A computational tool based on voxel geometry for dose reconstruction of a radiological accident due to external exposure

Reconstructive dosimetry and radiation dose evaluation of workers and public due to a Brazilian radiological accident in industrial radiography

Radiological accidents occur mainly in the practices recognized as high risk and which are classified by the International Atomic Energy Agency (IAEA) as Categories 1 and 2: radiotherapy, industrial irradiators and industrial radiography. In Brazil, five important cases in industrial gamma radiography occurred from 1985 to 2018, involving seven radiation workers and 19 members of the public. The accidents caused localized radiation lesions on the hands and fingers. One of these accidents is the focus of this paper. In this accident, a 3.28 TBq¹⁹²Ir radioactive source was left unshielded for 9 h in a non-destructive testing (NDT) company parking lot, and many radiation workers, employees and public, including teachers of a primary school were exposed. The radioactive source was also directly handled by a security worker for about 1.5 min causing severe radiation injuries in the hand and fingers. This paper presents radiation dose estimates for all accidentally exposed individuals. Four scenarios were considered, and three internationally recognised and updated reconstructive dosimetry techniques were used, named, Brazilian visual Monte Carlo Dose Calculation (VMC), virtual environment for radiological and nuclear accidents simulation (AVSAR) and RADPRO Calculator®. The main radiation doses estimated by VMC were the absorbed dose of 34 Gy for the security worker's finger and his effective dose of 91 mSv; effective doses from 43 to 160 mSv for radiation workers and NDT employees; and effective doses of 9 mSv for teachers in the schoolyard.

Physical dosimetry reconstructions of significant radiation exposure at an industrial accelerator facility in Tianjin (China)

The goal of this thesis is to estimate the physical radiation doses for two victims who were accidently exposed to an industrial electron beam at an industrial accelerator facility on 7 July 7 2016 in Tianjin, China. On the basis of the radiation source parameters, irradiation situation and irradiation time, physical dose reconstruction was carried out at the accident site by using a Bottle-Manikin-Absorption (BOMAB) phantom and an Alderson Radiation Therapy (ART) phantom. With thermoluminscent dosimeters (TLDs), skin estimation was conducted for the feet, calves, upper arms, left side of the body and neck, and the mean dose was estimated to be 14.1 ± 5.6 Gy. The foot and leg skin received the highest dose, which was >16.3 Gy. In addition, the mean dose estimated for the eye lens was 0.18 ± 0.07 Gy. The organ effective dose estimated and the total organs effective dose estimated were 0.46-4.94 mSv and 0.21 Sv, respectively. In the course of the accident, the damage caused by the electron radiation field to the exposed person was mainly to the skin, and the contributions to other radiation-sensitive organs were small. The damage to the organs other than the skin was mainly caused by the X-rays generated by the bremsstrahlung of the electron beam from the environment or the human body.

Reconstructive dosimetry for cutaneous radiation syndrome

According to the International Atomic Energy Agency (IAEA), a relatively significant number of radiological accidents have occurred in recent years mainly because of the practices referred to as potentially high-risk activities, such as radiotherapy, large irradiators and industrial radiography, especially in gammagraphy assays. In some instances, severe injuries have occurred in exposed persons due to high radiation doses. In industrial radiography, 80 cases involving a total of 120 radiation workers, 110 members of the public including 12 deaths have been recorded up to 2014. Radiological accidents in industrial practices in Brazil have mainly resulted in development of cutaneous radiation syndrome (CRS) in hands and fingers. Brazilian data include 5 serious cases related to industrial gammagraphy, affecting 7 radiation workers and 19 members of the public; however, none of them were fatal. Some methods of reconstructive dosimetry have been used to estimate the radiation dose to assist in prescribing medical treatment. The type and development of cutaneous manifestations in the exposed areas of a person is the first achievable gross dose estimation. This review article presents the state-of-the-art reconstructive dosimetry methods enabling estimation of local radiation doses and provides guidelines for medical handling of the exposed individuals. The review also presents the Chilean and Brazilian radiological accident cases to highlight the importance of reconstructive dosimetry.

Development of modified voxel phantoms for the numerical dosimetric reconstruction of radiological accidents involving external sources: implementation in SESAME tool

Estimating the dose distribution in a victim's body is a relevant indicator in assessing biological damage from exposure in the event of a radiological accident caused by an external source. When the dose distribution is evaluated with a numerical anthropomorphic model, the posture and morphology of the victim have to be reproduced as realistically as possible. Several years ago, IRSN developed a specific software application, called the simulation of external source accident with medical images (SESAME), for the dosimetric reconstruction of radiological accidents by numerical simulation. This tool combines voxel geometry and the MCNP(X) Monte Carlo computer code for radiation-material interaction. This note presents a new functionality in this software that enables the modelling of a victim's posture and morphology based on non-uniform rational B-spline (NURBS) surfaces. The procedure for constructing the modified voxel phantoms is described, along with a numerical validation of this new functionality using a voxel phantom of the RANDO tissue-equivalent physical model.

Accurate dose assessment system for an exposed person utilising radiation transport calculation codes in emergency response to a radiological accident

A system has been developed to assess radiation dose distribution inside the body of exposed persons in a radiological accident by utilising radiation transport calculation codes-MCNP and MCNPX. The system consists mainly of two parts, pre-processor and post-processor of the radiation transport calculation. Programs for the pre-processor are used to set up a 'problem-dependent' input file, which defines the accident condition and dosimetric quantities to be estimated. The program developed for the post-processor part can effectively indicate dose information based upon the output file of the code. All of the programs in the dosimetry system can be executed with a generally used personal computer and accurately give the dose profile to an exposed person in a radiological accident without complicated procedures. An experiment using a physical phantom was carried out to verify the availability of the dosimetry system with the developed programs in a gamma ray irradiation field.

SESAME: a software tool for the numerical dosimetric reconstruction of radiological accidents involving external sources and its application to the accident in Chile in December 2005

Estimating the dose distribution in a victim's body is a relevant indicator in assessing biological damage from exposure in the event of a radiological accident caused by an external source. This dose distribution can be assessed by physical dosimetric reconstruction methods. Physical dosimetric reconstruction can be achieved using experimental or numerical techniques. This article presents the laboratory-developed SESAME--Simulation of External Source Accident with MEdical images--tool specific to dosimetric reconstruction of radiological accidents through numerical simulations which combine voxel geometry and the radiation-material interaction MCNP(X) Monte Carlo computer code. The experimental validation of the tool using a photon field and its application to a radiological accident in Chile in December 2005 are also described.

New approach to radiation burn treatment by dosimetry-guided surgery combined with autologous mesenchymal stem cell therapy

The therapeutic management of severe radiation burns remains a challenging issue. Conventional surgical treatment (excision and skin autograft or rotation flap) often fails to prevent unpredictable and uncontrolled extension of the radiation necrotic process. We report here an innovative therapeutic strategy applied to the victim of a radiation accident (December 15, 2005) with an iridium gammagraphy radioactive source (192Ir, 3.3 TBq). The approach combined numerical dosimetry-guided surgery with cellular therapy using mesenchymal stem cells. A very severe buttock radiation burn (2000 Gy at the center of the skin surface lesion) of a 27-year-old Chilean victim was widely excised (10 cm in diameter) using a physical and anatomical dose reconstruction in order to better define the limit of the surgical excision in apparently healthy tissues. A secondary extension of the radiation necrosis led to a new excision of fibronecrotic tissues associated with a local cellular therapy using autologous expanded mesenchymal stem cells as a source of trophic factors to promote tissue regeneration. Bone marrow-derived mesenchymal stem cells were expanded according to a clinical-grade technique using closed culture devices and serum-free medium enriched in human platelet lysate. The clinical evolution (radiation pain and healing progression) was favorable and no recurrence of radiation inflammatory waves was observed during the 11 month patient's follow-up. This novel multidisciplinary therapeutic approach combining physical techniques, surgical procedures and cellular therapy with adult stem cells may be of clinical relevance for improving the medical management of severe localized irradiations. It may open new prospects in the field of radiotherapy complications.

Dosimetric comparison of Monte Carlo codes (EGS4, MCNP, MCNPX) considering external and internal exposures of the Zubal phantom to electron and photon sources

This paper aims at comparing dosimetric assessments performed with three Monte Carlo codes: EGS4, MCNP4c2 and MCNPX2.5e, using a realistic voxel phantom, namely the Zubal phantom, in two configurations of exposure. The first one deals with an external irradiation corresponding to the example of a radiological accident. The results are obtained using the EGS4 and the MCNP4c2 codes and expressed in terms of the mean absorbed dose (in Gy per source particle) for brain, lungs, liver and spleen. The second one deals with an internal exposure corresponding to the treatment of a medullary thyroid cancer by 131I-labelled radiopharmaceutical. The results are obtained by EGS4 and MCNPX2.5e and compared in terms of S-values (expressed in mGy per kBq and per hour) for liver, kidney, whole body and thyroid. The results of these two studies are presented and differences between the codes are analysed and discussed.

Current developments at IRSN on computational tools dedicated to assessing doses for both internal and external exposure

The paper presents the OEDIPE (French acronym that stands for tool for personalised internal dose assessment) and SESAME (for simulation of external source accident with medical images) computational tools, dedicated to internal and external dose assessment, respectively, and currently being developed at the Institute for Radiological Protection and Nuclear Safety. The originality of OEDIPE and SESAME, by using voxel phantoms in association with Monte Carlo codes, lies in their ability to construct personalised voxel phantoms from medical images and automatically generate the Monte Carlo input file and visualise the expected results. OEDIPE simulates in vivo measurements to improve their calibration, and calculates the dose distribution taking both internal contamination and internal radiotherapy cases into account. SESAME enables radiological overexposure doses to be reconstructed, as also victim, source and accident environment modelling. The paper presents the principles on which these tools function and an overview of specificities and results linked to their fields of application.