EGSnrc Monte Carlo school for Medical Physics

Monte Carlo techniques are the most rigorous tools for modelling radiation transport in medical physics. The widespread availability of general-purpose software (EGSnrc, Penelope, Geant4) and increasing computational capabilities have led to their extensive adoption in recent years. As a fundamental calculation tool, the Monte Carlo method will continue to drive accurate dosimetry and innovative research in radiation therapy, medical imaging, and radiation protection.

This school provides participants with comprehensive training in Monte Carlo methods for medical applications in therapy and diagnostics, including hands-on simulation laboratories using the EGSnrc toolkit. The target audience is students and professionals with an interest in Monte Carlo Simulation applied to medical physics and ionizing radiation.

Previous events:

• 2024 ICTP/IAEA workshop, Trieste, Italy
• 2017 ICTP/IAEA workshop, Trieste, Italy
• 2014 I Brazilian Congress of Ionizing Radiation Metrology, Rio de Janeiro, Brasil (no link)
• 2011 ICTP/IAEA workshop, Trieste, Italy

• Matheus Rebello do Nascimento (National Institute of Metrology, Quality and Technology (INMETRO) and Universidade do Estado do Rio de Janeiro (UERJ), Brazil)
• Eric Matos Macedo (Instituto Federal de Educação, Ciência e Tecnologia da Bahia (IFBA), Brazil)
• Frédéric Tessier (National Research Council of Canada (NRC), Canada)

 

It is highly recommended that participants in this event prepare poster presentations.

Participants that have accepted abstracts will be able to submit a complete paper to a special issue of the Brazilian Journal of Radiation Sciences at the end of the school.

 

Announcement:

Application is now closed

Participants list here

Lecturers

• Ernesto Mainegra-Hing ( National Research Council of Canada, Canada)
• Frédéric Tessier (National Research Council of Canada, Canada)
• Reid Townson (National Research Council of Canada, Canada)

 

Announcement:

Application is now closed

 

• Aman, Enzo Hilario (Departamento de Física, Facultad de Ciencias Exactas, UNLP, Argentina): SIMIND Montecarlo and thyroid digital twins for assesing the accuracy of dosimetric calculations in real clinical settings.

Enzo Hilario Aman (1) , Cecilia Yamil Chain (2, 3) and Luis Héctor Illanes (1). 1- Departamento de Física, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, La Plata, Argentina 2- Departamento de Ciencias Biológicas, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, La Plata, Argentina 3- Instituto de Investigaciones Fisicoquímicas Teóricas y Aplicadas (INIFTA), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Universidad Nacional de La Plata, La Plata, Argentina Introduction: Digital twins are emerging tools for simulating and optimizing therapeutic protocols in personalized nuclear medicine (NM). In this study, we construct patient-specific thyroid digital twins (TDT) aimed at assesing the accuracy of dosimetric calculations in hyperthyroidism 131 I- radiopharmaceutical therapy in a real clinical setting. Materials & Methods: The methodology integrates two components: (i) anatomical TDTs of big and medium goitre, generated by processing patient CT scans with 3D Slicer Image computing software; and (ii) a virtual NM image acquisition using SIMIND Montecarlo code. SIMIND MC emulating a gamma camera Picker Prism 2000 XP is used to obtain planar images of TDTs containing 3.7 MBq of 131 I, using a high- energy collimator and a distance thyroid/ detector of 15 cm, as real clinical setting. The quantification of 131 I activity in the simulated images is performed using Fiji and dosimetric calculations are based on a monocomportamental 131 I biodistribution model, assuming a post- administration time of 24 hours, and an effective elimination time of 5.5 days. The accuracy in the dose calculation is evaluated in terms of the relative standard deviation (% RSD) from the D value that emerges from considering 3.7 MBq as the “true” activity in the gland. Results: The anatomical TDTs of goitre have been successfully implemented, yielding with SIMIND MC realistic, patient-morphed thyroid scintigraphies. Activity quantification and dose calculation are in course. Conclusion: This study illustrates the utility of SIMIND MC simulations together with TDTs to trully ponderate the accuracy of a dosimetric calculation that could emerge in a realistic clinical setting. It is foreseen that the knowledge of the expected maximun accuracy in the dose calculation, even when it is not optimal, will persuade physicians to planificate treatments based on dosimetric approaches, enabling safer and more effective personalized radiopharmaceutical therapies.

• Apaza Blanco, Oscar Abel (Facultad de Matemática Astronomía y Física, Universidad Nacional de Córdoba, Bolivia): Commissioning PRIMO (PENELOPE) for Patient-Specific QA of 3D-CRT and IMRT on a Varian Clinac iX.

Purpose. To implement PRIMO (PENELOPE) as an independent Monte Carlo (MC) dose verification tool for patient-specific QA of 3D-CRT and IMRT on a Varian Clinac iX at the Instituto Oncológico del Oriente Boliviano. The poster describes the commissioning protocol and early experience planned for the next 2–3 months. Methods (planned). A 6 MV beam model for the Clinac iX will be built and tuned to match Golden Beam Data and TPS reference data—PDDs and lateral profiles in water. Commissioning will follow IAEA-TECDOC-1583 and AAPM TG-119 style test fields. MC transport cutoffs, voxel size, variance-reduction options, and statistical-uncertainty targets will be optimized for accuracy versus runtime. Validation will compare MC doses with the clinical TPS and measurements for standard tests and initial patient plans. Endpoints include point-dose differences, 2D/3D gamma pass rates (3%/2 mm), and computation times. Expected outcome. We aim to deliver a practical, shareable protocol checklists, parameter sets, and automation scripts demonstrating the feasibility of PRIMO for independent QA on Clinac iX. Preliminary results will be included.

• Bonatto, Alexandre (Federal University of Health Sciences of Porto Alegre (UFCSPA), Brazil): Bayesian optimization of a laser-plasma accelerator aiming the production of high-energy electron beams for VHEE radiotherapy

Radiotherapy is one of the main treatment modalities for malignant tumors, but dose-limiting toxicity to normal tissues remains a significant obstacle. Proton therapy addresses this issue through superior dose localization, but its high infrastructure cost limits widespread adoption. Very high energy electrons (VHEE), with energies from 50 to 250 MeV, offer deep tissue penetration and are particularly suited for FLASH radiotherapy, which can reduce toxicity to normal tissues while delivering higher curative doses. Laser-plasma accelerators are being investigated as a compact, lower-cost alternative to conventional accelerators for producing VHEE beams. In this work, we use Bayesian optimization to tune a laser-plasma accelerator for VHEE radiotherapy applications. The optimization employs an objective function designed to favor electron beams peaking at selected energies—100, 200, and 300 MeV—with a separate optimization sequence for each target energy. The goal is to produce quasi-monoenergetic spectra suitable for VHEE radiotherapy at different penetration depths, obtained by setting the laser-plasma accelerator input parameters to the optimal values identified in each optimization sequence.

• Campos, Luciana Tourinho (Universidade do Estado do Rio de Janeiro, Brazil): Estimation of computed tomography dose index with Monte Carlo using EGSnrc

Abstract: Computed tomography is a method that extends the clinical capabilities of X-ray imaging. Dosimetry in CT is increasingly based on Monte Carlo studies that define the dose in the patient as a function of air kerma at the isocenter and periphery. To understand and quantify the risk associated with examinations, efforts have been made to determine the radiation dose more accurately to individual radiosensitive organs, which are important quantities to calculate the effective dose and cancer risk. Although dosimetric studies are classically developed through experimental tests, CT dosimetry is increasingly based on Monte Carlo studies. This work aimed to estimate computed tomography dose index values using the EGSnrc code and compare them with experimentally obtained results, to demonstrate the feasibility of CT dosimetry with the EGSnrc code. The dosimetric setup was modeled with the EGSnrc geometry package, based on instruments and materials used in the experimental dosimetry phase. The results of this study showed the capability of the EGSnrc simulation code in extracting the air kerma of CT machines. The simulated CTDI results for the x-ray output of 120 kVp were obtained for the central and peripheral positions. The agreement for the CTDI100 head and body phantom between measured and simulation was less than 2.5% and this comparison shows an agreement between measured and simulation CTDIw of less than 2.5%. Then, the application egs_kerma was used to estimate the CTDI100 values in the central and peripheral holes of two standard PMMA phantoms. The discrepancies between the measured and simulated CTDI100 values for the two phantoms were up to 2.65%. Comparison of simulated and measured values of CTDIw showed a good agreement, less than 2.5 % for the head and body phantom. The good agreement between the measured and calculated results indicates that the X-ray characteristics of the CT scanner were obtained accurately, which directly affects the absorbed dose to the phantom. The results of this study showed the capability of EGSnrc and the user code egs_kerma in extracting the real fluence, primary, and scattered spectra used. This study may be useful to establish a patient-specific dosimetry for CT systems in the future. Keywords: Computed tomography, CTDI, dosimetry, Monte Carlo, EGSnrc

• Cárdenas Urrego, Laura Sofía (Universidad Distrital Francisco José de Caldas, Colombia): Monte Carlo simulation of radiation–matter interaction: state of the art and simulation codes

Monte Carlo simulation methods are widely used to model radiation-matter interaction due to their ability to describe complex stochastic processes with high precision. This paper presents a review of the state of the art of the main Monte Carlo codes used in radiation transport simulation, analyzing their general characteristics, applications, and limitations. The review seeks to provide an updated overview that facilitates the understanding and selection of these tools in different scientific and technological contexts.

• Cortez, Enzo (Centro de Desenvolvimento da Tecnologia Nuclear, Brazil): Estimating the dose rate constant for Iodine-131 via Monte Carlo simulation: A first step towards pediatric radioiodine therapy shielding evaluation

Accurate shielding design in nuclear medicine facilities is essential for radiation safety. Recent evidence demonstrates that patient self-absorption reduces ambient radiation fields, directly impacting barrier requirements. Particularly, to our best estimate, there are no results regarding pediatric body attenuation, where patient anatomy changes the photon spectrum emitted by the incorporated source. For I-131, a widely used therapeutic radioisotope, there is no agreement on the dose rate constant for ambient dose equivalent estimation (ΓH*(10)). Several values are reported in the literature, leading to potential shielding overdesign or under protection. Therefore, the first step to estimating body attenuation for pediatric patients under radioiodine therapy is establishing a consensus ΓH*(10) for I-131. This study aims to estimate the ΓH*(10) for I-131-point sources through Monte Carlo (MC) simulations. Two different photon spectrum data (ICRP 107 and RADAR database) were studied. Two modelling scenarios were considered. First, a point source following the geometry from Paixão and Fonseca (2023 and 2024) was modeled, with monoenergetic photon emissions and air kerma scoring. ICRP 74 coefficients (Sv/Gy) were interpolated for the ΓH*(10) calculation in this case. Additionally, the same scenario was used to score photon fluence, and ICRP 74 coefficients (Sv cm²) were selected appropriately for the ΓH*(10) calculation. Secondly, the ICRU sphere setup, which included a parallel source, was modeled to obtain the dose rate constant. Simulations were performed with up to 1×1010 particle histories on a 35-core computational cluster. The results demonstrated strong agreement between different photon spectra and methodology. In the first scenario and kerma scoring, the results were 6.41×10-2 and 6.40×10-2 mSv·m²/GBq·h for ICRP 107 and RADAR photon database, respectively. For fluence scoring, the results were 6.35 mSv·m²/GBq·h for RADAR database spectrum and 6.36 ×10-2 mSv·m²/GBq·h for ICRP 107 spectrum. The ICRU sphere simulation scenario result was 6.4×10-2 mSv·m²/GBq·h for both photon spectra. The uncertainty of all MC results is less than or equal to 3%. Comparative analysis of our results to nine literature values showed differences ranging from -17% to 23%. Our results are close to the DIN 6844-3 recommended value (maximum -4%) and the average literature value (maximum -1%), confirming methodological consistency. This validation supports the reliability of the Monte Carlo simulations of this work. Future work will focus on patient body attenuation, modeling a pediatric voxel phantom with a biodistributed I-131 source.

• Da Silva, Ellen Vilarinho (Rio de Janeiro State University, Brazil): Application of the egs_kerma code for Estimating Air Kerma and Mean Glandular Dose in Digital Mammography

The breast is an organ sensitive to radiation, and optimizing imaging procedures based on ionizing radiation is essential to reduce risks. This study evaluated the applicability of the egs_kerma code, from the EGSnrc system, to estimate the air kerma and the mean glandular dose (MGD) in digital mammography. Different anode/filter combinations (Mo/Mo, Rh/Rh, and W/Rh) and breast compositions (75%, 50%, and 25% glandular tissue) were simulated. The air kerma was determined using egs_kerma, and the MGD values were calculated from conversion factors. The results showed that lower mean energy beams, such as Mo/Mo, produced higher doses, while more penetrating beams, such as W/Rh, resulted in lower doses consistent with the expected physical principles. Differences observed in high-glandularity combinations were attributed to the geometric simplifications of the model and the idealized beam conditions. The obtained values showed good agreement with reference data, with percentage differences generally below 2%. It is concluded that egs_kerma is a robust, accurate, and efficient tool for estimating air kerma and MGD in mammography, proving useful for quality control and optimization of clinical protocols. Future studies may include more realistic geometries and spectral adjustments to increase result accuracy and expand the method’s applications in medical dosimetry.

• Da Silva, Gustavo Freire Pereira (Institute of Physics of University of São Paulo, Brazil): Feasibility of a Bench-Top PCCT Model for Electronic Density and Effective Atomic Number Measurements: A Monte Carlo Simulation Study

The goal of this study was to evaluate the feasibility of a bench-top PCCT using Geant4 Monte Carlo simulations. The study applied Geant4 to determine the electronic density and the effective atomic number of materials as a function of CT number at two energy ranges. A focal spot-to-image detector distance of 1045 mm was adopted. A 50 kV X-ray spectrum measured by a CdTe spectrometer was incorporated. The virtual photon-counting detector consisted of a 128 × 128 matrix, and an ideal efficiency was modeled. Low- and high-energy acquisition ranges with an energy threshold of 30 keV were adopted, and CT images were reconstructed using SIRT (Simultaneous Iterative Reconstruction Technique). Two virtual cylindrical phantoms were designed: a calibration phantom with internal cylindrical inserts of polypropylene, nylon, silicon dioxide, polycarbonate, and Teflon; and a test phantom with paraffin and polyethylene inserts, with all materials immersed in water. The axes of these phantoms were positioned at 225 mm from the focal spot. ROIs were selected, and the CT numbers were measured for the reconstructed images from the low- and high-energy ranges. A dual-energy subtracted quantity was estimated as DHU, and the reduced CT number, linear with the CT number for the low-energy image, was calculated. A linear regression between the electronic densities and DHU, and between the effective atomic number and the reduced CT number, was fitted based on the calibration phantom and used to estimate these quantities for the test phantom materials. Using the test phantom, electronic densities and effective atomic numbers were estimated with accuracies better than 8.5% and 6.6%, respectively. Based on MC simulations, the proposed experimental model was demonstrated to be a viable alternative for characterizing materials based on their electronic densities.

• De Souza Junior, José Elizeu (Universidade Federal de Minas Gerais, Brazil): Influence of Photon Emission Spectra on Air Kerma Rate Constants for Radionuclides: A Monte Carlo EGSnrc Study

The air kerma rate constant is a critical parameter of utmost importance in several applications, including instrument calibration, radiation shielding design, and absorbed dose estimation in brachytherapy (Stabin, 2008).Traditionally, these values are determined using deterministic methods ( Wasserman and Groenewald (1988) and Ninkovic and Adrovic (2005)). The feasibility of estimating air kerma rate constants through Monte Carlo simulations via EGSnrc code was recently demonstrated ( Paixao and Fonseca (2024)). The influence of the photon spectrum database on the results of MC simulations was around 1%. Overall, the DECDATA photon spectrum database provided more detailed spectra and results for a better understanding of the discrepancies observed when comparing with other databases, such as RADARDecay. The results demonstrate good agreement with reference data. The results of this study will contribute to evaluating the impact of photon spectra on the air Kerma rate constant.

• Farias, Whoody Alem (UFPE, Brazil): A platform for creating computational exposure models: personalized dosimetry with Holmium-166

Blender is an open-source 3D environment widely used for Boundary Representation (BREP) phantom modeling. The PHAntoms MAnufacturing (PHAMA) add-on was developed to extend Blender with specific tools for creating, editing, and exporting BREP phantoms, as well as for generating radiation source algorithms. Integrated with other add-ons and the in-house Digital Imaging Processing (DIP) and Monte Carlo software, PHAMA optimizes workflows for computational dosimetry in EGSnrc and other Monte Carlo codes compatible with BREP or voxel phantoms. PHAMA’s core functionalities include: (i) modeling intestines in polygonal meshes; (ii) segmentation of bone tissues for skeletal dosimetry; (iii) detection and closure of gaps that may prevent phantom voxelization; (iv) identification of intersections that may make tetrahedralization of phantoms unfeasible; (v) automation for geometric and topological adjustments; (vi) calculation of object volumes and estimation of the number of voxels; and (vii) export of phantoms with axis mapping suitable for simulations in EGSnrc. The aim of this study is to demonstrate a complete workflow for personalized dosimetry in radioembolization: from the construction of a BREP phantom in Blender from DICOM computed tomography images, through the generation of a Holmium-166 source, voxelization and coupling to EGSnrc for simulations, to the determination of the resulting personalized S-values.

• Franco, Flavia (Universidade de São Paulo, Brazil): Study and Development of an Experimental Setup for Spatially Fractionated Radiation Therapy Using Monte Carlo Simulation

Unlike conventional radiotherapy, which aims for homogeneous dose delivery, Spatially Fractionated Radiation Therapy (SFRT), such as GRID therapy, proposes a heterogeneous dose distribution within the tumor volume. This technique holds promise for treating bulky tumors, which often exhibit high radioresistance and hypoxic areas that hinder cell death by apoptosis. By utilizing narrow beams, GRID fosters radiation-induced effects, such as the Bystander and Abscopal effects, potentially contributing to optimized therapeutic responses. This work aimed to develop and implement a GRID irradiation system for in vitro and small animal studies, as well as to analyze the generated dose distributions. Monte Carlo simulations using the PENELOPE v. 2008 package were employed alongside an experimental approach to characterize aluminum and brass GRID collimators. The collimators, featuring a 5 cm diameter and 3.5 cm thickness, were simulated for beam energies ranging from 50 to 120 kVp, matching the characteristics of the Isovolt Titan E X-ray tube. In the experimental phase, the collimators were manufactured and coupled to the X-ray unit. EBT3 radiochromic films were used for dosimetric characterization, allowing for the assessment of experimental dose distributions, GRID profiles, and Valley-to-Peak Dose Ratios (VPDR). Furthermore, simulations enabled the acquisition of Percentage Depth Dose (PDD) curves and the evaluation of other materials. For Aluminum and Brass GRID collimators with a 1:1 hole-to-shielding ratio and 3 mm apertures, the experimental VPDR values were 0.164 and 0.146, respectively. Cerrobend and Lead collimators showed similar VPDRs. Both experimental and simulated results demonstrated better VPDRs compared to collimators made of lower atomic number materials. The study concluded that brass collimators present superior dosimetric performance for the GRID system. Dosimetry with radiochromic films proved effective, and Monte Carlo simulations successfully expanded the experimental analysis, indicating that collimator material and geometry directly influence the quality of GRID dose distribution.

• Freitas, Juliana Campos De (Universidade Estadual de Maringá – UEM, Brazil): Generation and Validation of Phase Spaces in LINACs via Monte Carlo for Clinical Use

This project aims to develop and validate a methodology for modeling radiotherapy LINAC beams using Monte Carlo simulations with GATE/Geant4. We focus on generating accurate phase-space files that describe the energy, position, and direction of particles as they travel through the accelerator head. The simulation includes detailed modeling of key components such as the target, flattening filter, and collimation systems to ensure realistic dose profiles. In collaboration with RT Medical Systems, the generated phase spaces will be compared with reference curves and experimental measurements using metrics such as the gamma index. Once validated, these files can support independent dose-verification tools, helping improve quality assurance in radiotherapy while reducing the need for extensive experimental tests. This work has practical potential for clinics, hospitals, and software developers in the radiotherapy field. By the end of the project, we expect to deliver a reliable methodology that could even be integrated into Treatment Planning Systems. The initiative also includes plans for intellectual property protection and offers valuable hands-on experience in technological development applied to medical physics.

• Giaccone Thomaz, Sofia (Universidade Estatual de Campinas, Brazil): Glandular Dose Map in Voxelized Phantoms Across Mammography and Contrast-enhanced Mammography obtained from Monte Carlo Simulations

Monte Carlo simulations are increasingly used in breast dosimetry due to their precision in estimating difficult-to-measure quantities, such as glandular dose. In breast imaging involving ionizing radiation, dosimetry studies are crucial for risk assessment; however, they typically provide information only on the average glandular dose (Dg), without accounting for dose hot spots. For this reason, this work explores dose mapping in the breast, a growing field aimed at improving the understanding of spatial dose distribution. This study focuses on mammography and contrast-enhanced mammography, and on how these modalities affect dose mapping depending on breast parameters such as phantom model, thickness, and glandularity, as well as imaging parameters such as the X-ray spectrum. The study was carried out using the MCGPU code and five different breast models: homogeneous, two different Gaussian models, a Fedon sampling model, and an anthropomorphic model.

• Larraga Gutierrez, Jose Manuel (Instituto Nacional de Neurologia y Neurocirugia, Mexico): Virtual source modeling of a medical linear accelerator using egs++ for detector correction factor calculations in very small and elongated photon fields

Accurate commissioning of treatment planning systems (TPS) for very small and elongated photon fields is particularly challenging, as standard beam models and clinical reference datasets are often insufficient. Furthermore, using vendor-supplied phase-space files (e.g., Varian phase spaces) for Monte Carlo benchmarking is limited by latent variance and restricted phase-space statistics, which may affect the accuracy of calculated output factors and dose distributions in small fields. I will introduce a virtual source modeling (VSM) approach for a medical linear accelerator implemented in egs++ (EGSnrc). The VSM models the linac head with three components: a primary photon source with an adjustable focal spot (circular or elliptical), an extrafocal photon source that accounts for head scatter, and an explicit electron-contamination source. These components are combined in an egs++ source collection with adjustable weights, enabling calibration against measured depth-dose curves and dose profiles for the clinical beam under commissioning. This project aims to develop a photon source that allows the calculation of detector correction factors for very small and elongated fields, overcoming the limitation of a finite number of photons in the available vendor-supplied phase-space files.

• Malagon Palomino, Angela Sofia (Universidad Distrital Francisco Jose de Caldas, Colombia): Comparative Analysis of Photon Attenuation Coefficients in Multilayer Shielding using Geant4 and MULASSIS

The interaction of ionizing radiation with matter is a critical factor in the design of biological shielding. This work develops a radiation-matter interaction simulation using the Monte Carlo method, specifically employing the Geant4 toolkit to model and calculate the attenuation coefficients of a photon beam within the 10^{-1} to 10^{1} MeV energy range for two multilayer material configurations: lead-concrete and polyethylene-aluminum. Rather than recording energy deposition, the implementation focuses on particle fluence by utilizing scoring planes positioned at the material interfaces. This approach enables the determination of the transmission factor and the Half-Value Layer (HVL) for each configuration. Data processing and statistical analysis are performed in Python using the Pandas library. Finally, the results are benchmarked against the European Space Agency’s (ESA) MULASSIS tool, allowing for a comparison of the attenuation coefficients obtained in composite material configurations for photons across this energy spectrum.

• Martínez Salazar, Itzel Amayrani (Universidad de Guanajuato, Mexico): Dosimetric Comparison of a Co-60 HDR Source in Water Using MCNP5 Simulations and Experimental Measurements

Brachytherapy is a cancer treatment modality that involves placing radioactive sources directly in or near the affected area. The radionuclides most commonly used in modern brachytherapy include Ir-192, Co-60, I-125, and Pd-103. In recent years, the use of Co-60 in high-dose-rate (HDR) brachytherapy has increased due to its long half-life (5.27 years) and its availability in miniaturized configurations comparable to HDR Ir-192 sources. The objective of this study is to model the radiation field in water for the Co-60 source model Co0.A86 (E&Z BEBIG), compute the dose-rate distribution around the source according to the TG-43 formalism, and compare these results with experimental measurements. Monte Carlo simulations were performed using MCNP version 5, a particle-transport code developed at Los Alamos National Laboratory. The detailed geometry and material composition of the Co0.A86 source were modeled in a water medium to obtain the dosimetric parameters defined in TG-43. For the experimental measurements, an A1SL ionization chamber was positioned inside a water phantom with a parallelepiped geometry (60.4 cm × 26.9 cm × 27.9 cm), enabling conditions comparable to clinical practice. This work presents the dosimetric simulation results in water for the Co-60 source model Co0.A86, as well as the direct comparison between the Monte Carlo dose-rate distribution and the experimentally measured values. In clinical brachytherapy, accurate dose-rate data for HDR sources are essential. The results obtained in this study help validate and complement the theoretical calculations derived from Monte Carlo simulations with MCNP5, thus contributing to the improvement of quality-assurance procedures and the strengthening of dosimetric verification in HDR brachytherapy.

• Muka, Paul (University of Sao Paulo, Brazil): MicroPET Performance Estimation Protocol Preclinical by Computer Simulation

This paper describes and presents the results of simulations of three hypothetical systems for tomographic imaging of positron emission radioactive emitters, based on the PETsys evaluation kit electronics, scintillation crystals, and silicon photomultipliers acquired as part of a FAPESP project to support the construction of the first preclinical microPET system in Brazil. The systems were modeled with detectors that matched the existing instrumentation: one system with four rotatable detectors, and two with 12 and 16 fixed detectors covering the entire field of view. These results will serve as a basis for estimating the expected results of systems with fixed detectors arranged circularly or forming a square with a field of view up to approximately 100 mm x 100 mm x 26 mm. The simulations use GATE 9.2 and considered crystal blurring effects in pulse detection. The analyses were performed using Root and Python programming codes. The results are consistent with expectations and demonstrate that the idealized virtual systems will serve as a reference for estimating the expected resolutions of systems with a larger number of detectors. Energy spectral of the Preclinical PET system was also simulated which shown 511 kev with the highest photopeak was observed

• Nunes, Bruno Silveira (IPEN, Brazil): Photon Production Efficiency in Bremsstrahlung Converters Aiming Mo-99 Production

High-Z materials are widely employed as bremsstrahlung converters in electron–photon sources, with tantalum being one of the most commonly used. In this study, Monte Carlo simulations using openTOPAS were performed to compare the photon production efficiency of converters made of Mo, Ta, W, Ir, Pt, Au, Hg, and Pb irradiated by a 40 MeV electron beam with a Gaussian energy spread. Photon yields were evaluated as a function of converter thickness and within selected energy ranges. The results indicate that Ta, W, Ir, Pt, and Au achieve similar maximum photon production, whereas Mo requires approximately twice the thickness to reach comparable yields and exhibits lower efficiency in the 10–20 MeV energy range. Despite their higher atomic numbers, Hg and Pb show reduced performance, attributed to electronic shell effects. Overall, these findings support the widespread use of tantalum, which offers a balance between conversion efficiency, practical material properties, and availability. This study contributes to the design of an experimental setup for the production of molybdenum-99 via photonuclear reactions, supporting the development of accelerator-based routes for radiopharmaceutical production.

• Paixão Reis, Lucas (Universidade Federal de Minas Gerais, Brazil): Students’ perceptions of Monte Carlo simulations as a learning tool in radiation physics

Introduction: Monte Carlo computational simulations have been extensively applied in various fields of radiation and medical physics, including radiology, radiotherapy, nuclear medicine, and radiation protection, for nearly seventy years. Implementing a traditional radiation physics laboratory involves challenges such as shielding, radiation source safety, and high costs. Therefore, computational laboratories represent a safer and more accessible alternative. In 2022, the book Computational Experiments in Radiation Physics was published, and the first class of the Computational Laboratory of Radiation Physics course began in 2023 at the Federal University of Minas Gerais School of Medicine, Brazil. Methods: Students enrolled in the course between March 2023 and July 2025, and other students who used the book in academic activities were invited to participate. A total of 38 invitations were distributed by email. Participation was voluntary and anonymous. The data collection instrument consisted of a structured online questionnaire with statements evaluated on a five-point Likert scale (1 – strongly disagree to 5 – strongly agree). Descriptive analyses were performed with a 95% confidence level. Results: Twenty-nine valid responses were obtained. Students expressed strong agreement that Monte Carlo simulations facilitated understanding of radiation–matter interactions, visualization of phenomena, and interest in radiation physics. Ten students provided comments suggesting increased course workload, improved clinical contextualization, a more detailed introduction to the Monte Carlo method and code, and better performance of virtual machines. Conclusion: The results indicate that Monte Carlo simulations are an effective tool for teaching radiation physics safely and engagingly.

• Ramos Borba, Tiago (Instituo Federal da Bahia, Brazil): Simulations of the VMAT Radiotherapy Technique Using TOPAS MC: Methodology and Perspectives for Clinical Applications

Volumetric Modulated Arc Therapy (VMAT) enables high dose conformity, reducing toxicity in healthy tissues. However, its complexity demands rigorous validation, particularly regarding beam modulation (MLC leaf motion, gantry rotation, and variable dose rate at each control point). Monte Carlo (MC) simulations are considered the gold standard for physical modeling. TOPAS MC, a Geant4-based platform, provides tools to simulate complex geometries and dynamic motion with high precision. This study aims to validate a virtual linear accelerator model for VMAT, highlighting replicable methodologies for Brazilian clinical scenarios.

• Scatena, Rafael Oddone (Universidade de São Paulo, Brazil): Modulo de Geometria de Malha Poligonal para o software de simulação Monte Carlo PENELOPE 2014

Monte Carlo (MC) simulation is considered the most accurate method for dose calculation in external beam radiotherapy, capable of reproducing complex physical phenomena such as electron backscattering in bone, air cavity perturbations, and detailed fluence distributions. However, the extremely high computational cost limits its clinical use, confining MC simulations mainly to verification or research purposes. To overcome this limitation, the PRISMATIC (PaRtIcle-matter SiMulATIon in Cuda) code was developed as a hybrid CPU/GPU Monte Carlo framework based on PENELOPE 2014, offloading the geometric transport routines to the GPU for massive parallelization. A second limitation of conventional Monte Carlo codes lies in their Constructive Solid Geometry (CSG) representation, which cannot accurately describe human anatomy. This work introduces a polygonal-mesh geometry module fully integrated into PRISMATIC, replacing the original quadratic Constructive Solid Geometry system. The new module supports efficient ray–triangle intersection tests and includes a GPU-optimized octree-based acceleration structure to enable fast particle transport on detailed anatomical phantoms. Additionally, the project includes tools for converting CT segmentations into polygonal meshes, storing material and body identifiers required by PENELOPE-based simulations, and 3D visualization routines for geometric and dosimetric analysis. The results demonstrate that polygonal meshes provide a more realistic anatomical description while maintaining computational efficiency through GPU acceleration, paving the way for next-generation Monte Carlo dose engines directly compatible with clinical platforms such as 3D Slicer RT. Keywords: Radiotherapy · Monte Carlo Simulation · GPU Computing · Polygonal Mesh Phantoms · PRISMMATIC

• Silva, Kyssylla (USP-RP, Brazil): Monte Carlo simulation for dosimetry study of electron beams with high dose rate.

Ionizing radiation can be used in processes for treatment, sterilization, and degradation of effluents. In this context, it is essential to accurately understand the characteristics of the beam used and the dose deposited in the material exposed to ionizing radiation. For this, dosimetry by Electron Paramagnetic Resonance (EPR) with Alanine dosimeters stands out due to its dosimetric properties, such as: linear response over a wide range of doses, non-destructive reading procedure, and no need for sample treatment prior to measurement. The aim of this work is to develop a dosimetric system with Alanine tablets for quality control of electron beams for high doses at energies of 500, 600, and 700 keV, in addition to using Monte Carlo simulation as a reference for obtaining and validating the dose maps generated by the developed systems. The beam is produced by an electron accelerator from the Nuclear and Energy Research Institute (IPEN), with energies of 500, 600, and 700 keV. The Monte Carlo simulation package Penelope was used to determine the dose-depth curve for the beams.

• Taube, Malena (Centro de Investigaciones y Transferencias Santa Cruz CONICET – Universidad Nacional de La Plata, Argentina): Evolution of Recovery Coefficients from Monte Carlo–Based PET Computational Simulations

In recent years, PET systems have become essential tools in medical diagnostics, where image quality strongly depends on both the acquisition procedures and the reconstruction algorithm implemented. In this work, we study the iterative algebraic method Maximum Likelihood Expectation Maximization MLEM under ideal conditions and including random events, using computational simulations based on the Monte Carlo method. We present preliminary results on the influence of voxel size on image quality, as well as the evolution of recovery coefficients as a function of the number of iterations for the NEMA image quality phantom, considering different contrast ratios between the background and the corresponding spheres. The main goal is to provide insights that contribute to a deeper understanding of the MLEM algorithm and, consequently, offering a valuable tool for nuclear medicine.

• Xavier, Francisco Harlley Dantas Hauradou (Programa de Engenharia Nuclear – UFRJ, Brazil): SPECT/CT-based Monte Carlo workflow for patient-specific internal dosimetry

Patient-specific internal dosimetry in radionuclide therapies relies on a realistic description of anatomy and activity distributions, as well as accurate radiation transport and energy deposition calculations. This work presents a workflow based on SPECT/CT images to estimate 3D absorbed dose-rate maps using Monte Carlo simulations. The CT image is used to build the voxelized geometry and assign tissue properties such as density and attenuation, while SPECT provides an image derived activity distribution after calibration preprocessing steps. These inputs feed the Monte Carlo simulation to obtain voxel-by-voxel energy deposition and the corresponding dose estimates in regions of interest. The workflow includes a staged validation plan, characterizing statistical uncertainty as a function of simulation runtime, and performing a preliminary sensitivity analysis to identify which imaging and modeling factors most impact the final dose estimates. As an expected outcome, this work aims to establish a reproducible and traceable calculation workflow for Monte Carlo–based absorbed dose estimates from SPECT/CT, supported by staged validation and uncertainty characterization.

 

2026-03-30

• 09:00 - Frédéric Tessier (National Research Council of Canada (NRC)): Introduction to the Monte Carlo Method
• 10:00 - Reid Townson (Carleton University): Looking inside EGSnrc
• 14:30 - Ernesto Mainegra-Hing (National Research Council of Canada (NRC)): Photon physics
• 15:30 - Frédéric Tessier (National Research Council of Canada (NRC)): Electron physics 1
• 16:30 - Frédéric Tessier (National Research Council of Canada (NRC)): Electron physics 2

2026-03-31

• 09:00 - Frédéric Tessier (National Research Council of Canada (NRC)): BEAMnrc introduction and main inputs
• 10:00 - Reid Townson (Carleton University): Phase space file analysis with beamdp
• 14:30 - Ernesto Mainegra-Hing (National Research Council of Canada (NRC)): DOSXYZnrc dose calculation in a phantom

2026-04-01

• 09:00 - Frédéric Tessier (National Research Council of Canada (NRC)): Fundamental geometry definition: howfar and hownear
• 10:00 - Reid Townson (Carleton University): egs++: the EGSnrc C++ library and geometries
• 14:30 - Reid Townson (Carleton University): egs++ particle sources
• 15:30 - Ernesto Mainegra-Hing (National Research Council of Canada (NRC)): EGSnrc scoring and egs++ ausgab objects

2026-04-02

• 09:00 - Reid Townson (Carleton University): egs++ applications
• 10:00 - Frédéric Tessier (National Research Council of Canada (NRC)): egs++ applications: egs_chamber
• 14:30 - Frédéric Tessier (National Research Council of Canada (NRC)): Verifying simulations with the Fano test
• 15:30 - Ernesto Mainegra-Hing (National Research Council of Canada (NRC)): FLURZnrc, SPRRZnrc and g

2026-04-03

• 09:00 - Ernesto Mainegra-Hing (National Research Council of Canada (NRC)): egs++ applications: egs_kerma, egs_cbct
• 14:30 - Reid Townson (Carleton University): Automation of EGSnrc simulations

Venue: The event will be held at IFT-UNESP, located at R. Jornalista Aloysio Biondi, 120 – Barra Funda, São Paulo. The easiest way to reach us is by subway or bus, See arrival instructions here.

Attention! Some participants in ICTP-SAIFR activities have received email from fake travel agencies asking for credit card information. All communication with participants will be made by ICTP-SAIFR staff using an e-mail “@ictp-saifr.org”. We will not send any mailings about accommodation that require a credit card number or any sort of deposit. Also, if you are staying at Hotel Intercity the Universe Paulista, please confirm with the Uber/Taxi driver that the hotel is located at Rua Pamplona 83 in Bela Vista (and not in Jardim Etelvina).

 

Attention! Some participants in ICTP-SAIFR activities have received email from fake travel agencies asking for credit card information. All communication with participants will be made by ICTP-SAIFR staff using an e-mail “@ictp-saifr.org”. We will not send any mailings about accommodation that require a credit card number or any sort of deposit. Also, if you are staying at Hotel Intercity the Universe Paulista, please confirm with the Uber/Taxi driver that the hotel is located at Rua Pamplona 83 in Bela Vista (and not in Jardim Etelvina).

BOARDING PASS: All participants, whose travel has been provided or will be reimbursed by ICTP-SAIFR, should bring the boarding pass  upon registration. The return boarding pass (PDF, if online check-in, scan or picture, if physical) should be sent to secretary@ictp-saifr.org by e-mail.

Visa information: Nationals from several countries in Latin America and Europe are exempt from tourist visa. Nationals from Australia, Canada and USA are required to apply for a tourist visa.

Accommodation: Participants, whose accommodation will be provided by the institute, will stay at Hotel Intercity the Universe Paulista. Hotel recommendations are available here.

Poster presentation: Participants who are presenting a poster MUST BRING A PRINTED BANNER . The banner size should be at most 1 m (width) x 1,5 m (length). We do not accept A4 or A3 paper.

Badge: You will receive an identification badge upon registration, which must be used during the entire event. Without the badge, it may not be possible to enter the venue.

Security issues: Although São Paulo is a relatively safe city, be careful when using cellphones on the street, avoid isolated areas at night, and be aware when crossing the street that cars may not stop for pedestrians. Also, please do not leave valuable items like laptops unattended even for short breaks.