TY  - JOUR
T1  - Development and Validation of a Random Forest Model for X-Ray Shielding Thickness Prediction in Myanmar Healthcare Facilities Based on NCRP Report 147
TT  - Development and Validation of a Random Forest Model for X-Ray Shielding Thickness Prediction in Myanmar Healthcare Facilities Based on NCRP Report 147
AU  - Oo, Zaw Lin
AU  - Laı, Theint Win
PY  - 2026
DA  - July
Y2  - 2026
DO  - 10.59474/nuclear.2023.69
JF  - Journal of Nuclear Sciences
PB  - Ankara University
WT  - DergiPark
SN  - 2147-7736
VL  - 10
IS  - 2
LA  - en
AB  - This study aims to develop and validate a machine learning model for accurate and rapid prediction of required brick wall thickness for diagnostic X-ray rooms, based on the methodology of NCRP Report No. 147, and to assess its performance using real-world clinical data from Myanmar. A synthetic dataset of 864,000 samples was generated based on the shielding design methodology in NCRP Report No. 147. A Random Forest regressor was trained and hyperparameter-tuned, then benchmarked against baseline models including Linear Regression and Decision Trees. The final model was externally validated using real-world data from 11 diagnostic X-ray installations (4 polyclinics, 4 general hospitals, and 3 diagnostic centers) across Myanmar. The best-performing model was deployed as a free open-access web application on Hugging Face Spaces. The optimized Random Forest model achieved excellent performance on the synthetic test set (MAE = 0.056 cm, RMSE = 0.156 cm, R² = 0.9996), markedly outperforming the best baseline model (Decision Tree, MAE = 0.615 cm). Feature importance analysis showed that tube voltage (kVp) and brick density were the dominant predictors, contributing 53.2% and 37.8% of the total importance, respectively. External validation on independent real clinical data yielded a mean absolute error of 3.22 cm, which remains within conservative safety margins for shielding design. The model also indicated that conventional 9-inch (22.9 cm) brick walls are frequently inadequate for typical clinical workloads in Myanmar. This study demonstrates that a well-trained Random Forest model can provide highly accurate and rapid predictions of X-ray shielding requirements. Through successful validation with real clinical data from a low-resource setting and the provision of an open-source web application, the proposed approach offers a practical and accessible tool for radiation safety officers and facility designers in Myanmar and similar healthcare environments.
KW  - X-ray shielding
KW  - Radiation protection
KW  - Diagnostic radiology
KW  - NCRP Report 147
KW  - Machine learning.
N2  - This study aims to develop and validate a machine learning model for accurate and rapid prediction of required brick wall thickness for diagnostic X-ray rooms, based on the methodology of NCRP Report No. 147, and to assess its performance using real-world clinical data from Myanmar. A synthetic dataset of 864,000 samples was generated based on the shielding design methodology in NCRP Report No. 147. A Random Forest regressor was trained and hyperparameter-tuned, then benchmarked against baseline models including Linear Regression and Decision Trees. The final model was externally validated using real-world data from 11 diagnostic X-ray installations (4 polyclinics, 4 general hospitals, and 3 diagnostic centers) across Myanmar. The best-performing model was deployed as a free open-access web application on Hugging Face Spaces. The optimized Random Forest model achieved excellent performance on the synthetic test set (MAE = 0.056 cm, RMSE = 0.156 cm, R² = 0.9996), markedly outperforming the best baseline model (Decision Tree, MAE = 0.615 cm). Feature importance analysis showed that tube voltage (kVp) and brick density were the dominant predictors, contributing 53.2% and 37.8% of the total importance, respectively. External validation on independent real clinical data yielded a mean absolute error of 3.22 cm, which remains within conservative safety margins for shielding design. The model also indicated that conventional 9-inch (22.9 cm) brick walls are frequently inadequate for typical clinical workloads in Myanmar. This study demonstrates that a well-trained Random Forest model can provide highly accurate and rapid predictions of X-ray shielding requirements. Through successful validation with real clinical data from a low-resource setting and the provision of an open-source web application, the proposed approach offers a practical and accessible tool for radiation safety officers and facility designers in Myanmar and similar healthcare environments.
CR  - [1] United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), Sources and effects of ionizing radiation, New York, United Nations, 2018.
CR  - [2] K. H. Ng and M. M. Rehani, X-ray imaging goes digital: benefits and risks, Biomedical Imaging and Intervention Journal, 4(2), e25, 2008.
CR  - [3] G. Sánchez-Barroso, M. Botejara-Antúnez, J. García-Sanz-Calcedo, and F. Zamora-Polo, A life cycle analysis of ionizing radiation shielding construction systems in healthcare buildings, Journal of Building Engineering, 41, 102387, 2021.
CR  - [4] N. Z. J. Jamaluddin, M. M. Bani-Ahmad, N. N. Z. Azman, and R. Ramli, Eggshell-enhanced composites: Innovative radiation shielding materials for diagnostic X-ray applications, Radiation Physics and Chemistry, 224, 112076, 2024.
CR  - [5] A. Dwiyanto, G. Hardiman, and W. Budi, Comparative study of the absorbed dose of secondary shield wall elements in a digital radiography room, International Journal of Scientific and Research Publications, 8(6), 352-358, 2018.
CR  - [6] A. S. Wibowo, E. Cahyono, R. S. Iswari, K. A. Amin, and M. Jannah, Comparison of the effectiveness of radiation shield wall between lead-layers and plastering brick-layers, Jurnal Riset Kesehatan, 11(2), 95-102, 2022.
CR  - [7] M. Haider, S. Shill, Q. Mohammad, R. Nizam, and M. Akramuzzaman, Shielding calculation based on NCRP methodologies for some diagnostic x-ray facilities in Bangladesh, International Conference on Physics for Sustainable Development, 2014.
CR  - [8] Department of Atomic Energy, Myanmar, Annual inspection reports of diagnostic X-ray facilities (Internal), Nay Pyi Taw: DAE, 2023.
CR  - [9] National Council on Radiation Protection and Measurements, Structural shielding design for medical X-ray imaging facilities, NCRP Report No. 147, Bethesda, MD: NCRP, 2004.
CR  - [10] B. R. Archer, Recent changes in medical X-ray imaging facility shielding design methodology, Medical Physics, 32(12), 3599-3601, 2005.
CR  - [11] D. J. Simpkin, Shielding requirements for constant-potential diagnostic X-ray tubes determined by Monte Carlo calculations, Health Physics, 108(2), 148-157, 2015.
CR  - [12] M. M. Rehani, Challenges and opportunities for radiation protection in developing countries, Radiation Protection Dosimetry, 165(1-4), 8-13, 2015.
CR  - [13] World Health Organization, Global atlas of medical devices, Geneva: WHO, 2017.
CR  - [14] Ministry of Health and Sports, Myanmar, Health in Myanmar 2022, Nay Pyi Taw: MOHS, 2022.
CR  - [15] L. Breiman, Random forests, Machine Learning, 45(1), 5-32, 2001.
CR  - [16] P. Probst, M. N. Wright, and A. L. Boulesteix, Hyperparameters and tuning strategies for random forest, WIREs Data Mining and Knowledge Discovery, 9(3), e1301, 2019.
CR  - [17] T. Chen and C. Guestrin, XGBoost: A scalable tree boosting system, Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785-794, 2016.
CR  - [18] H. Akyıldırım, F. Waheed, K. Günoğlu, and İ. Akkurt, Investigation of buildup factor in gamma-ray measurement, Acta Physica Polonica A, 132(3), 1203-1206, 2017.
CR  - [19] A. K. Patel, R. J. Smith, and D. L. Brown, A Python-based Monte Carlo tool for radiation shielding design in medical facilities, Nuclear Instruments and Methods in Physics Research A, 1012, 165623, 2021.
CR  - [20] L. Zhang, W. Chen, and H. Li, Concrete shielding design for medical accelerators: a comparative study of NCRP 151 and Monte Carlo approaches, Journal of Applied Clinical Medical Physics, 21(4), 112-120, 2020.
CR  - [21] B. R. Archer, J. I. Thornby, and S. C. Bushong, Diagnostic x-ray shielding design based on an empirical model of photon attenuation, Health Physics, 44(5), 507-517, 1983.
CR  - [22] F. Pedregosa, G. Varoquaux, A. Gramfort, et al., Scikit-learn: Machine learning in Python, Journal of Machine Learning Research, 12, 2825-2830, 2011.
CR  - [23] Ministry of Construction, Myanmar, National building code of Myanmar: radiation protection facilities, Yangon: MOC, 2015.
CR  - [24] Hugging Face, Hugging Face Spaces documentation, Available from: https://huggingface.co/docs/hub/spaces (Accessed date: 05.04.2026), 2026.
CR  - [25] J. D. Hunter, Matplotlib: A 2D graphics environment, Computing in Science & Engineering, 9(3), 90-95, 2007.
CR  - [26] W. McKinney, Data structures for statistical computing in Python, Proceedings of the 9th Python in Science Conference, 51-56, 2010.
CR  - [27] G. Van Rossum and F. L. Drake, Python 3 Reference Manual, Scotts Valley, CA: CreateSpace, 2009.
CR  - [28] C. R. Harris, K. J. Millman, S. J. van der Walt, et al., Array programming with NumPy, Nature, 585(7825), 357-362, 2020.
CR  - [29] P. Virtanen, R. Gommers, T. E. Oliphant, et al., SciPy 1.0: fundamental algorithms for scientific computing in Python, Nature Methods, 17(3), 261-272, 2020.
CR  - [30] IAEA, Design of radiotherapy facilities, IAEA Safety Standards Series No. SSG-9, Vienna: IAEA, 2019.
CR  - [31] ICRP, The 2007 recommendations of the International Commission on Radiological Protection, ICRP Publication 103, Oxford: Pergamon Press, 2007.
UR  - https://doi.org/10.59474/nuclear.2023.69
L1  - https://dergipark.org.tr/en/download/article-file/5985104
ER  -