On the Use of a Diagnostic X-Ray Machine for Calibrating Personal Dosimeters A. T. Baptista Neto, T. A. Da Silva Centro de Desenvolvimento da Tecnologia Nuclear, Comissão Nacional de Energia Nuclear, Rua Prof. Mario Werneck, s/n, P. O. Box 941, cep. 30123-970, Belo Horizonte, Brazil. E-mail: annibal@cdtn.br, silvata@cdtn.br Abstract. X-ray machines that comply with the ISO requirements for calibrating personal dosimeters sometimes are not available in a world region, mainly in developing countries. This is used as an excuse for a lack of personal dosimeter calibration and, consequently, a lack of reliability in the individual monitoring programs. Considering the availability of x-ray machines in many hospitals, this work shows the steps for implementing a calibration beam in a x-ray diagnostic machine and it compares the calibration results of an Agfa film based monitoring system in both conditions, i.e., with the implemented x-ray beam and with the ISO reference radiations from the Brazilian Metrology Laboratory. Results showed that in spite of the difference between the beam qualities used, the performance of the chosen personal dosimeter complied with the accuracy requirements for individual monitoring. 1. Introduction The exposure of workers to radiation must be controlled and monitored in order to comply with regulatory requirements. Dose values must be obtained by reliable personal dosimeters that are type-tested and calibrated under internationally agreed conditions [1]. Gamma and x-ray reference radiations for calibrating dosimeters and for determining their energy response as a function of photon energy were established by the International Organization for Standardization. Detailed information on the beam production and dosimetry is provided in the ISO standards [2,3]. Many countries have no x-ray machine that complies with the ISO requirements and individual monitoring programs suffer of a lack of reliability due to the lack of calibration. Because most hospitals have diagnostic x- ray machines, this work studied the feasibility of implementing a x-ray beam (obviously not similar to the ISO reference radiations) and using it for calibrating personal dosimeters. Instead of looking for very complex steps for a high accuracy, the work adopted minimum and very simple steps for a minimum reliability of the radiation geometry and conditions. X-ray parameters, such as voltage, half-value-layer, beam uniformity, filtration were measured and compared to those established by ISO reference radiation. As the influence on the monitoring results depends on the energy dependence of the personal dosimeter, a personal dosimetry system based on Agfa film monitoring was used to compare its response under the ISO and the implemented radiation conditions. 2. Implementation of the x-ray beam quality A diagnostic VMI x-ray machine, model Pulsar 800 Plus, installed at the Calibration Laboratory of the Centro de Desenvolvimento da Tecnologia Nuclear (CDTN) was used. Geometry conditions and x-ray parameters were verified in order to establish the personnel dosimeter calibration procedure. 2.1. Beam alignment The light beam for positioning patients in the x-ray beam was used for positioning ionization chambers and personal dosimeters. The alignment between the light and the radiation beams was verified by exposing a radiographic film following a recommended national methodology [4]. The shift between the light and the radiation beams did not exceed 1 cm at 100 cm distance from the x-ray tube. The image of a cylindrical tool for the test showed that the primary beam could be considered to be perpendicular to the irradiation plane within ± 1.5. 2.2. Accuracy and reproducibility of the high voltage The high voltage applied to the x-ray tube was measured with a calibrated Radcal Corporation kvp meter, model 4075, following the adopted methodology [4]. Results of 10 measurements (Table 1) show that the maximum deviation between the nominal and the measured values did not exceeded 2%; the high voltage reproducibility (standard deviation of the 10 measurements) also did not exceed 3%. 1
Table I Accuracy and reproducibility (standard deviation of 10 measurements) of the high voltage applied to the x-ray tube. Nominal kv value (kv) Measured kv value (kv) Standard deviation (%) 60 60.3 1.16 80 79.0 0.50 110 110.0 0.55 2.3. Beam uniformity Radiation fields of 240x180 mm 2, 120x90 mm 2 and 60x50 mm 2 were investigated as far as the beam uniformity. Air kerma measurements were carried out with a small ionization chamber (Radcal Corporation 6 cc chamber) at 9 points of the same plane of the radiation beam at 100 cm distance. Results obtained for the beam quality of 60 kv with total filtration of 4.1 mmal plus 0.3 mmcu (Table 2) show that the air kerma rate measured values varied from 1.6% to 3.2% for the smallest to the largest field sizes. It means that the obtained results complied with the ISO requirement of 5% for the maximum variation value of air kerma rate in the field area where dosimeters are to be irradiated. Table 2 - Uniformity of the radiation field. Field Size Maximum Variation (mm 2 ) (%) 240 x 180 3.2 120 x 90 1.9 60 x 50 1.6 2.4. Inherent filtration The inherent filtration of the x-ray machine was evaluated based on the measurements with a small volume (6 cc) and very low energy dependent (less than 3%) Radcal ionization chamber model 10X5-6, under low scatter geometry condition. The procedure recommended by ISO [2] was adopted but instead of high purity absorbers (very expensive and usually not available) commercially available filters were used. Based on the measurements the inherent filtration was estimated to be 2.1mmAl; this value was used to adjust to the total filtration required by ISO. 2.5 The choice of the beam quality The x-ray beam quality was chosen based on the calibration requirement of the chosen Agfa film dosimeter. A 137 Cs and a 48 kev mean energy radiation beams are required to calibrate the Agfa film based personal dosimetry system because the Simplex is used as the dose evaluation method. A similar ISO reference radiation from the narrow spectrum series with 60 kv (N60) is expected to be reproduced but due to its low air kerma rate a 60 kv reference radiation of the wide spectrum series (W60) with mean energy of 45 kev was chosen to be implemented. The total filtration established by ISO is 4.0 mmal plus 0.3 mmcu by adding high purity absorbers; the total filtration implemented was 4.1 mmal plus 0.3 mmcu by adding commercially available absorbers. The half-value layer was measured according the procedure suggested by Lacerda [5]; the small volume chamber (Radcal ionization chamber model 10X5-6) was used. Table 3 shows the results of the first and the second half-value layers measured with the ionization chamber, calculated by the XCOMP5R and those established by ISO. The homogeneity coefficient is also showed. 2
Table 3 Comparison among the half-value layers for the 60 kv radiation beam. 1 st HVL (mmcu) 2 nd HVL (mmcu) Homogeneity Coefficient Measured values 0.19 0.23 0.83 Calculated values (XCOMP5R) 0.19 0.22 0.86 ISO established values 0.18 0.21 0.86 3. Calibration of the personal dosimetry system 3.1 Characteristics of the personal dosimetry system A Brazilian routine personal dosimetry system was chosen to verify the influence on its calibration with the implemented x-ray beam and the ISO reference radiation. The system is based on an Agfa film, which is inserted in a plastic badge with 5 windows (open, plastic, aluminum, cupper 1, copper 2 and lead plus tin). The Agfa film based dosimetry system uses the Simplex method for dose evaluation purposes in the range 0,20 msv to 2.0 Sv, for photon energies from 15 kev to 1250 kev; this results in a variation with the radiation energy of about ±15%, at lower energies, related to 137 Cs (Fig. 1). Correction factor (normalized to Cs-137) 1,2 1,1 1,0 0,9 0,8 0 20 40 60 80 100 Mean energy (kev) Figure 1. Energy dependency of the Agfa film based Brazilian personal dosimetry system 3.2. Irradiation procedure for calibration and evaluation purpose At the CDTN Calibration Laboratory, dosimeters were exposed to the 60 kv x-ray beam, which was implemented at the VMI diagnostic x-ray machine, to doses from 0.10 to 5 msv (higher doses were not possible in order to save the machine). A set of dosimeters was also exposed to a 137 Cs gamma beam with doses from 0.10 to 30 msv. At the Brazilian Metrology Laboratory, dosimeters were exposed to the ISO reference radiation N60 or W60, in a Pantak x-ray machine, to doses from 0.10 to 30 msv. In both cases, processing, evaluation and generation of the calibration curve were done following the well-established routine procedure of the CDTN Individual Monitoring Service. 3.3. Comparison between calibrations Mean values of optical densities of the dosimeters irradiated at the Brazilian Metrology Laboratory were used to generate the calibration curve of the personal dosimetry system, which was used to evaluate each dosimeter individually. Fig. 2 shows that the calibration and evaluation procedures cause dispersion and overestimation of the evaluation doses. The curves (called trumpet-curves) represent the acceptable accuracy limits for dose evaluations. Fig. 3 shows the accuracy of the calibrated system during the evaluation of dosimeters irradiated at CDTN Calibration Laboratory at the diagnostic x-ray machine. 3
2,50 0,10 1 10 Figure 2 Accuracy of the personal dosimetry system calibrated at the Brazilian Metrology Laboratory during the evaluation of dosimeters irradiated at the same Laboratory. 2,50 0,10 1 10 Figure 3 Accuracy of the personal dosimetry system calibrated at the Brazilian Metrology Laboratory during the evaluation of dosimeters irradiated at the CDTN Calibration Laboratory with a diagnostic x-ray machine. Following with the same procedure, Fig. 4 and 5 show the accuracy of the dosimetry system calibrated with dosimeters irradiated in the diagnostic x-ray machine during the evaluation of each dosimeter and during the evaluation of dosimeters irradiated at the Brazilian Metrology Laboratory. Results show that calibration and evaluations at both conditions, i.e., at the Brazilian Metrology Laboratory with ISO reference radiation and at CDTN Calibration Laboratory with a diagnostic x-ray machine agreed within 10% to doses up to 1 msv and within 5% to higher doses. In all cases, the accuracy was acceptable and it complied with the trumpet-curves requirements. Although the calibrations with the diagnostic x-ray machine was done up to 5 msv, no bad results was found during the evaluation of dosimeters irradiated at the Brazilian Metrology Laboratory with doses up to 30 msv.. 4
2,50 0,10 1 10 Figure 4 Accuracy of the personal dosimetry system calibrated at the CDTN Calibration Laboratory with the diagnostic x-ray machine during the evaluation of dosimeters irradiated at the same Laboratory. 2,50 0,10 1 10 Figure 5 Accuracy of the personal dosimetry system calibrated at the CDTN Calibration Laboratory with the diagnostic x-ray machine during the evaluation of dosimeters irradiated at the Brazilian Metrology Laboratory. 5
4. Evaluation of user doses under both calibrated conditions To confirm the acceptable performance as far as dose accuracy, 30 dosimeters from real users were evaluated in both conditions, i.e., with the personal dosimetry system calibrated with ISO reference radiation by the Brazilian Metrology Laboratory and with a x-ray beam close but obviously not similar to ISO reference radiation which was implemented at the CDTN Calibration Laboratory. Fig. 6 shows the dose ratio between both dose evaluations. 1,20 1,10 Dose ratio 0,90 0,80 0,1 1,0 10,0 Dose value (msv) Fig. 6 Dose ratios of dosimeters from real users when are evaluated with the personal dosimetry system calibrated with the ISO reference radiation and with a beam from a diagnostic x-ray machine. Results show that 26 of 30 (87%) dose evaluations agreed within ±10% and all doses agreed within ±15%. One believes that the variation is mainly due to the energy dependence of the personal dosimetry system used in this work. It is important to emphasize that the use of the calibration procedure proposed in this work is valid to the personal dosimetry system here described and extensively to others systems with lower energy dependence. Dosimeters with high-energy dependence would suffer influence of the energy spectrum that could cause unacceptable dose errors. REFERENCES 1. International Atomic Energy Agency (IAEA), International Basic Safety Standards for Protection against Ionizing Radiation and for Safety of Radiation Sources, Safety Series 115, Safety Standards, IAEA, Vienna (1996). 2. International Organization for Standardization (ISO), X and gamma reference radiations for calibrating and determining their response as a function of photon energy. ISO 4037/Part 1: Radiation characteristics and production methods, ISO, Geneva (1996). 3. International Organization for Standardization (ISO), X and gamma reference radiations for calibrating and determining their response as a function of photon energy. ISO 4037/Part 2: Dosimetry for radiation protection over the energy range 8 kev to 1.3 MeV and 4 MeV to 9 MeV, ISO 4037 Part 3: Calibration of area and personal dosemeters and the measurement of their energy and angle of incidence, ISO, Geneva (1998). 4. Mota, H. C. et al. Protocol for radiological protection evaluation in radio diagnostic area. Ministério da Saúde. Rio de Janeiro, RJ (2000) (in Portuguese). 5. Lacerda, M.A.S. On the measurement of the half-value layer in diagnostic radiology. M.Sc. Thesis, Universidade Federal de Minas Gerais, Belo Horizonte, Brasil (2002) (in Portuguese), 6