NN models, each model was compared through an error matrix. As a result of the evaluation, the accuracy of kNN was 0.967 area under curve(AUC) was 0.993, and the accuracy of SVM was 0.992 AUC was 1.000. The accuracy of NN was 0.992 and AUC was 0.999, which was slightly lower in kNN, but all three models recorded high accuracy and AUC. In this study, right-handed PA, Obl, Lat, Fan lat images were classified and predicted using the machine learning classification models, kNN, SVM, and NN models. The prediction showed that SVM and NN were the same at 0.992, and AUC was similar at 1.000 and 0.999, indicating that both models showed high predictive power and were applicable to educational simulators.

Keywords

Acknowledgement

This work was supported by the research grant of the Dongnam Health University.

References

  1. KDCA. Assessment of radiation exposure of Korean population by medical radiation. Korea Disease Control and Prevention Agency; 2020.
  2. Jeon SM, Lee YK, Ahn SM. A Study on the exposure dose of workers and frequent workers in the radiology department. Journal of the Korean Society of Radiology. 2021;15(3):355-9. https://doi.org/10.7742/JKSR.2021.15.3.355
  3. Kang SS, Kim CS, Choi SY, Go SJ, Kim JH. Evaluation of present curriculum for development of dept. of radiological science curriculum. The Journal of the Korea Contents Association. 2011;11(5):242-51. https://doi.org/10.5392/JKCA.2011.11.5.242
  4. Kil JW, Park JH, Kim YG. Study on the planning and operation of training education in radiologic science for reduced x-ray exposure. Journal of the Institute of Electronics and Information Engineers. 2014;51(12):174-9. https://doi.org/10.5573/IEIE.2014.51.12.174
  5. Shim JG, Kwon SM. Analysis of learning effect through the development and application of virtual reality(VR) education content for radiology students. Journal of Radiological Science and Technology. 2020;43(6):519-24. https://doi.org/10.17946/JRST.2020.43.6.519
  6. Park HH, Shim JG, Kwon SM. Mixed reality based radiation safety education simulator platform development: Focused on medical field. Journal of Radiological Science and Technology. 2021;44(2):123-31. https://doi.org/10.17946/JRST.2021.44.2.123
  7. Hong DH. Comparison of CT exposure dose prediction models using machine learning-based body measurement information. Journal of Radiological Science and Technology. 2020;43(6):503-9. https://doi.org/10.17946/JRST.2020.43.6.503
  8. Kwon CM. Python Machine Learning Complete Guide. Gyeonggi: Wikibooks; 2019.
  9. Park DR, Ahn JM, Jang JH, Yu WJ, Kim WY, Bae YK, et al. The Development of software teaching-learning model based on machine learning platform. Journal of The Korean Association of Information Education. 2020;24(1):49-57. https://doi.org/10.14352/jkaie.2020.24.1.49
  10. Kang IW, Sharma R, Jeon SM, Park S, Lee SH, Na YH, et al. Optimized KNN/SVM algorithm for efficent indoor location. Proceedings of the Korea Information Processing Society Conference. 2011;18(2):602-5.
  11. Lee BW, Kim CG. A Study on the convergence perception of students in radiology on the reorganization of safety management system by person with frequent access of nuclear safety act. Journal of the Korea Convergence Society. 2019;10(6):89-94. https://doi.org/10.15207/JKCS.2019.10.6.089
  12. Choi PS, Min IS. A Predictive model for the employment of college graduates using a machine learning approach. Journal of Vocational Education & Training. 2018;21(1):31-54. https://doi.org/10.36907/KRIVET.2018.21.1.31
  13. Lee GS, Lee JC. A Classification of medical and advertising blogs using machine learning. Journal of Korea Academia-Industrial cooperation Society. 2018;19(11):730-7. https://doi.org/10.5762/KAIS.2018.19.11.730
  14. Eom JS, Lee SW, Kim BY. A Feasibility study on the improvement of diagnostic accuracy for energy-selective digital mammography using machine learning. Journal of Radiological Science and Technology. 2019;42(1):9-17. https://doi.org/10.17946/JRST.2019.42.1.9
  15. Lee IJ, Lee JH. Predictive of osteoporosis by tree-based machine learning model in post-menopause woman. Journal of Radiological Science and Technology. 2020;43(6):495-502. https://doi.org/10.17946/JRST.2020.43.6.495