Technology and Health Care 22 (2014) 345–350 DOI 10.3233/THC-140790 IOS Press
345
Monte Carlo simulation for correlation analysis of average glandular dose by breast thickness and glandular ratio in breast tissue Sang-Tae Kima and Jung-Keun Chob,∗
a Radiation
Safety Division, Nuclear Safety and Security Commission, Seoul, Korea of Radiological Science, Jeonju University, Jeonju, Korea
b Department
Received 26 October 2013 Accepted 22 January 2014 Abstract. A glandular breast tissue is a radio-sensitive tissue. So during the evaluation of an X-ray mammography device, Average Glandular Dose (AGD) measurement is a very important part. In reality, it is difficult to measure AGD directly, Monte Carlo simulation was used to analyze the correlation between the AGD and breast thickness. As a result, AGDs calculated through the Monte Carlo simulation were 1.64, 1.41 and 0.88 mGy. The simulated AGDs mainly depend on the glandular ratio of the breast. With the increase of glandular breast tissue, absorption of low photon-energy increased so that the AGDs increased, too. In addition, the thicker the breast was, the more the AGD became. Consequently, this study will be used as basic data for establishing the diagnostic reference levels of mammography. Keywords: Glandular, average glandular dose, Monte Carlo simulation, breast thickness
1. Introduction With concern and anxiety regarding man’s radiation exposure continuing to rise after the disaster at Fukushima Nuclear Power Plant (2011), it is critical for subjects to have accurate information about their exposure doses during an X-ray test since it can help them reduce their anxiety and their radiologists or doctors improve their awareness of lower exposure doses. In particular, it is very important to establish an accurate system for assessing patient exposure doses during an mammography, which requires relatively longer radiation exposure time per test. Essential to the establishment of a more accurate patient doses assessment system is information about dose distributions by considering interactions between the major organs of the patient exposed to radiation and the radiation. Also needed to assess effective doses by multiplying each organ dose by their corresponding tissue weighting factors and adding them up is information about the dose distributions of each organ. It is difficult to directly measure the equivalent doses of various organs when the body is exposed to radiation, which is why humanoid and mathematical phantoms have been used in most patient exposure doses ∗ Corresponding author: Jung-Keun Cho, Department of Radiological Science, Jeonju University, 303, Cheonjam-ro, Wansangu, Jeonju-si, Jeollabuk-do, Korea. Tel.: +82 63 220 3129; Fax: +82 63 220 2054; E-mail: [email protected].
c 2014 – IOS Press and the authors. All rights reserved 0928-7329/14/$27.50
346
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose
assessments [1–3]. X-ray mammography, a breast cancer test method using X-ray with high contrast and resolution is the most frequently used early diagnosis of breast cancer currently. Yet, with the increasing frequency of use of mammography X-ray unit, interests in radiation injuries due to X-ray mammography are heightened [4]. Since a glandular breast tissue is a tissue with a high radio sensitivity, a carcinogenic risk in an X-ray mammography continues to be raised. In X-ray mammography test, Average Glandular Dose (AGD) measurement is an important item for the quality assessment of mammography X-ray unit [5]. The Committee on Quality Assurance in Mammography (CQAM) of the American College of Radiology (ACR) provides that the AGD is less than 3 mGy in breast thickness 4.2 cm with a breast tissue ratio of adipose tissues of 50% and glandular tissues of 50% [6]. Yet, since AGD in fact, differs depending on breast thickness, even though it does not exceed the advised degree of 3 mGy, you cannot say that the glandular dose to which the patient is exposed is less. This study attempted to draw out the correlation of AGD according to a variety of breast thickness and glandular ratio using Monte Carlo simulation. 2. Monte Carlo simulation GEANT (Geometry And Tracking) 4, the calculation code used in Monte Carlo simulation in this study is a code developed in C++, which can carry out computer simulations of microscopic phenomena for the interactions between particles and substances. In addition, it is widely used in various domains such as high energy physics, medical physics and astrophysics, and it has a variety of physical models which can be applied to interaction research on photons, electronics and hadrons. GEANT 4 provides three types of model, that is, standard, low energy and Penelope model for electromagnetic interactions. This study used the low energy model. The low energy model can apply to the energy range from 250 eV to 100 GeV, and considers the fluorescence of excited atoms including photoelectric effect, Compton scattering, Rayleigh scattering, bremsstrahlung and ionization [7]. The version of GEANT4 used for calculation in this study is 9.4.p01. The phantom simulated for the correlation analysis of the AGD by breast thickness and glandular ratio has been reported in preceding studies [8]. The standard breast phantom is wagon headed with a diameter of 160 mm and a thickness of 45 mm, and in the middle domain, adipose tissues and glandular tissues are mixed at the ratio of 50:50, and surrounded by 100% adipose tissues with a thickness of 5 mm. For the correlation analysis of the AGD by breast thickness and glandular ratio, the thickness of the standard breast phantom was simulated increasing by 1 cm from 1 cm to 10 cm. The correlation was analyzed while glandular tissues in the middle domain were increased by 2% from 0% to 100%. The X-ray spectrum applied to simulate the source of X-ray mammography device was set up using the technical documents of the X-ray shooting equipment of this study and IPEM-78 [9]. IPEM-78 is a code providing the spectrum of the emitted X-ray, with three target substances of tungsten, rhodium and molybdenum, the range of tube voltage of 30 ∼ 150 kVp, the target angle of 6 ∼ 22◦ , and information about filter materials of aluminum, beryllium, copper, air and bone, which is composed so that the user can use it adjusted to the characteristics of the relevant X-ray device. Based on Tables 1–3, the results of X-ray spectrum generation in IPEM-78 were shown in Fig. 1. To simulate it with the source closest to the continuous spectrum, the photons integrated at an interval of 0.5 kV were applied in the obtained energy spectrum. To increase efficiency by increasing the simulation speed, a geometric approximation method like the solid angle limitation of radiation source was used [10]. Point sources were assumed and approximated to a fan shape at an angle proper for each radiation field. The distance from the breast support to the focus is 605 mm and the central beam axis is perpendicular to the surface of the phantom.
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose
347
Table 1 Density and elemental ratio of compression paddle and PMMA breast phantom Density (g/cm3 ) Compression paddle PMMA breast phantom
1.2 1.19
H 0.0556 0.0805
Component C 0.7925 0.5998
O 0.1509 0.3196
Fig. 1. Result of X-ray spectrum generation in IPEM-78.
The focus is an isotropic point source, and the solid angle was calculated as an X-ray beam in a half cone-shape with the radius of 80 mm at 480 mm point from the focus (Fig. 2). 2 × 109 beam histories of the used photon were used. In X-ray mammography, the compression paddle is included in the X-ray radiation field, which may affect the AGD measurement by grade and thickness due to X-ray attenuator, so it must be simulated. The simulated compression paddle is a thickness of 2 mm. The compression paddle to apply Monte Carlo simulation, the density and elemental ratio of the PMMA breast phantom, X-ray exposure condition and HVL by tube voltage were shown in Tables 1–3. 3. Results A standard PMMA phantom with a diameter of 200 mm and a thickness of 45 mm was simulated to measure AGD. The thickness of the PMMA phantom decreased to 45, 35 and 25 mm to calculate the AGD. The AGDs calculated to the Monte Carlo simulation were 1.64, 1.41 and 0.88 mGy (Table 4). For the correlation analysis of the AGD by the simulated breast thickness and glandular ratio, a wagonheaded standard breast phantom with a diameter of 160 mm and a thickness of 45 mm was simulated. The glandular tissues of the standard breast phantom were set to 0% (100% adipose tissue), 25% (25% glandular, 75% adipose tissue), 50% (50% glandular, 50% adipose tissue), 75% (75% glandular, 25% adipose tissue) and 100% (100% glandular), respectively, and the thickness of each phantom was increased by 1 cm from 1 cm to 10 cm to calculate the AGD at each thickness (Fig. 3). The AGDs at the thicknesses of 1–10 cm of the phantom with glandular tissues of 50% was calculated at 1.18–1.57 mGy. The correlation was analyzed while the glandular tissues in the middle domain of the
348
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose Table 2 X-ray exposure condition Target Filter Tube Voltage [kV] Tube Current [mAs]
Mo Mo, Rh 25 60
Table 3 HVL set point by tube voltage provided in IEC Tube Voltage [kV] 25 28 30
1st HVL [mmAl] 0.28 0.31 0.33
Table 4 AGDs by PMMA thickness obtained through Monte Carlo simulation Thickness of PMMA Simulated phantom [mm] AGD [mGy] 45 1.64 35 1.41 25 0.88
Fig. 3. Changes in AGDs by compressed breast thickness according to the ratio of adipose tissues and glandular tissues.
Fig. 2. Geometric structure for application of Monte Carlo simulation. (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/THC-140790)
Fig. 4. Correlation between glandular ratio and AGD in 45 mm standard breast phantom.
standard breast phantom of 45 mm were increased by 2% from 0% to 100%. The irradiation condition for the simulation was simulated at 29 kV and 60 mAs. When the glandular tissues were 0–100%, the AGDs were calculated at 1.23–2.04 mGy, respectively (Fig. 4). 4. Discussion In an X-ray mammography, a compression paddle should be included in the X-ray radiation field, and by grade and thickness, it occupies a part of the total capacity of the filtration as X-ray attenuator.
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose
349
Thus, since the compressor may affect measurements of the quality of radiation and AGD, it should be included in the radiation field in actual experiment or simulation. This study understood the grade and thickness of the compression paddle used in the experiment to carry out accurate simulation, which was included in the radiation field. The half-value-layer measured to calculate AGD is an index showing the quality of radiation characteristics for kVp set up regularly, and using the kVp used in clinics, it was measured for the beam that penetrated the compression paddle. Since the quality of radiation greatly affects the contrast, and the AGD calculated from the incident air Kerma differs depending on HVL, the HVL measurement is an important element to determine the AGD of a digital mammography system. Mammography Quality Standards Act (MQSA) provided in Part 900 of the Code of Federal Regulations regulates the HVL in 30 kVp at least 0.3 mmAl for an Mo or Mo-W alloy target. The value provided in the IEC was used for the first half-value-layer of the system used in this study. The second half-valuelayer is defined as the difference between the Al attenuator thickness needed to reduce the glandular dose to 1/4 and the first HVL while the homogeneity coefficient is defined as the ratio of the first HVL and the second HVL, to which 0.8 was applied. The homogeneity coefficient is an index showing the width of an X-ray spectrum, calculated at 0 to 1, and the greater the value, the narrow the spectrum becomes. The homogeneity coefficient used in diagnostic radiology is 0.7 ∼ 0.9. To enhance the precision of measurement of quality of radiation, it is important to establish a standardized energy correction table on the half-value-layer 0.3 ∼ 0.4 mmAl equivalent beam. The AGDs in most cases of actual calculation turned out to be less than 3 mGy, which was provided in the International Standard Regulations. The additional magnification view mammography and Stereotactic mammography belong to the test in the diagnosis domain, which might not be treated the same as the radiation injury, but it is a glandular dose which cannot be ignored if the exposure of the domain in which relatively high glandular dose is received. In addition, AGDs differ depending on breast thickness, so even if it does not exceed the advised degree of 3 mGy, you cannot say that the glandular dose to which the patient is exposed is less. In the simulation, as a result of the calculation of AGD while the thickness of the standard breast phantom was increased by 1 cm from 1 cm to 10 cm, it was found that the value of AGD gradually increased with the increase of breast thickness. The thicker the breast, the smaller the difference in the absorption between structures gets by the beam hardening effect, which lowers the subject contrast. In addition, the influence of the scattered rays lowers the subject contrast and increases the glandular dose exposed to the patient, too. To minimize this, the breast is pressed in X-ray mammography. Pressing the breast firmly can reduce the scattered rays and minimize the blurring by movement. For the young women with a low weight, the ratio of dense breast is relatively high, so in X-ray mammography, it involves the patients’ pain. Thus, efforts to minimize the patients’ pain with proper press are necessary. In the simulation, the AGDs were calculated while the glandular tissues were increased by 2% from 0% to 100% in the middle domain of the standard breast phantom. The more the glandular tissues, the more the AGD gradually became. It seems that by the glandular tissues with higher density than that of adipose tissues, the absorption of relatively low energy photons increased so that the overall X-ray absorption rate increased. An Asian woman’s breast has more fiber than a Western woman’s breast, so it is said that there is great possibility of misdiagnosis since there is a lot of difficulty in reading X-ray mammography images [11]. In addition, Asian women have dense breast with relatively high glandular tissues in their lower ages of 30s and 40s. Therefore, as a result of this study, the glandular tissues of the breast have a close relationship with AGD, so it is desperately necessary to establish diagnostic reference levels for Asian women.
350
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose
5. Conclusions To overcome the limitations of the unified phantom for AGD measurement, correlation was drawn out using Monte Carlo simulation by measuring AGDs by a variety of breast thickness. Recently, international organizations like ICRP have reinforced the regulation of dose constraint by lowering it from 3 mGy to 2 mGy. Each country should verify the validity of its existing diagnostic reference levels and set up detailed technical instructions for them. This study is judged to be used as basic data to determine this guideline and target value of glandular dose. References [1]
Snydner WS, Ford MR and Warner GG. Estimates of Specific Absorbed fraction for Photon Sources Uniformly Distributed in Various Organs of a Heterogeneous Phantom. Society of Nuclear Medicine. New York: MIRD Pamphlet (5). Revised; 1978. [2] Cristy M and Eckerman KF. Specific Absorbed Fractions of Energy at Various Ages from Internal Photon Sources. 1. Methods: Appendix A. Description of the Mathematical Phantoms. Health and Safety Research Division. RNL; 1987. [3] Briesmeister Judith F. MCNP-A General Monte Carlo Code for Neutron and Photon Transport. LA7396-M; 1986. [4] Archie Bleyer, and Gilbert Welch H. Effect of Three Decades of Screening Mammography on Breast-Cancer Incidence. New England Journal of Medicine. 2012; 367(21): 1998-2005. [5] Walter Huda, Anthony Sajewicz M, Kent M Ogden, and David R Dance. Experimental investigation of the dose and image quality characteristics of a digital mammography imaging system. Medical Physics. 2003; 30(3): 442-448. [6] Mammography quality control manual Reston. Va: American College of Radiology. Committee on Quality Assurance in Mammography; 1999. [7] Fletcher E WL, Baum JD, and Draper G. The risk of diagnostic radiation of the newborn. British Journal of Radiology. 1986; 59(698): 165-270. [8] Janse JTM, Veldkamp WJH, Thijssen MAO, S van Wouenberg and Zoetelief J. Method for determination of the mean fraction of glandular tissue in individual female breasts using mammography. Physics in Medicine and Biology. 2005; 50(24): 5953-5967. [9] IPEM, Catalogue of Diagnostic X-Ray Spectra and other Data, 1997. [10] White DR, Widdowson EM, Woodard HQ, Dickerson WT. The composition of body tissues (II). Fetus to young adult. British Journal of Radiology.1991; 64(758): 149-159. [11] Lee HD, Park HB, Koo JY, Oh SM, Lim JY, Cha KH, Kim DY. Study for Mammographic Patterns of Korean Breast Cancer. Korean Journal of Breast Cancer. 1999; 2(1): 96-94.
Copyright of Technology & Health Care is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
345
Monte Carlo simulation for correlation analysis of average glandular dose by breast thickness and glandular ratio in breast tissue Sang-Tae Kima and Jung-Keun Chob,∗
a Radiation
Safety Division, Nuclear Safety and Security Commission, Seoul, Korea of Radiological Science, Jeonju University, Jeonju, Korea
b Department
Received 26 October 2013 Accepted 22 January 2014 Abstract. A glandular breast tissue is a radio-sensitive tissue. So during the evaluation of an X-ray mammography device, Average Glandular Dose (AGD) measurement is a very important part. In reality, it is difficult to measure AGD directly, Monte Carlo simulation was used to analyze the correlation between the AGD and breast thickness. As a result, AGDs calculated through the Monte Carlo simulation were 1.64, 1.41 and 0.88 mGy. The simulated AGDs mainly depend on the glandular ratio of the breast. With the increase of glandular breast tissue, absorption of low photon-energy increased so that the AGDs increased, too. In addition, the thicker the breast was, the more the AGD became. Consequently, this study will be used as basic data for establishing the diagnostic reference levels of mammography. Keywords: Glandular, average glandular dose, Monte Carlo simulation, breast thickness
1. Introduction With concern and anxiety regarding man’s radiation exposure continuing to rise after the disaster at Fukushima Nuclear Power Plant (2011), it is critical for subjects to have accurate information about their exposure doses during an X-ray test since it can help them reduce their anxiety and their radiologists or doctors improve their awareness of lower exposure doses. In particular, it is very important to establish an accurate system for assessing patient exposure doses during an mammography, which requires relatively longer radiation exposure time per test. Essential to the establishment of a more accurate patient doses assessment system is information about dose distributions by considering interactions between the major organs of the patient exposed to radiation and the radiation. Also needed to assess effective doses by multiplying each organ dose by their corresponding tissue weighting factors and adding them up is information about the dose distributions of each organ. It is difficult to directly measure the equivalent doses of various organs when the body is exposed to radiation, which is why humanoid and mathematical phantoms have been used in most patient exposure doses ∗ Corresponding author: Jung-Keun Cho, Department of Radiological Science, Jeonju University, 303, Cheonjam-ro, Wansangu, Jeonju-si, Jeollabuk-do, Korea. Tel.: +82 63 220 3129; Fax: +82 63 220 2054; E-mail: [email protected].
c 2014 – IOS Press and the authors. All rights reserved 0928-7329/14/$27.50
346
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose
assessments [1–3]. X-ray mammography, a breast cancer test method using X-ray with high contrast and resolution is the most frequently used early diagnosis of breast cancer currently. Yet, with the increasing frequency of use of mammography X-ray unit, interests in radiation injuries due to X-ray mammography are heightened [4]. Since a glandular breast tissue is a tissue with a high radio sensitivity, a carcinogenic risk in an X-ray mammography continues to be raised. In X-ray mammography test, Average Glandular Dose (AGD) measurement is an important item for the quality assessment of mammography X-ray unit [5]. The Committee on Quality Assurance in Mammography (CQAM) of the American College of Radiology (ACR) provides that the AGD is less than 3 mGy in breast thickness 4.2 cm with a breast tissue ratio of adipose tissues of 50% and glandular tissues of 50% [6]. Yet, since AGD in fact, differs depending on breast thickness, even though it does not exceed the advised degree of 3 mGy, you cannot say that the glandular dose to which the patient is exposed is less. This study attempted to draw out the correlation of AGD according to a variety of breast thickness and glandular ratio using Monte Carlo simulation. 2. Monte Carlo simulation GEANT (Geometry And Tracking) 4, the calculation code used in Monte Carlo simulation in this study is a code developed in C++, which can carry out computer simulations of microscopic phenomena for the interactions between particles and substances. In addition, it is widely used in various domains such as high energy physics, medical physics and astrophysics, and it has a variety of physical models which can be applied to interaction research on photons, electronics and hadrons. GEANT 4 provides three types of model, that is, standard, low energy and Penelope model for electromagnetic interactions. This study used the low energy model. The low energy model can apply to the energy range from 250 eV to 100 GeV, and considers the fluorescence of excited atoms including photoelectric effect, Compton scattering, Rayleigh scattering, bremsstrahlung and ionization [7]. The version of GEANT4 used for calculation in this study is 9.4.p01. The phantom simulated for the correlation analysis of the AGD by breast thickness and glandular ratio has been reported in preceding studies [8]. The standard breast phantom is wagon headed with a diameter of 160 mm and a thickness of 45 mm, and in the middle domain, adipose tissues and glandular tissues are mixed at the ratio of 50:50, and surrounded by 100% adipose tissues with a thickness of 5 mm. For the correlation analysis of the AGD by breast thickness and glandular ratio, the thickness of the standard breast phantom was simulated increasing by 1 cm from 1 cm to 10 cm. The correlation was analyzed while glandular tissues in the middle domain were increased by 2% from 0% to 100%. The X-ray spectrum applied to simulate the source of X-ray mammography device was set up using the technical documents of the X-ray shooting equipment of this study and IPEM-78 [9]. IPEM-78 is a code providing the spectrum of the emitted X-ray, with three target substances of tungsten, rhodium and molybdenum, the range of tube voltage of 30 ∼ 150 kVp, the target angle of 6 ∼ 22◦ , and information about filter materials of aluminum, beryllium, copper, air and bone, which is composed so that the user can use it adjusted to the characteristics of the relevant X-ray device. Based on Tables 1–3, the results of X-ray spectrum generation in IPEM-78 were shown in Fig. 1. To simulate it with the source closest to the continuous spectrum, the photons integrated at an interval of 0.5 kV were applied in the obtained energy spectrum. To increase efficiency by increasing the simulation speed, a geometric approximation method like the solid angle limitation of radiation source was used [10]. Point sources were assumed and approximated to a fan shape at an angle proper for each radiation field. The distance from the breast support to the focus is 605 mm and the central beam axis is perpendicular to the surface of the phantom.
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose
347
Table 1 Density and elemental ratio of compression paddle and PMMA breast phantom Density (g/cm3 ) Compression paddle PMMA breast phantom
1.2 1.19
H 0.0556 0.0805
Component C 0.7925 0.5998
O 0.1509 0.3196
Fig. 1. Result of X-ray spectrum generation in IPEM-78.
The focus is an isotropic point source, and the solid angle was calculated as an X-ray beam in a half cone-shape with the radius of 80 mm at 480 mm point from the focus (Fig. 2). 2 × 109 beam histories of the used photon were used. In X-ray mammography, the compression paddle is included in the X-ray radiation field, which may affect the AGD measurement by grade and thickness due to X-ray attenuator, so it must be simulated. The simulated compression paddle is a thickness of 2 mm. The compression paddle to apply Monte Carlo simulation, the density and elemental ratio of the PMMA breast phantom, X-ray exposure condition and HVL by tube voltage were shown in Tables 1–3. 3. Results A standard PMMA phantom with a diameter of 200 mm and a thickness of 45 mm was simulated to measure AGD. The thickness of the PMMA phantom decreased to 45, 35 and 25 mm to calculate the AGD. The AGDs calculated to the Monte Carlo simulation were 1.64, 1.41 and 0.88 mGy (Table 4). For the correlation analysis of the AGD by the simulated breast thickness and glandular ratio, a wagonheaded standard breast phantom with a diameter of 160 mm and a thickness of 45 mm was simulated. The glandular tissues of the standard breast phantom were set to 0% (100% adipose tissue), 25% (25% glandular, 75% adipose tissue), 50% (50% glandular, 50% adipose tissue), 75% (75% glandular, 25% adipose tissue) and 100% (100% glandular), respectively, and the thickness of each phantom was increased by 1 cm from 1 cm to 10 cm to calculate the AGD at each thickness (Fig. 3). The AGDs at the thicknesses of 1–10 cm of the phantom with glandular tissues of 50% was calculated at 1.18–1.57 mGy. The correlation was analyzed while the glandular tissues in the middle domain of the
348
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose Table 2 X-ray exposure condition Target Filter Tube Voltage [kV] Tube Current [mAs]
Mo Mo, Rh 25 60
Table 3 HVL set point by tube voltage provided in IEC Tube Voltage [kV] 25 28 30
1st HVL [mmAl] 0.28 0.31 0.33
Table 4 AGDs by PMMA thickness obtained through Monte Carlo simulation Thickness of PMMA Simulated phantom [mm] AGD [mGy] 45 1.64 35 1.41 25 0.88
Fig. 3. Changes in AGDs by compressed breast thickness according to the ratio of adipose tissues and glandular tissues.
Fig. 2. Geometric structure for application of Monte Carlo simulation. (Colours are visible in the online version of the article; http://dx.doi.org/10.3233/THC-140790)
Fig. 4. Correlation between glandular ratio and AGD in 45 mm standard breast phantom.
standard breast phantom of 45 mm were increased by 2% from 0% to 100%. The irradiation condition for the simulation was simulated at 29 kV and 60 mAs. When the glandular tissues were 0–100%, the AGDs were calculated at 1.23–2.04 mGy, respectively (Fig. 4). 4. Discussion In an X-ray mammography, a compression paddle should be included in the X-ray radiation field, and by grade and thickness, it occupies a part of the total capacity of the filtration as X-ray attenuator.
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose
349
Thus, since the compressor may affect measurements of the quality of radiation and AGD, it should be included in the radiation field in actual experiment or simulation. This study understood the grade and thickness of the compression paddle used in the experiment to carry out accurate simulation, which was included in the radiation field. The half-value-layer measured to calculate AGD is an index showing the quality of radiation characteristics for kVp set up regularly, and using the kVp used in clinics, it was measured for the beam that penetrated the compression paddle. Since the quality of radiation greatly affects the contrast, and the AGD calculated from the incident air Kerma differs depending on HVL, the HVL measurement is an important element to determine the AGD of a digital mammography system. Mammography Quality Standards Act (MQSA) provided in Part 900 of the Code of Federal Regulations regulates the HVL in 30 kVp at least 0.3 mmAl for an Mo or Mo-W alloy target. The value provided in the IEC was used for the first half-value-layer of the system used in this study. The second half-valuelayer is defined as the difference between the Al attenuator thickness needed to reduce the glandular dose to 1/4 and the first HVL while the homogeneity coefficient is defined as the ratio of the first HVL and the second HVL, to which 0.8 was applied. The homogeneity coefficient is an index showing the width of an X-ray spectrum, calculated at 0 to 1, and the greater the value, the narrow the spectrum becomes. The homogeneity coefficient used in diagnostic radiology is 0.7 ∼ 0.9. To enhance the precision of measurement of quality of radiation, it is important to establish a standardized energy correction table on the half-value-layer 0.3 ∼ 0.4 mmAl equivalent beam. The AGDs in most cases of actual calculation turned out to be less than 3 mGy, which was provided in the International Standard Regulations. The additional magnification view mammography and Stereotactic mammography belong to the test in the diagnosis domain, which might not be treated the same as the radiation injury, but it is a glandular dose which cannot be ignored if the exposure of the domain in which relatively high glandular dose is received. In addition, AGDs differ depending on breast thickness, so even if it does not exceed the advised degree of 3 mGy, you cannot say that the glandular dose to which the patient is exposed is less. In the simulation, as a result of the calculation of AGD while the thickness of the standard breast phantom was increased by 1 cm from 1 cm to 10 cm, it was found that the value of AGD gradually increased with the increase of breast thickness. The thicker the breast, the smaller the difference in the absorption between structures gets by the beam hardening effect, which lowers the subject contrast. In addition, the influence of the scattered rays lowers the subject contrast and increases the glandular dose exposed to the patient, too. To minimize this, the breast is pressed in X-ray mammography. Pressing the breast firmly can reduce the scattered rays and minimize the blurring by movement. For the young women with a low weight, the ratio of dense breast is relatively high, so in X-ray mammography, it involves the patients’ pain. Thus, efforts to minimize the patients’ pain with proper press are necessary. In the simulation, the AGDs were calculated while the glandular tissues were increased by 2% from 0% to 100% in the middle domain of the standard breast phantom. The more the glandular tissues, the more the AGD gradually became. It seems that by the glandular tissues with higher density than that of adipose tissues, the absorption of relatively low energy photons increased so that the overall X-ray absorption rate increased. An Asian woman’s breast has more fiber than a Western woman’s breast, so it is said that there is great possibility of misdiagnosis since there is a lot of difficulty in reading X-ray mammography images [11]. In addition, Asian women have dense breast with relatively high glandular tissues in their lower ages of 30s and 40s. Therefore, as a result of this study, the glandular tissues of the breast have a close relationship with AGD, so it is desperately necessary to establish diagnostic reference levels for Asian women.
350
S.-T. Kim and J.-K. Cho / Monte Carlo simulation for correlation analysis of average glandular dose
5. Conclusions To overcome the limitations of the unified phantom for AGD measurement, correlation was drawn out using Monte Carlo simulation by measuring AGDs by a variety of breast thickness. Recently, international organizations like ICRP have reinforced the regulation of dose constraint by lowering it from 3 mGy to 2 mGy. Each country should verify the validity of its existing diagnostic reference levels and set up detailed technical instructions for them. This study is judged to be used as basic data to determine this guideline and target value of glandular dose. References [1]
Snydner WS, Ford MR and Warner GG. Estimates of Specific Absorbed fraction for Photon Sources Uniformly Distributed in Various Organs of a Heterogeneous Phantom. Society of Nuclear Medicine. New York: MIRD Pamphlet (5). Revised; 1978. [2] Cristy M and Eckerman KF. Specific Absorbed Fractions of Energy at Various Ages from Internal Photon Sources. 1. Methods: Appendix A. Description of the Mathematical Phantoms. Health and Safety Research Division. RNL; 1987. [3] Briesmeister Judith F. MCNP-A General Monte Carlo Code for Neutron and Photon Transport. LA7396-M; 1986. [4] Archie Bleyer, and Gilbert Welch H. Effect of Three Decades of Screening Mammography on Breast-Cancer Incidence. New England Journal of Medicine. 2012; 367(21): 1998-2005. [5] Walter Huda, Anthony Sajewicz M, Kent M Ogden, and David R Dance. Experimental investigation of the dose and image quality characteristics of a digital mammography imaging system. Medical Physics. 2003; 30(3): 442-448. [6] Mammography quality control manual Reston. Va: American College of Radiology. Committee on Quality Assurance in Mammography; 1999. [7] Fletcher E WL, Baum JD, and Draper G. The risk of diagnostic radiation of the newborn. British Journal of Radiology. 1986; 59(698): 165-270. [8] Janse JTM, Veldkamp WJH, Thijssen MAO, S van Wouenberg and Zoetelief J. Method for determination of the mean fraction of glandular tissue in individual female breasts using mammography. Physics in Medicine and Biology. 2005; 50(24): 5953-5967. [9] IPEM, Catalogue of Diagnostic X-Ray Spectra and other Data, 1997. [10] White DR, Widdowson EM, Woodard HQ, Dickerson WT. The composition of body tissues (II). Fetus to young adult. British Journal of Radiology.1991; 64(758): 149-159. [11] Lee HD, Park HB, Koo JY, Oh SM, Lim JY, Cha KH, Kim DY. Study for Mammographic Patterns of Korean Breast Cancer. Korean Journal of Breast Cancer. 1999; 2(1): 96-94.
Copyright of Technology & Health Care is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
