Radiation Physics and Engineering

Abstract Radiation shielding is essential for minimizing exposure to harmful ionizing radiation by employing materials that effectively absorb or block radiation. This study investigates the shielding potential of five waste-derived materials -human hair, waste glass, plastic, waste cement, and medical gloves- against β and γ radiations emitted from Thallium-204, Cesium-137, Strontium-90, and Cobalt-60 sources. Circular disc samples (2 ± 0.2 mm thick, 20 ± 0.08 mm diameter, 0.8 ± 0.11 g) were prepared in pure and composite forms (0–100 wt%). Shielding properties were quantified using a GM counter and gamma spectrometry, while structural and functional characterizations were performed using X-ray diffraction (XRD) and Fourier-transform infrared spectroscopy (FTIR). Results show that waste cement exhibited the highest shielding efficiency with η = 27.61% and AF = 1.38 at 100 wt%, effectively reducing Sr-90 penetration below ~0.90 counts/s. Medical gloves showed moderate but consistent attenuation (η ≈ 24.32%, AF ≈ 1.32), whereas hair and plastic demonstrated weaker performance (η = 10.00–25.59%, AF = 1.12–1.34). Glass exhibited low shielding capacity due to its intrinsic radioactivity (η = 1.58%, AF = 1.02). FTIR and XRD analyses confirmed that inorganic groups such as Si–O, CO₃²⁻, and SO₄²⁻ in cement and glass enhance density and rigidity, improving photon attenuation, while polymeric and organic matrices offer limited protection. Transmission studies using Co-60 validated that none of the low-density materials achieved attenuation beyond 28%. Overall, waste cement demonstrates strong potential for sustainable, low-cost radiation shielding, which can be further enhanced through composite reinforcement with dense waste materials.

Abstract This feasibility study explores a computational method for estimating air-kerma calibration factors (N_K) of narrow-beam X-ray dosimeters at Secondary Standard Dosimetry Laboratories (SSDLs) in resource-limited settings. Using mass attenuation coefficient ratios interpolated from NIST data, analytical calculations yielded calibration factors for the 30-cc PTW spherical ionization chamber under the ISO 4037 N-60 (mean energy of 47.9 keV) and N-80 (mean energy of 65.2 keV) qualities, referenced to Co-60 and Cs-137. Deviations from IAEA-traceable values were recorded as 1.1% (for N-80) and 6.8% (for N-60) using the Co-60 as reference energy, and 3.0% (for N-80) and 8.9% (for N-60) using the Cs-137. Independent MCNP4C Monte Carlo simulations captured energy-dependent trends but showed larger discrepancies (8.6-13.4%). This method serves as a useful supplementary tool that supports standard experimental calibration (with expanded uncertainty, k=2: 1.8-3.5%). However, it has limitations due to the lack of new experimental measurements and reliance on computational assumptions, which require additional validation.

Abstract In this study, environmental gamma radiation dose rates are measured at Jamalpur 250 Beded General Hospital in Jamalpur district. Part of the Directorate General of Health Services (DGHS), the Jamalpur 250 Beded General Hospital is a public hospital located in Jamalpur, Bangladesh. The background radiation levels in this area must therefore be continuously monitored both indoors and outdoors. The measurement was performed using a portable Gamma-Scout detector. Total 35 measuring points were selected for collection of gamma-radiation in the outdoor and the indoor environment at different places at Jamalpur 250 Beded General Hospital. The measuring points were marked out using Global Positioning System (GPS) navigation. The measured outdoor mean dose rates ranged from 0.1136±0.0039 µSv.h^-1 to 0.1700±0.0056 µSv.h^-1 and the measured indoor mean dose rates ranged from 0.1150±0.0025 µSv.h^-1 to 0.2030±0.0031 µSv.h^-1. The annual effective dose rate in the outdoor environment ranges from 0.1991±0.0069 mSv to 0.2978±0.0098 mSv and the indoor environment, the value ranged from 0.8063±0.0176 mSv to 1.4226±0.0218 mSv. The outdoor Excess Lifetime Cancer Risk (ELCR) ranged from 0.0008 to 0.0012 and the indoor ELCR ranged from 0.0033 to 0.0059. However, the lowest ELCR is 0.0008 which is measured at outdoor and the highest ELCR is 0.0059 which is measured at indoor.

Abstract ‎This study focuses on the ability of both ZnO and Gd₂O₃ used as doping agents in glass in order to replace shielding for lead purposes‎. ‎The choices of metal-transition oxide and rare earth oxide doping aim to compare shielding results‎, ‎given their different properties‎. ‎A thermal study of two series of Zinc oxide and Gadolinium oxide in different percentages is performed‎. ‎The radiation attenuation properties of these doped glasses were developed‎. ‎Experimental measurements were performed using a photon spectrometer‎. ‎Photons incident from Co-60 and Cs-137 sources with main energies of 662‎, ‎1173‎, ‎and 1332 keV were applied to the shield‎. ‎This study demonstrated that adding either Gadolinium oxide or Zinc oxide to medical glass enhances the radiation attenuation ability‎, ‎with a preference given to Gadolinium Oxide glass doping‎. ‎In terms of radiation attenuation‎, ‎the results obtained with Gadolinium Oxide are comparable to those obtained with lead oxide-doped glass‎. ‎In addition to the attenuation factors‎, ‎the thermal properties of each of the mentioned samples were studied under temperature changes‎.

Abstract Monte Carlo (MC) methods are considered a complementary method for dose verification in radiation therapy‎. ‎This study aims to simulate the Artiste head and the Siemens 160 Multileaf Collimator (MLC) using MCNPX 2.6.0 to enhance dose verification accuracy in Intensity-Modulated Radiation Therapy (IMRT) treatment plans‎. The MC-based calculations were benchmarked against commissioning-measured data and an MLC test field‎. ‎A comparison between MC-based and treatment planning system (TPS)-based dose maps was made for beams of a typical complicated IMRT plan‎. ‎The results demonstrated a 3D gamma passing rate (GPR) of 97.1% with 3%/3mm criteria and a 10% dose threshold‎, ‎indicating the accuracy of the MC model‎. Based on the acceptable GPRs‎, ‎the provided model has sufficient accuracy‎. ‎It has been confirmed that the MC calculations can be carried out within a reasonable computation time‎, ‎taking approximately 10 minutes per beam and less than 2 hours for a typical 9-beam IMRT plan‎. ‎This is possible with a specific‎, ‎powerful CPU configuration used for MC verification of such a complicated IMRT plan‎.

Abstract To counter the growing threat of illicit radioactive material trafficking, we developed the Radioactive Detection System (RDS), a scalable, low-cost sensor network for real-time radioisotope identification and localization. Each node combines inexpensive detectors (NaI(Tl) scintillation spectrometers or plastic scintillators) with existing surveillance cameras. Machine-vision algorithms fuse radiation measurements with visual tracking to simultaneously quantify intensity and precisely locate moving sources in complex and dynamic urban settings. Laboratory and field tests using concealed sources carried at pedestrian speeds (0.5-1.2 m.s^-1) showed: (i) detection and visual localization of a 100 µCi Cs-137 source in 8-25 s; (ii) reliable spectroscopic identification of the Cs-137 within approximately 100 seconds under the reported laboratory testbed configuration. The multi-modal design substantially enhances sensitivity, specificity, and robustness against background fluctuations. With low per-node cost and straightforward integration into existing infrastructure, RDS enables practical large-scale deployment for continuous monitoring of public spaces, critical infrastructure, border checkpoints, and high-security areas, providing an effective tool for radiological threat mitigation.

Abstract Dual-purpose cask (DPC) is used for transporting and interim storage of spent fuel assemblies, since such casks are an attractive option due to their flexibility and economy efficiency. In point of economical view, construction of high-capacity DPC is more suitable. The cask could be used to transport the spent fuel assemblies of long cooling time with its full capacity. At emergency situations the same cask potentially could be used to transport the spent fuels with short cooling time using some modifications inside the canister. The present study would investigate the highest capacity of Tehran Research Reactor (TRR) DPC for transport and storage of 2-year cooled spent fuel assemblies. MCNPX2.7.0 computational code was used to calculate the DPC body maximum gamma and neutron dose rates. The obtained results showed that maximum six 2-year cooled SFAs of TRR could be transported by the present DPC. Moreover, filling the empty places of canister with carbon-steel shield blocks, some modifications on the cask door and floor is needed to pass the determined gamma dose rate limit (2 mSv.h^-1). By adding 6 cm thick carbon-steel on the DPC door and its bottom the goal is obtained while the modifications increase the cask weight about 1 ton.

Abstract Accurate segmentation of organs at risk (OARs) in head and neck CT scans is essential for effective radiotherapy planning. Although U-Net based deep learning models have achieved strong performance, small or anatomically complex OARs remain challenging to segment accurately. This study investigates how image slicing (cropping) influences segmentation accuracy and efficiency when using a Residual U-Net for head and neck OAR segmentation. A total of 63 CT scans from a public dataset and an institutional dataset were used. Slice-based preprocessing was applied by cropping regions surrounding target masks. Residual U-Net models were trained and tested using both full-size and sliced (cropped) images for 41 OARs. Segmentation accuracy was evaluated using Dice Similarity Coefficient (DSC) and Intersection over Union (IoU). Additional experiments incorporating dropout layers and longer training epochs were performed to improve optic chiasm and optic nerves segmentation. Slice-based networks achieved a 4.1% increase in IoU and a 3.2% increase in Dice score compared with full-size networks. For 11 complex structures, IoU and Dice scores improved by 10.9% and 9.0%, respectively. The standard deviation of both metrics decreased, indicating more consistent performance. Slice-based networks also demonstrated 130% faster training and about 8× faster prediction times. Adding dropout layers and extending training epochs further improved segmentation of the optic chiasm and optic nerves. Slice-based preprocessing combined with a Residual U-Net architecture improves segmentation accuracy and computational efficiency in head and neck CT imaging. This approach shows strong potential for practical use in radiotherapy planning.