Jai-Ki Lee

Korean adult male voxel model KORMAN segmented from magnetic resonance images

A voxel model of Korean adult male, KORMAN, was developed by processing whole-body magnetic resonance (MR) images of a healthy volunteer who represents an approximately average Korean in height and weight. Layer by layer the MR images were semi-automatically segmented and indexed using a graphic software and digitizer to construct data arrays consisting of 250 x 120 x 170 voxels of a size of 2 x 2 x 10 mm3. To assess the utility of the model, some illustrative dosimetric calculations were made to obtain organ absorbed doses and effective doses to the KORMAN placed in broad parallel photon fields with energies ranging from 0.05 to 10 MeV. The results were compared with those based on the medical internal radiation dose (MIRD)-type models given in ICRP74. The effective doses of ICRP74 were higher than those of KORMAN with percent differences ranging from 6% (LLAT, 10 MeV) to 30% (PA, 0.05 MeV). Significant differences of more than 40% were observed in organ absorbed doses for some organs including bone surface (AP), stomach (PA), and testes (LAT) for low photon energy. These are mainly caused by difference in trunk thickness between MIRD-type model and KORMAN, and differences in organ positions in the body.

Implementation of Japanese Male and Female Tomographic Phantoms to Multi-particle Monte Carlo Code for Ionizing Radiation Dosimetry

Japanese male and female tomographic phantoms, which have been developed for radio-frequency electromagnetic-field dosimetry, were implemented into multi-particle Monte Carlo transport code to evaluate realistic dose distribution in human body exposed to radiation field. Japanese tomographic phantoms, which were developed from the whole body magnetic resonance images of Japanese average adult male and female, were processed as follows to be implemented into general purpose multi-particle Monte Carlo code, MCNPX2.5. Original array size of Japanese male and female phantoms, 320x160x866 voxels and 320x160x804 voxels, respectively, were reduced into 320x160x433 voxels and 320x160x402 voxels due to the limitation of memory use in MCNPX2.5. The 3D voxel array of the phantoms were processed by using the built-in repeated structure algorithm, where the human anatomy was described by the repeated lattice of tiny cube containing the information of material composition and organ index number. Original phantom data were converted into ASCII file, which can be directly ported into the lattice card of MCNPX2.5 input deck by using in-house code. A total of 30 material compositions obtained from International Commission on Radiation Units and Measurement (ICRU) report 46 were assigned to 54 and 55 organs and tissues in the male and female phantoms, respectively, and imported into the material card of MCNPX2.5 along with the corresponding cross section data. Illustrative calculation of absorbed doses for 26 internal organs and effective dose were performed for idealized broad parallel photon and neutron beams in anterior-posterior irradiation geometry, which is typical for workers at nuclear power plant. The results were compared with the data from other Japanese and Caucasian tomographic phantom, and International Commission on Radiological Protection (ICRP) report 74. The further investigation of the difference in organ dose and effective dose among tomographic phantoms for other irradiation geometries should be carried out by employing additional tomographic phantoms to find out the dosimetric difference of Asian human phantoms from Caucasian-based phantoms.

Robust fuzzy neural network sliding mode control scheme for IPMSM drives

This article proposes a robust fuzzy neural network sliding mode control (FNNSMC) law for interior permanent magnet synchronous motor (IPMSM) drives. The proposed control strategy not only guarantees accurate and fast command speed tracking but also it ensures the robustness to system uncertainties and sudden speed and load changes. The proposed speed controller encompasses three control terms: a decoupling control term which compensates for nonlinear coupling factors using nominal parameters, a fuzzy neural network (FNN) control term which approximates the ideal control components and a sliding mode control (SMC) term which is proposed to compensate for the errors of that approximation. Next, an online FNN training methodology, which is developed using the Lyapunov stability theorem and the gradient descent method, is proposed to enhance the learning capability of the FNN. Moreover, the maximum torque per ampere (MTPA) control is incorporated to maximise the torque generation in the constant torque region and increase the efficiency of the IPMSM drives. To verify the effectiveness of the proposed robust FNNSMC, simulations and experiments are performed by using MATLAB/Simulink platform and a TI TMS320F28335 DSP on a prototype IPMSM drive setup, respectively. Finally, the simulated and experimental results indicate that the proposed design scheme can achieve much better control performances (e.g. more rapid transient response and smaller steady-state error) when compared to the conventional SMC method, especially in the case that there exist system uncertainties.

External dose-rate conversion factors of radionuclides for air submersion, ground surface contamination and water immersion based on the new ICRP dosimetric setting

For the assessment of external doses due to contaminated environment, the dose-rate conversion factors (DCFs) prescribed in Federal Guidance Report 12 (FGR 12) and FGR 13 have been widely used. Recently, there were significant changes in dosimetric models and parameters, which include the use of the Reference Male and Female Phantoms and the revised tissue weighting factors, as well as the updated decay data of radionuclides. In this study, the DCFs for effective and equivalent doses were calculated for three exposure settings: skyshine, groundshine and water immersion. Doses to the Reference Phantoms were calculated by Monte Carlo simulations with the MCNPX 2.7.0 radiation transport code for 26 mono-energy photons between 0.01 and 10 MeV. The transport calculations were performed for the source volume within the cut-off distances practically contributing to the dose rates, which were determined by a simplified calculation model. For small tissues for which the reduction of variances are difficult, the equivalent dose ratios to a larger tissue (with lower statistical errors) nearby were employed to make the calculation efficient. Empirical response functions relating photon energies, and the organ equivalent doses or the effective doses were then derived by the use of cubic-spline fitting of the resulting doses for 26 energy points. The DCFs for all radionuclides considered important were evaluated by combining the photon emission data of the radionuclide and the empirical response functions. Finally, contributions of accompanied beta particles to the skin equivalent doses and the effective doses were calculated separately and added to the DCFs. For radionuclides considered in this study, the new DCFs for the three exposure settings were within ±10 % when compared with DCFs in FGR 13.

The effect of unrealistic thyroid vertical position on thyroid dose in the MIRD phantom.

Anatomically, the thyroid gland is placed in the lower neck, extending from the level of the fifth cervical vertebra down to the first thoracic vertebra. However, the thyroid of the Medical Internal Radiation Dose (MIRD) phantom, which has been widely used for dosimetric calculation, is located right above the top of the torso and is completely included in the neck region. To investigate the effect of the unrealistic position of the thyroid in the MIRD phantom on dose calculation, thyroid absorbed doses at various vertical positions were calculated for the idealized external broad parallel photon beam from anterior-posterior, posterior-anterior, right lateral (RLAT), and left lateral (LLAT) direction using the Monte Carlo method. The thyroid absorbed dose decreased by as much as about 60% for 0.05 MeV photon in both RLAT and LLAT irradiations when the thyroid was relocated to realistic position (inserted into the torso). The effective dose also decreased by 10%, consequently. The thyroid dose of the widely accepted stylized model, the MIRD phantom, is overestimated in RLAT and LLAT irradiation geometries.

Radiological risk assessment for field radiography based on two dimensional Monte Carlo analysis.

Probabilistic risk assessment studies use probability distributions for one or more variables of the risk equation in order to quantitatively characterize the variability and uncertainty. In this study, an advanced technique called the two-dimensional Monte Carlo analysis (2D MCA) is applied to estimation of radiological risk for worker and member of the public in the vicinity of the work place for field radiological system in Korea. The variables of the risk model along with the parameters of these variables are described in terms of probability density functions (PDFs). Because the frequencies of normal tasks were far higher than those of accidents, the total risk associated with normal tasks was higher than the accidental risk. The result derived from this work can be used as guidance for the decision-making in controlling the radiological risk in the field of radiography area.

A new perspective on severe nuclear accidents

The reactions of the public in Korea to the nuclear accident at the Fukushima Daiichi plants in Japan, particularly over-reactions, are reviewed, with the conclusion that significant radioactive contamination of a small country could lead to a severe national crisis. The most important factor is the socio-economic damage caused by stigma, which in turn is caused by a misunderstanding of the radiation risk. Given that nuclear power is an important choice in the face of the threat of climate change, the publics perceptions need to be changed at any cost, not only in those countries operating nuclear power plants but globally as well.

Development of Point-Kernel Code for Skin Dose Calculation

Abstract The new computer code K-SKIN has been developed for use in skin dose assessment. The K-SKIN code calculates the dose distribution over the contaminated area using point kernels of monoenergetic electrons. These kernels are averaged over the beta spectra of contaminated radionuclides to obtain the dose distributions. Then, beta dose rates to the skin are calculated by numerical integration of point-kernel data over the contaminated area. Photon dose rates, if involved, are calculated using the specific gamma-ray constant for the radionuclides. Three predefined source types are arranged: point, disk, and cylinder. Backscattering correction, source self-shielding of a volume source, and reduction by the shielding material and air gap are considered during dose calculation. K-SKIN employs MATLAB as the coding tool and provides a graphical user interface. To verify K-SKIN, the dose rates from the point and disk source of several radionuclides over 1.0-cm2 area at 70 μm skin depth were calculated and compared with results obtained from another point-kernel code VARSKIN 3 and the Monte Carlo simulation code MCNPX. The calculated results agreed within ±20%. The skin dose at various depths showed that the inclusion of energy-loss straggling in the point kernel improves the accuracy of the beta dose calculation at the deep region. The K-SKIN computer code will facilitate assessment of skin exposure at nuclear facilities.

Charge Coupled Device Camera–Scintillator System for Two Dimensional Proton Dose Measurements

A two-dimensional dosimeter composed of a charge coupled device (CCD) camera and a scintillating screen has been tested for measuring the dose distributions of therapeutic proton beams in situ. The dosimeter is the main component of the dosimetry system designed for the quality assurance of scanning beams in the course of patient treatments. A thin-wall parallel plate ionization chamber was built as part of the system to scale the beam currents, and a collimator and variable-thickness phantom being the other components. The system has been tested using a 40 MeV proton beam in a nontherapeutic beam line. The light output from the screen was linear within 0.1% at the typical dose rates of 1–2 Gy/min. The depth dose distributions were measured using the CCD camera system, and agreed well with the measurements by a calibrated ionization chamber. The CCD camera was also used to measure the beam intensity distributions produced with a Cu scatterer in the beam line. The comparison with the results of Monte Carlo simulations using MCNPX was useful for validating the code.

Internal dosimetry for intake of 18FDG using spot urine sample

In nuclear medicine, workers handle unsealed radioactive materials. Among the materials, (18)FDG is the most widely used in PET/CT technique. Because of the short half-life of (18)F, it is very challenging to monitor internal exposure of nuclear medicine workers using in vitro bioassay. Thus, the authors developed the new in vitro bioassay methodology for short half-life nuclides. In the methodology, spot urine sample is directly used without normalisation to 1-d urine sample and the spot urinary excretion function was newly proposed. In order to estimate the intake and committed dose for workers dealing (18)FDG, biokinetic models for FDG was also developed. Using the new methodology and biokinetic model, the in vitro bioassay for workers dealing (18)FDG was successfully performed. The authors expect that this methodology will be very useful for internal monitoring of workers who deal short-lived radionuclides in the all field as well as the nuclear medicine field.