Digital breast tomosynthesis: comparison of different methods to calculate patient doses Poster No.: C-2220 Congress: ECR 2011 Type: Scientific Paper Authors: A. Jacobs 1, L. Cockmartin 1, D. R. Dance 2, H. Bosmans 1 ; 1 Leuven/ BE, 2 Guildford/UK Keywords: Breast DOI: 10.1594/ecr2011/C-2220 Any information contained in this pdf file is automatically generated from digital material submitted to EPOS by third parties in the form of scientific presentations. References to any names, marks, products, or services of third parties or hypertext links to thirdparty sites or information are provided solely as a convenience to you and do not in any way constitute or imply ECR's endorsement, sponsorship or recommendation of the third party, information, product or service. ECR is not responsible for the content of these pages and does not make any representations regarding the content or accuracy of material in this file. As per copyright regulations, any unauthorised use of the material or parts thereof as well as commercial reproduction or multiple distribution by any traditional or electronically based reproduction/publication method ist strictly prohibited. You agree to defend, indemnify, and hold ECR harmless from and against any and all claims, damages, costs, and expenses, including attorneys' fees, arising from or related to your use of these pages. Please note: Links to movies, ppt slideshows and any other multimedia files are not available in the pdf version of presentations. www.myesr.org Page 1 of 8
Purpose Digital breast tomosynthesis (DBT) has been proposed as a promising new development in the field of breast imaging. This pseudo three-dimensional (3D) technique provides images with reduced structure noise from overlaying tissues. While clinical studies are ongoing to evaluate the clinical performance of DBT systems, the associated radiation dose should also be examined. It is important to survey the dose delivered to the patient especially when it is aimed to use DBT systems in a breast screening program. Methods and Materials Data collection was performed on two systems at the University Hospital in Leuven, a Siemens MAMMOMAT Inspiration TOMO breast tomosynthesis system (Siemens, Erlangen, Germany) and a Hologic Selenia Dimensions tomosynthesis system (Hologic, Bedford, USA). Data from 100 clinical cases were collected for both medio-lateral-oblique (MLO) and cranio-caudal (CC) views on the Siemens tomosynthesis system. On the Hologic Dimensions, data of only 18 clinical cases were available. The average glandular dose (AGD) within the patient's breast was calculated using different approaches. The first method is described by Dance et al [1-3], and the following formula is used: AGD = Kgcs#t(#) (1) where K is the incident air kerma for a single projection, measured at zero degree projection angle and the remaining factors were obtained from Monte Carlo simulations. The g factor is the incident air kerma to average glandular dose conversion factor for 50% glandularity breasts, the c factor allows a correction for different glandularities and the s factor corrects for the use of different spectra. These factors are the same as used for 2D breast dose calculations. t(#) is the tomo-factor which is dependent on the projection angle # measured from the center of rotation. For each patient, we calculated a c- factor both for the glandularity as observed in the patient and for an age based glandularity estimate. The second method is described by Sechopoulos et al [4-5]. A different formula is used for the dose calculations: AGD = X CR D g N 0 #RGD(#) (2) Page 2 of 8
where AGD is the average glandular dose, X CR is the 'free in air' exposure of a single projection measured at a reference point on the breast support plate (expressed in röntgen) without the compression plate. This reference point is the point where the central x-ray meets the support plate. D g N 0 is the normalized glandular dose per unit exposure for the zero degree projection and RGD(#) is the relative glandular dose coefficient which is calculated for every tomosynthesis projection angle (#) measured from the support plate. The different methods were used to calculate the doses of the two x-ray systems. In a first step, the different methods were compared. Then, the calculated AGD values were compared with the doses indicated on the tomosynthesis systems. Finally, the doses from the tomosynthesis acquisition were compared with 2D doses from a large patient dosimetry study. Results 1. Comparison of different methods Doses were calculated once using the Dance et al method based on individual breast glandularity and once based on the Sechopoulos et al method. The ratio between those two AGD values was made. System 1 System 2 All CC MLO All CC MLO Mean ratio 1.23 1.15 1.27 1.13 1.10 1.14 Standard deviation 0.244 0.014 0.293 0.088 0.048 0.099 Maximum deviation 0.976 0.053 0.937 0.253 0.101 0.243 Table 1: Mean ratio of AGD values calculated by the Dance et al method based on breast glandularity and Sechopoulos' method, for all views and CC and MLO view separately. Also the standard deviation and maximum deviation from the mean value are shown. Table 1 shows a difference of 23 % for the Siemens system (system 1) and 13% for the Hologic system (system 2), which is mainly due to differences in the breast model and the fact that Sechopoulos et al method takes the heel effect into account and Dance et al method does not. The mean ratio for the CC view is 1.15 for system 1 and 1.10 for Page 3 of 8
system 2 with a maximum deviation from the mean of 5% for the first system and 10% for the second. 2. Comparison with dose indicated on the system The dose indicated on the system is compared with the calculated AGD values in order to know whether the dose indicated on the system is a good estimation of the doses calculated with the methods described by Dance and Sechopoulos. The results of the Dance et al method based on age was the closest to the system AGD (fig. 1). For both systems, the method of Dance et al based on age provided the best results with only a difference of 5 % between the indicated and the calculated value for the Siemens tomosynthesis system and a 12% difference for the Hologic Dimensions. The differences are larger when the other methods are used. 3. Comparison of AGD values for 2D mammography and breast tomosynthesis On the Siemens tomosynthesis system, a comparison was made between the calculated 3D AGD values and the 2D dose values from a large patient dosimetry study. 2D doses were calculated for about 1000 patients collected on the same system, during the same period of data collection of the DBT patient doses. The 2D dose calculation is based on the formula of Dance et al with an c factor for a screening population and therefore the same method is used for the 3D doses. Fig. 2 shows both the 2D and 3D dose values as a function of the compressed breast thickness. Table 2 presents an overview of the 2D and 3D AGD values for different ranges of compressed breast thicknesses. From this table, it can be seen that, on the Siemens system, doses in one view DBT are almost doubled compared to doses in one view 2D digital mammography. Compressed thickness (mm) breast 2D AGD (mgy) 20-30 0.58-31 - 40 0.70 1.32 41-50 0.86 1.67 51-60 1.11 1.84 61-70 1.29 2.07 71-80 1.57 2.59 > 80 1.90 2.64 3D AGD (mgy) Table 2: Overview of 2D and 3D AGD values for different ranges of compressed breast thickness. Page 4 of 8
Images for this section: Fig. 1: Comparison between AGD indicated on the system and AGD calculated based on the age of the patient. Page 5 of 8
Fig. 2: Comparison between 2D and 3D doses in function of compressed breast thickness. The blue dots represent the 2D AGD values and the red squares are the 3D dose values based on the age of the patient. Also the 2D acceptable and achievable dose levels are shown. Page 6 of 8
Conclusion The methods proposed to calculate the AGD for breast tomosynthesis were applied and compared on the available dataset in our hospital. It is possible to calculate this AGD in a similar way as for 2D mammography using different methods that could be considered relatively easy and applicable on a larger dataset. There is a small but significant difference between the two methods which can be explained by the different breast models used in the simulations and by the fact that the heel effect was taken into account in Sechopoulos' method but not in the method of Dance et al. The calculated dose values were compared with the system AGD. A good agreement was found between the latter method based on the age of the patient and the indicated dose for both tomosynthesis systems with only a bias of 5% in case of Siemens and 12% in case of Hologic. Finally, we showed that, on the Siemens system, doses in one view DBT are almost double of the doses in one view 2D digital mammography. References [1] D R Dance, C L Skinner, K C Young, J R Beckett and C J Kotre, Additional factors for the estimation of mean glandular breast dose using the UK mammography dosimetry protocol, Phys. Med. Biol. 45 (2000) 3225-3240. [2] D R Dance, K C Young and R E van Engen, Further factors for the estimation of mean glandular dose using the United Kingdom, European and IAEA breast dosimetry protocols, Phys. Med. Biol. 54 (2009) 4361-4372. [3] D R Dance, K C Young and R E van Engen, Estimation of mean glandular breast dose for digital tomosynthesis: factors for use with the UK, European and IAEA breast dosimetry protocols. Phys. Med. Biol. 56 (2011) 453-471. [4] I Sechopoulos, S Suryanarayanan, S Vendantham, C J D'Orsi and A Karellas, Computation of the glandular radiation dose in digital tomosynthesis of the breast, Med. Phys. 34 (1), January 2007, 221-232. [5] I Sechopoulos and C J D'Orsi, Glandular radiation dose in tomosynthesis of the breast using tungsten targets, Journal of Applied Clinical Medical Physics, Volume 9, Number 4, Fall 2008, 161-171. Page 7 of 8
Personal Information This work was funded by the OPTIMAM project which is supported by CR-UK & EPSRC Cancer Imaging Programme, in association with the MRC and Department of Health (England). Page 8 of 8