Radiation Dose Estimations to the Thorax Using Organ-Based Dose Modulation

Cardiopulmonary Imaging Original Research Lungren et al. Organ-Based Dose Modulation Cardiopulmonary Imaging Original Research Matthew P. Lungren 1 Terry T. Yoshizumi 1 Samuel M. Brady 1 Greta Toncheva 1 Colin Anderson-Evans 1 Carolyn Lowry 1 Xiaodong R. Zhou 2 Donald Frush 1 Lynne M. Hurwitz 1 Lungren MP, Yoshizumi TT, Brady SM, et al. Keywords: breast, CT, dose, organ-based dose modulation, thoracic CT DOI:10.2214/AJR.11.7798 Received August 27, 2011; accepted after revision November 8, 2011. D. Frush and L. M. Hurwitz received funding from Siemens Healthcare. 1 Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC 27710. Address correspondence to M. P. Lungren (matthew.lungren@duke.edu). 2 Siemens Medical Solutions, Malvern, PA. WEB This is a Web exclusive article. AJR 2012; 199:W65 W73 0361 803X/12/1991 W65 American Roentgen Ray Society Radiation Dose Estimations to the Thorax Using Organ-Based Dose Modulation OBJECTIVE. The purpose of this study was to assess the radiation dose distribution and image quality for organ-based dose modulation during adult thoracic MDCT. MATERIALS AND METHODS. Organ doses were measured using an anthropomorphic adult female phantom containing 30 metal oxide semiconductor field-effect transistor detectors on a dual-source MDCT scanner with two protocols: standard tube current modulation thoracic CT and organ-based dose modulation using a 120 radial arc. Radiochromic film measured the relative axial dose. Noise was measured to evaluate image quality. Breast tissue location across the anterior aspect of the thorax was retrospectively assessed in 100 consecutive thoracic MDCT examinations. RESULTS. There was a 17 47% decrease (p = < 0.05) in anterior thoracic organ dose and a maximum 52% increase (p = < 0.05) in posterior thoracic organ dose using organ-based dose modulation compared with tube current modulation. Effective dose (SD) for tube current modulation and organ-based dose modulation were 5.25 ± 0.36 msv and 4.42 ± 0.30 msv, respectively. Radiochromic film analysis showed a 30% relative midline anterior-posterior gradient. There was no statistically significant difference in image noise. Adult female breast tissue was located within an average anterior angle of 155 (123 187 ). CONCLUSION. Organ-based dose modulation CT using an anterior 120 arc can reduce the organ dose in the anterior aspect of the thorax with a compensatory organ dose increase posteriorly without impairment of image quality. Laterally located breast tissue will have higher organ doses than medially located breast tissue when using organ-based dose modulation. The benefit of this dose reduction must be clinically determined on the basis of the relationship of the irradiated organs to the location of the prescribed radial arc used in organ-based dose modulation. A ssessment of radiation dose to individual organs during diagnostic thoracic MDCT has been primarily focused on the breast, lungs, and bone marrow because of their relative high sensitivity to radiation compared with other organs in the thorax [1 5]. Radiation exposure to the breasts of young women has been of particular concern with current MDCT techniques, with absorbed breast doses reported in the range of 4.3 66 mgy [2, 4, 5]. Several available techniques have been shown to reduce radiation dose during MDCT imaging, including reduction in tube output, reduction in tube voltage, reduction in z-axis coverage, and variation in pitch [6 11]; reported dose reduction for these techniques ranges from 22% to 67% with variable effects on image noise. These techniques, however effective, lack specificity for targeted dose re- duction of radiosensitive organs included in the FOV, such as breast tissue. Currently, the only available organ-specific technique aimed at reducing the absorbed breast dose during thoracic MDCT is the use of bismuth shields, which has shown maximal dose reduction to the breast in the range of 43 47% [5, 12]. Tube current modulation software uses a constant image quality factor to change the tube output in the plane (x, y, and z) of the patient, depending on variations in body habitus and body size and has been shown to reduce radiation dose while maintaining constant image quality [11]. A novel form of tube current modulation has recently been developed that, in addition to modulation along the z-axis, modulates the beam in the x- and y-plane output for a constant 120 of the gantry rotation and correspondingly increases the tube output over the remaining 240 of the gantry W65

Lungren et al. rotation so that image quality is maintained centrally within the body. As a result of this technique, known as organ-based dose modulation, centrally located organs have a constant radiation exposure while superficially located organs experience reduced exposure within the prescribed 120 radial arc and increased exposure throughout the remaining 240 radial arc. Organ-based dose modulation can be specified to apply the region of the 120 radial arc of reduced tube current along the anterior, posterior, or lateral aspects of the beam relative to patient position (supine, prone, or lateral), and although the location of the dose reduction area can be changed according to patient orientation, the arc of the dose reduction remains constant. Relative to constant tube current, organ-based dose modulation has been shown to reduce tube current by 75% in the anterior 120 radial arc and increase tube current by 25% during the remainder of the gantry rotation, with variation in the ramp-up and ramp-down time depending on the gantry rotation time [13]. Organ-based dose modulation has been proposed as a method to reduce radiation exposure to the breast during thoracic MDCT, with maintenance of overall radiation exposure to the thorax and without impairing image noise. The purpose of this study was to evaluate absorbed organ dose, effective dose, radiation dose distribution, and image noise for organ-based dose modulation during adult thoracic MDCT imaging. Materials and Methods Phantom An adult anthropomorphic phantom (model 701- D, CIRS) with breast attachments (no extremities) was used that has been validated for human organ dosimetry measurements [4]. The specific dimensions of the phantom are as follows: weight, 73 kg; length, 173 cm; and thorax, 23 32 cm. The phantom includes bone, lung, brain, and soft-tissue compositions (Fig. 1) and is subdivided into 39 contiguous 2.5-cmthick sections. Each section contains several 5-mmdiameter holes through which detectors are placed for organ dose measurements. The phantom comes with assignable anatomic locations, and each hole location is optimized for precise dosimetry within a specific internal organ that can be referenced to the manufacturer s user manual. Tissue-equivalent plugs fill the holes when they are not being used. The breasts are attachable to the main body and are made of a 50:50 glandular-to-adipose formulation. The breasts for this phantom assume an erect and symmetric shape and are 5.5 cm in height and 13 cm in diameter at the base. Specific holes are located in the breasts 1 cm below the skin surface for detector placement. Organ Dose Organ dose measurements were performed using 30 high-sensitivity metal oxide semiconductor fieldeffect transistor (MOSFET) (Model TN-1002RD-H, Best Medical) dosimeters. MOSFET detectors were inserted into the predrilled holes of the anthropomorphic phantoms and distributed into the following locations listed in Table 1: skin left axilla, skin center left breast, skin anterior midline, skin center right breast, skin right axilla, posterior right skin, posterior left skin, anterior left lung, anterior right lung, posterior left lung, posterior right lung, upper thoracic spine, lower thoracic spine, sternum, anterior lateral left ribs, anterior lateral right ribs, posterior left ribs, posterior right ribs, left lateral breast, left medial breast, right lateral breast, right medial breast, esophagus, thymus, thyroid, liver, kidneys, intestine, uterus, and stomach. The MOSFET reader was connected to a laptop computer, and the data were read immediately after each CT examination. The software (model TN-RD-60 and model TN-RD-70-W, Best Medical) stored acquired data in centigrays. The phantom was placed in a supine position on the CT bed and centered within the gantry and scanned three times per protocol. Organ doses were averaged to provide a mean value and SD. Relative Axial Dose Distribution Radiochromic film (XRQA, International Specialty Products) that has been validated for use in radiation dosimetry was used; change in optical density along the film after exposure to ionizing radiation is linearly proportional to the amount of radiation absorbed (the reported range of detection is 0.1 20 cgy) [14]. The film was placed between two sections of the anthropomorphic phantom at the level of the breasts such that the plane of the film was oriented parallel to the image plane to measure radiation dose distribution along the axial plane. After exposure, the radiochromic films were digitized using an XL10000 flatbed scanner (Epson) to determine the optical density (OD) [14 19]. To remove any optical scanner induced artifacts, the net OD (netod) was calculated as the difference of the OD of the film before and after exposure to ionizing radiation. NetOD is determined by the log transformation of the ratio of preirradiation (I pre ) and postirradiation (I post ) scanner light intensity values [20]: I 0 netod = OD post OD pre = I log 0 10 log I 10 post I post = log 10 I pre I post MDCT Protocols A dual-source 64-MDCT scanner (Somatom Definition, Siemens Healthcare) was used for all experiments. All scanning was performed using a single-tube mode of scanning. Radiation dose distribution and absorbed organ doses for organbased dose modulation were compared with our standard clinical thoracic CT protocol (tube current modulation) that uses real-time tube current modulation in the x-, y-, and z-axes (Caredose 4D, Siemens Healthcare) and a pitch of 0.8. The organ-based dose modulation software evaluated (X- CARE, Siemens Healthcare) has a radial beam reduction arc of 120 and a fixed pitch of 0.6 and, for the purposes of this study, the radial beam reduction arc was prescribed to be anteriorly located over the phantom. To provide comparative assessment, the overall image quality was set to be of the same value for the tube current modulation and the organ-based dose modulation protocols, using a quality reference of 150 mas for all examinations. The quality reference tube current is the value used by this manufacturer to provide equivalent image quality between protocols and patients and is based on a 70-kg male. The specifics of the protocols are listed in Table 2. The z-axis coverage for each protocol was constant, extending from the thoracic inlet to the diaphragm. The phantom was placed in the isocenter of the gantry according to laser calipers and a large scanning FOV of 50 cm was used for all protocols. The volume CT dose index (CTDI vol ) and dose-length product were recorded from the CT console at the time of scanning for each protocol before and after each acquisition. The scanning time for each protocol was recorded. Image Quality Image quality was quantitatively assessed within the anthropomorphic phantom by measuring the difference in noise in specific anatomic locations in the thorax. Three separate 100-mm 2 circular regions of interest were placed in the anterior mediastinum in the midline, in the posterior soft tissues in the midline, and centrally within the mediastinum in identical locations for each examination (Fig. 2). The SD of the mean attenuation value (in Hounsfield units) was recorded as the noise level and the results were averaged for each location. Images were qualitatively assessed by two fellowship-trained radiologists with 13 and 25 years of experience reading thoracic CT. The axial images from the tube current modulation and the organbased dose modulation protocols were placed side by side on the viewing console (window and level settings of 440 and 40, respectively), and by consensus read the two radiologists, blinded to the protocols, evaluated the image quality for differences in noise and streak artifact (from MOSFET detectors). Images were graded for the relative amount of noise and streak artifact for the organ-based dose modulation dataset compared with the tube current modulation protocol on a 5-point scale. A score of 5 represented moderately better image quality with W66

Organ-Based Dose Modulation Fig. 1 Adult female anthropomorphic phantom with breast attachments. A, Photograph shows phantom with breast attachments (left), with cross-sectional view at level of breasts where single radiochromic film was placed for axial dose distribution determination (right). B E, Images show locations of 30 metal oxide semiconductor field effect transistor detectors (indicated by gray ovals): thyroid (B, top left); thymus (B, top right); sternum bone marrow, anterior and posterior right and left lung (B, bottom left); superior thoracic spine bone marrow (B, bottom right); lateral and posterior right and left rib bone marrow (C, top left); esophagus (C, top right); liver (C, bottom left); stomach and kidney (C, bottom right); inferior thoracic bone marrow (D, top left); uterus (D, top right); upper lower intestine (D, bottom left); skin paraspinal right and left (D, bottom right); medial and lateral right and left breast (E, top); and skin right axilla, top of right breast, anterior midline, top of left breast, and left axilla (E, bottom). Cross-sectional appearance of lungs (white), mediastinum (light gray), bone (dark gray), and breasts (attached anteriorly) is illustrated. (Fig. 1 continues on next page) moderately less noise and streak artifact than the tube current modulation protocol. A score of 4 represented slightly better image quality with slightly less noise and streak artifact than the tube current modulation protocol. A score of 3 represented no significant difference in image quality between the two datasets. A score of 2 represented slightly inferior image quality with mildly more noise and streak artifact than the tube current modulation protocol. A score of 1 represented moderately inferior image quality with moderately more noise and streak artifact than the tube current modulation protocol. This was performed three times in total so that each of the three datasets acquired using organ-based dose modulation was compared with the three datasets acquired using tube current modulation. Effective Dose Effective dose was calculated as the sum of measured organ doses multiplied by the organspecific tissue weighting factor as noted in International Commission on Radiologic Protection (ICRP), publication 103 [1]. Statistical Analysis Percentage organ dose reduction for the organbased dose modulation protocol was calculated as the difference in the absorbed organ dose for the organ-based dose modulation protocol divided by the absorbed organ dose for the tube current modulation protocol. Statistical analyses were performed using GraphPad Prism, version 4.03 software (GraphPad Software). To study the organ dose distribution within the 120 arc when using organ-based dose modulation and the compensatory absorbed dose increase posteriorly, the organ A B C W67

Lungren et al. locations within the thorax were grouped into anterior, posterior, lateral, and internal groups. The differences in the absorbed radiation dose for the organs within each of these groups were compared for the tube current modulation and the organ-based dose modulation protocols using an unpaired Student t test; a two-tailed p value of less than 0.05 was considered statistically significant D E Fig. 1 (continued) Adult female anthropomorphic phantom with breast attachments. B E, Images show locations of 30 metal oxide semiconductor field effect transistor detectors (indicated by gray ovals): thyroid (B, top left); thymus (B, top right); sternum bone marrow, anterior and posterior right and left lung (B, bottom left); superior thoracic spine bone marrow (B, bottom right); lateral and posterior right and left rib bone marrow (C, top left); esophagus (C, top right); liver (C, bottom left); stomach and kidney (C, bottom right); inferior thoracic bone marrow (D, top left); uterus (D, top right); upper lower intestine (D, bottom left); skin paraspinal right and left (D, bottom right); medial and lateral right and left breast (E, top); and skin right axilla, top of right breast, anterior midline, top of left breast, and left axilla (E, bottom). Cross-sectional appearance of lungs (white), mediastinum (light gray), bone (dark gray), and breasts (attached anteriorly) is illustrated. with 95% confidence. The mean noise values for the three locations in the thorax (anterior, posterior, and central) for the tube current modulation and organ-based dose modulation protocols were compared using an unpaired Student t test; a twotailed p value of less than 0.05 was considered statistically significant. Clinical Breast Tissue Location One hundred consecutive thoracic CT examinations on adult women that were performed between January 1, 2009, and January 1, 2010, were retrospectively reviewed for localization of the breast tissue across the anterior aspect of the thorax. Waiver of consent was obtained by the local institution review board for this portion of the study. Women who had undergone mastectomy or other breast surgery, wore a bra or other supportive clothing, or had breast tissue extending outside the FOV were excluded. The center of the FOV was identified on each scan and an angle drawn to include the visible breast tissue with the vertex of the angle being the center of the FOV (Fig. 3). The breast tissue was defined as containing fatty and soft-tissue elements and superficially defined by surface contour change. Results Organ Dose The highest organ doses with organ-based dose modulation were to the thyroid (10.45 mgy), posterior right rib (9.83 mgy), and right paraspinal skin (10.09 mgy). The highest average absorbed organ doses for tube current modulation protocol were to the thyroid (12.60 mgy), lateral left rib (12.37 mgy), and medial left breast (10.85 mgy) (Table 1). Assessment of relative change in radiation dose by location within the thorax for organ-based dose modulation compared with tube current modulation showed a statistically significant (p = 0.014) reduction to the anterior organs (thyroid, sternum, breast left me- W68

Organ-Based Dose Modulation TABLE 1: Absorbed Organ Dose for Tube Current Modulation and Organ-Based Dose Modulation MDCT Protocols Tube Current Modulation Protocol Organ-Based Dose Modulation Protocol Organ or Location Organ Dose (mgy) ± SD Organ Dose (mgy) ± SD Thyroid 12.6 ± 3.1 10.5 ± 1.5 BM left lateral rib 12.4 ± 4.7 7.9 ± 0.8 Breast left medial 10.9 ± 0.1 6.0 ± 0.5 BM right lateral rib 10.5 ± 1.6 7.2 ± 1.6 Liver 10.5 ± 1.0 4.5 ± 1.2 Skin anterior midline 9.8 ± 2.0 5.2 ± 0.8 Breast right lateral 9.5 ± 1.1 6.8 ± 0.5 Skin left axilla 9.1 ± 0.4 9.0 ± 0.4 Lung anterior right 8.8 ± 0.9 8.7 ± 1.7 Lung posterior right 8.6 ± 0.6 8.0 ± 0.2 Breast right medial 8.6 ± 0.4 6.9 ± 0.5 Thymus 8.4 ± 1.5 7.8 ± 1.3 Lung posterior left 8.3 ± 2.4 6.8 ± 1.8 Esophagus 8.0 ± 0.4 8.5 ± 0.6 Sternum 7.8 ± 1.2 6.5 ± 1.2 Skin center left breast 7.7 ± 0.9 5.8 ± 0.7 Skin center right breast 7.5 ± 1.7 5.0 ± 0.5 Skin left paraspinal 7.5 ± 0.9 9.2 ± 0.5 Skin right axilla 7.4 ± 0.8 8.1 ± 1.5 Stomach 7.3 ± 0.5 5.8 ± 1.6 Kidney 7.3 ± 0.6 7.4 ± 0.8 Skin right paraspinal 7.2 ± 1.3 10.1 ± 0.5 Lung anterior left 7.0 ± 1.4 6.2 ± 1.8 Breast left lateral 6.8 ± 0.7 6.2 ± 0.6 BM right posterior rib 6.5 ± 2.2 9.8 ± 1.5 BM left posterior rib 6.4 ± 1.3 9.1 ± 0.9 BM upper thoracic spine 6.3 ± 0.6 6.1 ± 0.3 BM lower thoracic-upper abdominal spine 2.2 ± 1.3 3.9 ± 1.5 Intestine-upper lower intestine 0 ± 0 0 ± 0 Uterus 0 ± 0 0 ± 0 Note Data are ± SD. BM= bone marrow. TABLE 2: Parameters for Tube Current Modulation and Organ-Based Dose Modulation MDCT Protocols Protocol Peak Beam Energy (kvp) Quality Reference Tube Current (mas) Effective Tube Current (mas) Pitch Tube Rotation (s) FOV (mm) CTDl vol (mgy) DLP (mgy cm) Scanning Time (s) Tube current modulation 120 150 106 0.8 0.5 340 7.71 226 10.58 Organ-based dose modulation 120 150 106 0.6 0.5 340 7.69 240 13.91 Note CTDI vol = volume CT dose index, DLP = dose-length product. dial, breast right medial, skin left breast central, skin anterior midline, and skin right breast central) of the thorax and a statistically significant (p = 0.024) increase in dose to the posterior organs (bone marrow [BM] left posterior rib, BM right posterior rib, BM upper thoracic spine, skin right paraspinal, and skin left paraspinal) of the thorax when using organ-based dose modulation compared with tube current modulation. The range in dose reduction for organ-based dose modulation compared with tube current modulation for the anterior organ doses was 17 47%, and the maximal dose increase for organ-based dose modulation compared with tube current modulation for the posterior organ doses was up 52%. The internal organs and the lateral organs did not show a statistically significant change in absorbed radiation dose (Table 3) for organ-based dose modulation compared with tube current modulation (p > 0.1) W69

Lungren et al. Relative Axial Dose Distribution Radiochromic film analysis showed that there was a gradual change in radiation distribution in the midline of the axial plane from the anterior-to-posterior direction for organ-based dose modulation compared with tube current modulation of approximately 30% (Fig. 4), A C Fig. 2 Axial CT image at the level of the breast in anthropomorphic phantom with region of interest (circles) locations used for noise measurement in anterior mediastinum, central mediastinum, and posterior chest wall (paraspinal) midline soft tissues. with a peak decrease in dose of 40% anterior in the midline and a peak relative increase in dose posterior in the midline of 30%. Image Quality There was no significant change in image noise centrally within the thorax between the Fig. 3 Breast tissue angle. A, Axial image from anthropomorphic phantom shows superimposed 122º anterior radial arc. B and C, Axial images of the thorax at level of breast from two patients in clinical cohort show 135º (B) and 177º (C) anterior radial arc to cover breast tissue. organ-based dose modulation and tube current modulation protocols. There was no significant difference in noise between anterior mediastinum and the right paraspinal musculature within each protocol or between the protocols (Table 4). For the qualitative assessment, a score of 3 was given for each review, indicating no perceivable difference in image quality between the organ-based dose modulation and tube current modulation datasets. Effective Dose Calculated effective dose was 5.25 ± 0.36 msv and 4.42 ± 0.3 msv, respectively, for the organ-based dose modulation protocol and the tube current modulation protocol. Clinical Breast Tissue Location The average patient age for the clinical cohort was 48 years (age range, 17 82 years). The B W70

Organ-Based Dose Modulation TABLE 3: Percentage Decrease in Absorbed Dose for Anteriorly, Laterally, Posteriorly, and Centrally Located Organs of the Thorax for Organ- Based Dose Modulation Compared With Tube Current Modulation Organ average angle found to contain all breast tissue was 155 (range, 123 187 ). The minimum arc that would be needed to contain all the breast tissue in 90% of the patients was 175. Discussion Organ-based dose modulation is a new technique for MDCT scanning that reduces the tube current over a predefined region of the body and is of specific interest in thoracic imaging to reduce absorbed breast dose. This study examined the absorbed radiation dose to organs of the thorax as well as the relative dose distribution within the axial plane through the thorax for clinically available Percentage Decrease in Dose Anterior organs (p = 0.014) Thyroid 17 Sternum 17 Breast left medial 45 Breast right medial 29 Skin left breast central 25 Skin anterior midline 47 Skin right breast central 33 Lateral organs (p = 0.123) BM left lateral rib 36 BM right lateral rib 31 Breast left lateral 9 Breast right lateral 19 Skin left axilla 1 Skin right axilla 10 Posterior organs (p = 0.024) BM left posterior rib -43 BM right posterior rib 52 BM upper thoracic spine 3 Skin right paraspinal 40 Skin left paraspinal 23 Internal organs (p = 0.304) Lung anterior left 12 Lung anterior right 2 Lung posterior left 18 Lung posterior right 7 Esophagus 6 Thymus 7 Note p value of < 0.05 is considered statistically significant. BM = bone marrow. organ-based dose modulation software that uses a radial arc of 120. Organ-based dose modulation was designed for use in the thorax to reduce radiation dose to the breast. Our results show that organ-based dose modulation can reduce the absorbed dose to the medial aspect of the breast to a similar degree as that reported when using bismuth breast shields [5, 12]. The average breast dose reduction for organbased dose modulation compared with standard thoracic CT using tube current modulation was approximately 37%, although this reduction was largely due to reduction at the medial breast location that was contained within the central aspect of the anterior 120 radial arc. Breast tissue located more laterally within the 120 radial arc recorded a smaller reduction in absorbed organ dose of 9 19%. The medially and anteriorly located organs (skin in the anterior midline, medial breasts, sternum, thyroid, and skin over the center of breast) as a group had a statistically significant dose reduction, with a maximum dose reduction of up to 47%; all of these measurements were taken within the central portion of the 120 radial arc (Table 3). Corresponding to the reduction in the anteriorly located organs, the posteriorly located organs (left posterior rib, right posterior rib, upper thoracic spine, and skin the right and left paraspinal locations) as a group had an increase in absorbed radiation that was statistically significant, with a maximal dose increase of 52%; this correlates to an increase in beam output posteriorly over the thorax to compensate for the decreased tube current output anteriorly. Optical density gradients from radiochromic film analysis confirmed the changes between the two protocols for the measured absorbed organ dose. Prior work by Vollmar and Kalender [5] evaluating experimental partial scanning techniques for thoracic CT showed a reduced dose throughout the breast tissue, a finding that was not confirmed in our study, likely due to the difference in breast positions. Duan et al. [13] recently reported a similar but smaller maximum surface dose reduction anteriorly and compensatory surface dose increase posteriorly in their evaluation of organ-based dose modulation technology in thoracic CT; in contrast to their study, our work directly assessed absorbed organ doses, which are used to assess radiation dose risks for cancer induction. Similar to previous studies assessing this and other forms of dose reduction techniques for thoracic CT protocols, our work shows that image noise can be preserved centrally within the thorax while reducing absorbed dose to the breast [2, 4 9, 12, 13]. Our study specifically focused on evaluating organ-based dose modulation in thoracic CT. Organ-based dose modulation may also be clinically valuable in other areas of the body for which reduction of radiation exposure to superficially located radiosensitive organs is desired. Recent phantom-based studies have reported relative dose reduction of 33 47% to the eye lens using organ-based dose modulation in during head CT without significant change in image quality [20]. Additionally, organ-based dose modulation has W71

Lungren et al. Net Optical Density 0.18 0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.02 Posterior TCM OBDM 5 10 15 20 Distance Along Midline (cm) been reported to reduce thyroid glandular dose during neck MDCT imaging [20]. The difference in dose reduction between the medial and lateral breast locations for organ-based dose modulation may have clinical implications. The most common location for breast cancer to arise is in the upper outer quadrant [21]. Using organ-based dose modulation resulted in a smaller dose savings to the lateral breast compared with the medial breast in the phantom model we used. The Anterior A C more modest dose saving to the lateral breast locations is likely related to the relationship of the breast tissue to the prescribed anterior radial arc of the organ-based dose modulation protocol rather than the ramp-up and rampdown time needed for the change in tube current because this change has been reported to be a small fraction of the anterior arc [13]. Furthermore, our analysis of breast tissue location in a clinical cohort found that the breast tissue may largely reside outside Fig. 4 Radiochromic film axial dose distribution for organ-based dose modulation (OBDM) and tube current modulation (TCM) thoracic CT. A and B, Radiochromic images show OBDM (A) and TCM (B) thoracic CT. Circles indicate skin dose locations, and asterisks indicate internal breast organ locations. Net OD = net optical density. C, Graph shows net optical density along posterior-to-anterior direction in midline of thorax for OBDM (red) and tube current modulation (black). TABLE 4: Quantitative Noise Measurements (HU) for Tube Current Modulation and Organ-Based Dose Modulation MDCT Protocols Protocol Anterior Mediastinum Midline Noise (HU) Central Mediastinum Posterior Chest Wall (Paraspinal) Midline Tube current modulation 29.0 ± 12.8 28.2 ± 12.1 21.0 ± 13.8 Organ-based dose modulation 27.5 ± 12.4 28.0 ± 12.3 20.6 ± 13.5 p 0.89 0.93 0.97 Note Except for p, values are ± SD. p value of < 0.05 is considered statistically significant. the area of greatest dose reduction afforded by current organ-based dose modulation software. Efforts to constrain the breast tissue to the medial aspect of the thorax so that it lies completely within the 120 radial arc or to investigate additional techniques to widen the angle of maximal beam-current reduction for organ-based dose modulation may be of benefit to provide similar dose savings to the lateral and medial breast tissue during organbased dose modulation. B W72

Organ-Based Dose Modulation More data are likely needed to fully describe glandular breast tissue location in supine patient positioning in different patient cohorts to determine the variation in breast dose. Clinical implementation of organ-based dose modulation should also involve consideration of the effect of improper patient centering within the center of the gantry. As has been noted with tube current modulation, improper patient centering within the gantry can alter absorbed radiation doses [22], and a similar phenomenon is likely in the setting of improper patient positioning within the gantry during the clinical use of organ-based dose modulation. Numerous conditions may contribute to changes in patient position and may alter the relative position of the breast tissue into the more lateral portion of the gantry rotation. The calculated effective dose was slightly higher for the organ-based dose modulation protocol compared with the tube current modulation protocol. Some of this difference may be due to using individual point sources for organ dose measurements. The CTDI vol (before scanning) was similar for both studies, 7.71 and 7.69 mgy, respectively, for the tube current modulation and organ-based dose modulation protocols; however, the CTDI vol (after scanning) was subsequently slightly lower for both studies, 6.94 and 7.51 mgy for tube current modulation and organ-based dose modulation, respectively. This difference in before and after scanning CTDI vol relates to the difference in the tube current modulation between protocols. Both protocols use tube current modulation along the z-axis that is based on the scout film; however, the tube current modulation protocol uses real time variable x- and y- (angular) tube current modulation during scanning, whereas the organ-based dose modulation protocol uses a fixed anterior-posterior x- and y-modulation. There were several limitations to this study. We used a single size anthropomorphic phantom, which limits generalization of the results to a population of heterogeneous body types. It is possible that the dose savings could be different with other morphologies. MOSFET detectors are limited to point detection of doses, and there may be some variation in dose (accounting for range in SD for each organ dose) that we cannot appreciate related to a change in the start point of the gantry. Thus, for each run, the beam location to any specific MOSFET detector may vary, and asymmetry of the anthropomorphic phantom may affect point dose measurements. Given these limitations, the supplementation with radiochromic film analysis corroborates the reported MOSFET data and confirms the trend in change in dose on the basis of the overall location of organs. Measurements were not performed on the phantom in the prone position, which may change breast tissue positioning and dose results. Our image quality analysis was limited to noise and streak artifacts, and no other diagnostic quality evaluation was performed; however, the lack of difference in noise supports that contrast differences are not expected on the basis of the mechanism of the organ-based dose modulation technology. Finally, table attenuation was not quantified and may account for up to a 30% decrease in photon flux posteriorly, and this may have implications if the patient was placed in a supine position and organ-based dose modulation was used. In summary, organ-based dose modulation that uses a 120 radial arc can significantly reduce the absorbed doses to anteriorly located organs during adult thoracic CT without change in image noise. The benefit of this dose reduction must be clinically determined on the basis of the relationship of the irradiated organs to the location of the prescribed radial arc used in organ-based dose modulation. References 1. [No authors listed]. The 2007 Recommendations of the International Commission on Radiological Protection. ICRP publication 103. Ann ICRP 2007; 37:1 332 2. Angel E, Yaghmai N, Jude CM, et al. Dose to radiosensitive organs during routine chest CT: effects of tube current modulation. AJR 2009; 193:1340 1345 3. Ellett WH. BEIR IV report. Science 1988; 241:1144 4. Hurwitz LM, Yoshizumi TT, Goodman PC, et al. Effective dose determination using an anthropomorphic phantom and metal oxide semiconductor field effect transistor technology for clinical adult body multidetector array computed tomography protocols. J Comput Assist Tomogr 2007; 31:544 549 5. Vollmar SV, Kalender WA. Reduction of dose to the female breast in thoracic CT: a comparison of standard-protocol, bismuth-shielded, partial and tubecurrent-modulated CT examinations. Eur Radiol 2008; 18:1674 1682 6. Gies M, Kalender WA, Wolf H, et al. Dose reduction in CT by anatomically adapted tube current modulation. Part I. Simulation studies. Med Phys 1999; 26:2235 2247 7. Kubo T, Lin PJ, Stiller W, et al. Radiation dose reduction in chest CT: a review. AJR 2008; 190:335 343 8. McCollough CH, Bruesewitz MR, Kofler JM Jr. CT dose reduction and dose management tools: overview of available options. RadioGraphics 2006; 26:503 512 9. Tack D. Radiation dose optimization in thoracic imaging. JBR-BTR 2010; 93:15 19 10. Kalra MK, Rizzo S, Maher MM, et al. Chest CT performed with z-axis modulation: scanning protocol and radiation dose. Radiology 2005; 237: 303 308 11. Schueller-Weidekamm C, Schaefer-Prokop CM, Weber M, et al. CT angiography of pulmonary arteries to detect pulmonary embolism: improvement of vascular enhancement with low kilovoltage settings. Radiology 2006; 241:899 907 12. Hurwitz LM, Yoshizumi TT, Goodman PC, et al. Radiation dose savings for adult pulmonary embolus 64-MDCT using bismuth breast shields, lower peak kilovoltage, and automatic tube current modulation. AJR 2009; 192:244 253 13. Duan X, Wang J, Christner JA, et al. Dose reduction to anterior surfaces with organ-based tubecurrent modulation: evaluation of performance in a phantom study. AJR 2011; 197:689 695 14. Brady S, Yoshizumi T, Toncheva G, et al. Implementation of radiochromic film dosimetry protocol for volumetric dose assessments to various organs during diagnostic CT procedures. Med Phys 2010; 37:4782 4792 15. Alva H, Mercado-Uribe H, Rodriguez-Villafuerte M, et al. The use of a reflective scanner to study radiochromic film response. Phys Med Biol 2002; 47:2925 2933 16. Devic S, Tomic N, Soares CG, et al. Optimizing the dynamic range extension of a radiochromic film dosimetry system. Med Phys 2009; 36:429 437 17. Ferreira BC, Lopes MC, Capela M. Evaluation of an Epson flatbed scanner to read Gafchromic EBT films for radiation dosimetry. Phys Med Biol 2009; 54:1073 1085 18. Paelinck L, De Neve W, De Wagter C. Precautions and strategies in using a commercial flatbed scanner for radiochromic film dosimetry. Phys Med Biol 2007; 52:231 242 19. Devic S, Seuntjens J, Sham E, et al. Precise radiochromic film dosimetry using a flat-bed document scanner. Med Phys 2005; 32:2245 2253 20. Hoang JK, Yoshizumi TT, Choudhury KR, et al. Organ-based dose current modulation and thyroid shields: techniques of dose reduction for neck CT. AJR 2012; 198:1132 1138 21. Sickles EA. Breast masses: mammographic evaluation. Radiology 1989; 173:297 303 22. Li J, Udayasankar UK, Toth TL, et al. Automatic patient centering for MDCT: effect on radiation dose. AJR 2007; 188:547 552 W73