- Email: [email protected]
Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection
Clinical Radiology xxx (2014) 1e7
Contents lists available at ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection G. Wang, J. Gao*, S. Zhao, X. Sun, X. Chen, X. Cui Department of Radiology, The General Hospital of Chinese People’s Armed Police Forces, Beijing, China
art icl e i nformat ion Article history: Received 17 November 2013 Received in revised form 13 April 2014 Accepted 16 April 2014
AIM: To develop a quantitative body mass index (BMI)-dependent tube voltage and tube current selection method for obtaining consistent image quality and overall dose reduction in computed tomography coronary angiography (CTCA). METHODS AND MATERIALS: The images of 190 consecutive patients (group A) who underwent CTCA with fixed protocols (100 kV/193 mAs for 100 patients with a BMI of <27 and 120 kV/175 mAs for 90 patients with a BMI of >27) were retrospectively analysed and reconstructed with an adaptive statistical iterative reconstruction (ASIR) algorithm at 50% blending. Image noise was measured and the relationship to BMI was studied to establish BMIdependent tube current for obtaining CTCA images with user-specified image noise. One hundred additional cardiac patients (group B) were examined using prospective triggering with the BMI-dependent tube voltage/current. CTCA image-quality score, image noise, and effective dose from groups B and C (subgroup of A of 100 patients examined with prospective triggering only) were obtained and compared. RESULTS: There was a linear relationship between image noise and BMI in group A. Using a BMI-dependent tube current in group B, an average CTCA image noise of 27.7 HU (target 28 HU) and 31.7 HU (target 33 HU) was obtained for the subgroups of patients with BMIs of >27 and of <27, respectively, and was independent of patient BMI. There was no difference between image-quality scores between groups B and C (4.52 versus 4.60, p > 0.05). The average effective dose for group B (2.56 mSv) was 42% lower than group C (4.38 mSv; p < 0.01). CONCLUSION: BMI-dependent tube voltage/current selection in CTCA provides an individualized protocol that generates consistent image quality and helps to reduce overall patient radiation dose. Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction Computed tomography (CT) coronary angiography (CTCA) has become a mainstream clinical application with * Guarantor and correspondent: J. Gao, Department of Radiology, The General Hospital of Chinese People’s Armed Police Forces, No. 69, Yongding Road, Beijing 100039, China. Tel.: þ86 10 57976744; fax: þ86 10 67873782. E-mail address: [email protected] (J. Gao).
the introduction of multidetector row CT (MDCT) with fast rotation speed and wide coverage that can cover the whole heart in <5 s.1e4 The need to reduce the x-ray dose to cardiac patients has also grown considerably due to the use of high spatial resolution, submillimetre section thicknesses, and small helical pitches in cardiac helical CT to satisfy the requirement of complete coverage of all cardiac phases for high-resolution coronary artery imaging and functional measurement.3e5 Many techniques have been developed in
0009-9260/$ e see front matter Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.crad.2014.04.016
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
2
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
cardiac CT applications to effectively reduce the x-ray dose required to obtain the desired image quality. These techniques have included both hardware and software approaches.6e9 Scan mode change can greatly reduce CTCA radiation dose. Recently, a prospectively triggered CTCA technique has been shown to reduce the dose by up to 83% compared to the retrospectively gated helical mode by minimizing scan overlap.4,10 Imaging parameter optimization provides a way to balance both dose and image quality. In CTCA, the tube current can be selected based on patientspecific information, such as patient size, weight, body mass index (BMI), or based on other more sophisticated methods with phantom calibration, scout scan attenuation measurement, and timing bolus analysis.11e16 Using lower tube voltages and statistical reconstruction algorithms for slim patients has also shown advantages in improving the signal-to-noise ratio (SNR) for contrast-enhanced CT.17,18 As the BMI is a standard parameter measured on every patient undergoing CT, the present authors have proposed a quantitative method to select both the tube voltage and current based on BMI values for obtaining pre-defined image noise levels for CTCA. The purpose of the present study was to evaluate the robustness of this quantitative method for obtaining consistent image quality and assess its dosereduction potential in MDCT coronary artery imaging.
Materials and methods The institutional review board waived the requirement for approval and informed consent for this study. The investigation was divided into two parts: (1) the establishment of a quantitative relationship between patient BMI and tube current required for obtaining pre-defined image noise levels, and (2) the evaluation of the robustness of the quantitative method for obtaining consistent image quality and optimizing radiation exposure. Patients with atrial fibrillation that caused irregular heartbeats; heart rates >70 beats/min after 50 mg oral metoprolol; liver or renal insufficiency; known allergy to iodinated contrast media; potential pregnancy or breastfeeding; or could not breathhold for 10 s were excluded from the study.
Patient selection One hundred and ninety consecutive patients with controlled heart rates of <70 beats/min and suspected of having coronary artery disease [group A, 138 men, 52 women, aged 27e80 years with a mean of 57.3 years, BMI values from 20e34.4 with a mean of 25.97, 100 patients with a BMI of <27 (mean 23.56), and 90 patients with a BMI of >27 (mean 28.6)] who underwent standard cardiac examinations at The General Hospital of Chinese People’s Armed Police Forces, from January 2012 to April 2012 were included to establish the BMI-based tube current selection method. A second set of 100 consecutive cardiac patients recruited from May 2012 to August 2012 (group B: 68 men, 32 women, aged 32e82 years with a mean of 55.1 years; BMI values from 17.9e32.1 with a mean of 25.34) was scanned with an individually adjusted tube current
established from group A to evaluate the robustness and clinical value of the method. Before imaging, the heart rates of all patients were controlled to <70 beats/min using oral metoprolol at 25e50 mg.
CT protocol All patients underwent CT coronary artery imaging using a Discovery CT750 HDCT machine (GE Healthcare, Waukesha, WI, USA) in the supine position using either retrospective gating or prospective triggering scanning mode in group A, and prospective triggering-only in group B. A single scout scan was performed to determine the scan range. Timing bolus scans were performed before CTCA to determine the scan start-delay time: a total of 15 ml contrast media (350 mg iodine/ml Ioversol) at the speed of 4.5 ml/s was injected intravenously. Cine scans were used to monitor the contrast enhancement in the aorta. A region of interest (ROI) was placed on the ascending aorta and a contrast-enhancement curve as function of time was measured to determine the blood circulation time between the vein of the elbow and the aorta. This circulation time was then used to determine the optimal scan delay time for CTCA. CTCA was then performed with the predetermined scan-delay time, and 70 ml contrast media was injected at a speed of 4.5 ml/s intravenously followed by 30 ml saline at the same injection rate. For patients in group A, standard low-dose imaging techniques were used for the CTCA examinations: retrospective gating with electrocardiography (ECG) tube current modulation (peak tube current for cardiac phases from 40% to 80%) for patients with heart rates between 65 and 70 beats/min; prospective triggering with padding time of 50 ms centred at 75% cardiac phase for patients with heart rates <65 beats/min. Four steps were performed to cover the whole heart from tracheal bifurcation to the diaphragm in the prospective triggering mode. Tube voltages and currents were 100 kV/550 mA and 120 kV/500 mA for patients with BMIs of <27 and >27, respectively. Other imaging parameters included 0.35 s rotation speed, 64 sections with 0.625 mm section thickness, small bowtie (25 cm display field of view). For patients in group B, there were three major changes from group A for the CTCA protocol: (1) patient heart rates were better controlled to <65 beats/min so that all CTCA examinations were performed using the prospective triggering mode, and the padding time was reduced to 0 ms after carefully analysing the cardiac phase selection in group A; (2) the number of scans was further reduced from four to three scans when possible; and (3) the tube current was dependent on the patients’ BMI values using the formula determined in group A after image noise analysis.
Image evaluation and statistical analysis All CTCA images were reconstructed with a section thickness of 0.625 mm using the adaptive statistical iterative reconstruction (ASIR) algorithm at 50% blending (50% ASIR). CTCA images were transferred to a standalone
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
3
Figure 1 Exemplary images for scoring CPR image quality. (a) Score of 5: excellent image quality with clear vessel wall definition and limited perceptible lumen noise, fully acceptable for diagnosis; (b) score of 4: good image quality with minimal lumen noise and well maintained vessel wall definition, fully acceptable for diagnosis; (c) score of 3: adequate image quality with moderate lumen noise and minimal limitation of vessel wall definition, acceptable for diagnosis; (d) score of 2: reduced image quality with high lumen noise and poor vessel wall definition or low lumen attenuation, diagnosis acceptable only under limited conditions; and (e) score of 1: impaired image quality limited by excessive lumen noise and poor vessel wall definition or poor lumen attenuation, unacceptable for diagnosis.
advanced workstation (AW4.5, GE Healthcare) for analysis and three-dimensional reconstruction and display. Image noise was measured in an ROI of 100 mm2 placed at the centre of the aortic root for all images. The average value of the standard deviations of three consecutive sections at this level was used to represent image noise to reduce measurement variation. A subset of 100 patients from group A that underwent the prospective triggering CTCA were selected to form group C (BMI values from 20.1e32.5 with a mean of 25.76) for image quality and radiation dose comparison with group B. Two experienced radiologists quantitatively evaluated the curved-plane reformat image quality blindly according to the diagnostic acceptability and Likert score9 in terms of image noise, overall image quality, and vessel sharpness using scores of 1e5 for images in group B and group C. The five-point scale was assigned based on the worst scored artery >2 mm in diameter: 1 ¼ impaired image quality limited by excessive lumen noise and poor vessel wall definition or poor lumen attenuation, unacceptable for diagnosis; 2 ¼ reduced image quality with high lumen noise and poor vessel wall definition or low lumen attenuation, diagnosis acceptable only under limited conditions; 3 ¼ adequate image quality with moderate lumen noise and minimal limitation of vessel wall definition, acceptable for diagnosis; 4 ¼ good image quality with minimal lumen noise and well maintained vessel wall definition, fully acceptable for diagnosis; and 5 ¼ excellent image quality with clear vessel wall definition and limited perceptible lumen noise, fully acceptable for diagnosis. The readers were instructed to ignore motion artefact and stair-step artefact. Final scores were obtained in consensus between the two reviewers and sample images are shown in Fig 1. The CT machine recorded the volume-weighted CT dose index (CTDIvol) and doseelength product (DLP) automatically for each patient when the examination was performed. The effective dose (E) was calculated using the formula E ¼ DLP 0.028 with the newest conversion factor of 0.028, which is specific for cardiac CT.19
Statistical analyses were performed on the patient BMI values, image-quality scores (IQS), image noise measurements, and the effective dose measurements from the two sets using SPSS version 10.0 software with p < 0.05 being statistically significant for differences between the two groups.
Desired image noise selection The desired CTCA image noise for group B was preselected based on image noise analysis from 500 previous cardiac examinations acquired earlier at The General Hospital of Chinese People’s Armed Police Forces.16 An image noise level of 28 HU was selected for patients scanned using 120 kV. As 100 kV enhances the contrast for the contrast medium in the vessel, a more relaxed image noise level of 33 HU was selected for patients scanned at a tube voltage of 100 kV to maintain a similar SNR for all patients.
Results CTCA image noise from group A was plotted against patient BMI, and fitted with a linear equation (Fig 2): SDðmAF Þ ¼ a*BMI þ b
(1)
where SD(mAF) is the CTCA image noise with a fixed tube current (mAF ¼ 550 mA with 100 kV and mAF ¼ 500 mA with 120 kV), and parameters a and b are obtained using the regression analysis. Good correlations were observed between image noise and patient BMI value with constant tube current for the two subgroups in group A (R ¼ 0.62 and R ¼ 0.67 for subgroups with BMIs of <27 and >27, respectively). Using the basic relationship between image noise and tube current, the following equation was obtained: ½SDðmAF Þ=SDðmAX Þ2 ¼ ½mAX =mAF
(2)
where SD(mAx) is a pre-defined or desired CTCA image noise, and mAx is the required tube current to obtain this desired image noise.
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
4
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
Figure 2 Linear relationship was obtained between image noise and patient BMI with constant tube current.
Inserting Eq. (1) into Eq. (2) we obtained the formula to automatically calculate tube current based on patient BMI for obtaining consistent image quality: mAX ¼ mAF ½ða*BMI þ bÞ=SDðmAX Þ2
(3)
If an acceptable image noise is pre-defined, the required tube current can be calculated for any patient with a given BMI using this relationship. Specifically, the following two expressions were obtained for patients with BMIs of <27 (with 100 kV) and >27 (with 120 kV) on the Discovery CT750 HD with 0.35 s gantry rotation speed, respectively: . (4) mAX ¼ 550 ½ð1:5457*BMI 6:6464Þ SDðmAX Þ2 and . mAX ¼ 500 ½ð1:1111*BMI 3:9861Þ SDðmAX Þ2
(5)
For ease of clinical application, a look-up table was established relating the required tube current to patient BMI for obtaining image noise levels of 33 and 28 HU, as demonstrated in Table 1. The patient’s BMI, IQSs, and image noise measurements, and effective dose calculations for groups B and C are listed
in Table 2. The average BMI value and IQS for groups B and C were 25.34, 4.52, 25.76, and 4.60, respectively. Statistical analysis indicated that there was no difference between the corresponding parameters for the two groups. The average CTCA image noise in group B with BMI-dependent tube current selection was 27.7 HU for the BMI >27 subgroup and 31.7 HU for the BMI <27 subgroup. These values were within 5% of the presets. Fig 3 shows the scatter plots of the image noise as a function of patient BMI for groups B and C. The overall image noise and IQS in group B was statistically the same as that in group C (p ¼ 0.10 and p ¼ 0.42; Table 2). Adequate SNR values of 17 and 14.5 were obtained for the 100 kV subgroup (BMI <27) and 120 kV subgroup (BMI >27) in group B, respectively. In addition, there was no difference in the overall SNR values between groups B and C (16.07 versus 16.71; Table 2). The CTDIw distributions for patients in groups B and C are shown in Fig 4. The average effective dose values for patients in groups B and C were 2.56 0.88 and 4.38 0.82 mSv, respectively. A 42% dose reduction was achieved on average and a dose reduction of approximately 69% (1.14 versus 3.7 mSv) was achieved for the smallest patient in the present cohort with the use of an adaptive tube current (Fig 5).
Discussion Table 1 Tube current values required to obtain an image noise of 33 HU [for a body mass index (BMI) of <27 with 100 kVp tube voltage] and 28 HU (for a BMI of >27 with 120 kVp tube voltage) at computed tomography coronary angiography (CTCA) based on patient BMI with a 0.35 s gantry rotation speed and 0.625 mm image section thickness. BMI (kg/m2)
Required tube voltage/ current (kV/mA) for an image noise of 33 HU
BMI (kg/m2)
Required tube voltage/current (kV/mA) for an image noise of 28 HU
18 19 20 21 22 23 24 24.5 25 26 26.5
100/225 100/260 100/295 100/335 100/380 100/420 100/470 100/490 100/515 100/570 100/595
27 27.5 28 28.5 29 30 31 32 33 34 35.5
120/430 120/450 120/470 120/490 120/505 120/550 120/590 120/635 120/680 120/730 120/800
In the present study, a quantitative low-dose imaging protocol for selecting tube voltage and current based on patient BMI in CTCA applications was proposed. The study demonstrated that by effectively combining many dosereduction techniques20 and adaptively adjusting tube voltage and current based on the patient’s BMI value, one Table 2 Statistical analysis for parameters in group B [with body mass index (BMI)dependent tube current values] and Group C (with fixed tube current values). Group B BMI Image-quality score Image noise Signal-to-noise ratio Doseelength product (mGy$cm) Effective dose (mSv)
25.34 4.52 30.23 16.07 91.43
Group C 2.85 0.59 2.79 2.75 31.43
2.56 0.88
25.76 4.6 29.01 16.71 156.43
p-Value 2.82 0.55 3.67 3.55 29.29
4.38 0.82
0.3 0.42 0.1 0.16 0 0
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
5
Figure 3 Image noise variation as function of patient BMI in group B [(aeb) with BMI-dependent tube current values] and group C [(ced) with fixed tube current values].
could optimize dose delivery to patients and obtain consistent and desired image quality. The use of a prospective triggering scan mode provides the initial huge dose reduction compared with the conventional retrospective-gated helical scan mode,4,10 which requires the use of sufficient b-blockers to control heart rate and heart rate variability.20 For group B, patient heart rates were controlled <65 beats/min, which allowed the use of the prospective-triggering mode for all patients in this group. Furthermore, due to the low and steady heart rates, the padding time for the prospective triggering could be reduced from the initial 50 ms to 0 ms for the study group. The reduction in padding time translated to a 30% dose reduction for the patients in the study group. Scan range could also be optimized to further reduce the radiation dose to patients. In the present study, the number of steps was reduced from four to three for approximately 16% of patients by tightly controlling the scan range, which yielded an additional 4% dose reduction to the study population.
Scan protocol optimization does not necessary mean dose reduction in all cases; in some cases a higher dose was required. However, it does optimize the patient dose distribution, and for smaller patients, especially for those with BMI <20, the tube current should be greatly reduced.21,22 In fact, for patients in group B with patient-dependent tube current selection, a wide range of tube currents was required. For some large patients, a high radiation dose (approximately 5 mSv) was required using the retrospective scan mode to produce the desired image quality. On the other hand, for smaller patients, the dose reduction could be very dramatic. An effective dose of only 1.14 mSv was achieved for the patient with a BMI of 17.1 kg/m2 in the present cohort with acceptable image quality, resulting in an almost 70% dose reduction. As demonstrated in Figs 3 and 4 for group C where a constant tube current was used, the effective dose was independent of patient BMI but the image noise had strong dependency on patient BMI values. Image noise ranged from 22e38 HU. This would
Figure 4 CTDIw values as function of patient BMI in group B (left with BMI-dependent tube current values) and group C (right with fixed tube current values). Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
6
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
Figure 5 An example of patient-dependent tube current/voltage adjustment that enabled excellent image quality for a large patient (BMI ¼ 32.1 kg/m2, CTDIw ¼ 12.2 mGy) and substantial dose reduction with equally good image quality for a smaller patient (BMI ¼ 19.5 kg/m2, CTDIw ¼ 3.39 mGy).
indicate over-exposure for smaller patients and possible under-exposure for large patients. On the other hand, for group B with the proposed adaptive tube current selection, image noise had a narrower range of 24e35 HU. The line fitted to the image noise and patient BMI for group B had a slope of close to zero indicating image noise in this group was independent of patient BMI values. The image noise level selection does have a very big impact on the overall dose delivery to patient population. Lower noise levels means higher overall doses and vice versa. On the other hand, clinical practice indicated that the incremental impact of image noise to diagnostic ability is limited when image noises change within a certain range.23 This provided us with the ability to moderately decrease the image noise requirement without negatively impacting diagnostic outcomes. Finding the highest image noise (lowest x-ray dose) for acceptable diagnostic quality would be desirable and would require systematic studies. Such systematic studies would be facilitated by the quantitative tube currenteBMI formula proposed in this study, which enabled a desired noise level to be achieved more consistently across the patient population. The correlation between cardiac image noise and patient BMI is only moderate. This is because BMI is only an average value. It does not specifically reflect the fat and muscle composition of the chest where the heart is located. Even with same BMI value, different patients could have different shapes and different fat and muscle distributions, resulting in cardiac image noise deviations from the pre-selected noise value. In the future, the noise prediction accuracy could be improved by incorporating individual information such as gender and age. In addition, more robust methods for selecting the tube voltage and current could be evaluated to obtain more consistent cardiac images. Another limitation is that the values determined in the present study are manufacturer (and been probably machine) specific given the proprietary ASIR, percent image blending, etc.,
which influence image noise; although the method could potentially be used by practitioners using other vendors equipment. In conclusion, the present study describes a quantitative and effective method for tube voltage and tube current selection based on patient BMI value for CTCA. This method provides an individualized protocol to obtain consistent image quality and optimized dose delivery across the patient population.
References 1. Taguchi K, Anno H. High temporal resolution for multislice helical computed tomography. Med Phys 2000;27:861e72. 2. Herzog C, Abolmaali N, Balzer JO, et al. Heart-rate-adapted image reconstruction in multidetector-row cardiac CT: influence of physiological and technical prerequisite on image quality. Eur Radiol 2002;12: 2670e8. 3. Choi HS, Choi BW, Choe KO, et al. Pitfalls, artifacts, and remedies in multi-detector row CT coronary angiography. RadioGraphics 2004;24: 787e800. 4. Hsieh J, Londt J, Vass M, et al. Step-and-shoot data acquisition and reconstruction for cardiac x-ray computed tomography. Med Phys 2006;33:4236e48. 5. Hurwitz LM, Reiman RE, Yoshizumi TT, et al. Radiation dose from contemporary cardiothoracic multidetector CT protocols with an anthropomorphic female phantom: implications for cancer induction. Radiology 2007;245:742e50. 6. Avinash G. Method and apparatus for enhancing discrete pixel images. US Patent, http://patft.uspto.gov/netacgi/ - h0http://patft.uspto.gov/ netacgi/ - h2; 2001. 6,208,763. 7. Jakobs TF, Becker CR, Ohnesorge B, et al. Multislice helical CT of the heart with retrospective ECG gating: reduction of radiation exposure by ECGcontrolled tube current modulation. Eur Radiol 2002;12:1081e6. 8. McCollough C, Bruesewitz M, Kofler Jr J. CT dose reduction and dose management tools: overview of available options. RadioGraphics 2006;26:503e12. 9. Leipsic J, Labounty TM, Heilbron B, et al. Adaptive statistical iterative reconstruction: assessment of image noise and image quality in coronary CT angiography. AJR Am J Roentgenol 2010;195:649e54. 10. Earls JP, Berman EL, Urban BA, et al. Prospectively gated transverse coronary CT angiography versus retrospectively gated helical technique:
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
11.
12.
13.
14.
15.
16. 17.
improved image quality and reduced radiation dose. Radiology 2008;246:742e53. Mahnken AH, Wildberger JE, Simon J, et al. Detection of coronary calcifications: feasibility of dose reduction with a body weight-adapted examination protocol. AJR Am J Roentgenol 2003;181:533e8. Herzog BA, Husmann L, Valenta I, et al. Determinants of vessel contrast in BMI-adapted low dose CT coronary angiography with prospective ECG-triggering. Int J Cardiovasc Imaging 2009;25:625e30. Das M, Mahnken A, Muhlenbruch G, et al. Individually adapted examination protocols for reduction of radiation exposure for 16-MDCT chest examinations. AJR Am J Roentgenol 2005;184:1437e43. Hur G, Hong S, Kim S, et al. Uniform image quality achieved by tube current modulation using SD of attenuation in coronary CT angiography. AJR Am J Roentgenol 2007;189:188e98. Qi W, Li J, Du X. Method for automatic tube current selection for obtaining a consistent image quality and dose optimization in a cardiac multidetector CT. Korean J Radiol 2009;10:568e74. Gao J, Li J, Earls J, et al. Individualized tube current selection for 64-row cardiac CTA based on analysis of the scout view. Eur J Radiol 2011;79:266e71. Marin D, Nelson RC, Schindera ST, et al. Low-tube-voltage, high-tubecurrent multidetector abdominal CT: improved image quality and
18.
19.
20.
21.
22.
23.
7
decreased radiation dose with adaptive statistical iterative reconstruction algorithm d initial clinical experience. Radiology 2010;254:145e53. Feuchtner GM, Jodocy D, Klauser A, et al. Radiation dose reduction by using 100-kV tube voltage in cardiac 64-slice computed tomography: a comparative study. Eur J Radiol 2010;75:e51e6. Gosling O, Loader R, Venables P, et al. Cardiac CT: are we underestimating the dose? A radiation dose study utilizing the 2007 ICRP tissue weighting factors and a cardiac specific scan volume. Clin Radiol 2010;65:1013e7. Raff GL, Chinnaiyan KM, Share DA, et al. Radiation dose from cardiac computed tomography before and after implementation of radiation dose-reduction techniques. JAMA 2009;301:2340e8. Hunold P, Vogt FM, Schmermund A, et al. Radiation exposure during cardiac CT: effective doses at multi-detector row CT and electron-beam CT. Radiology 2003;226:145e52. Mahnken AH, Wildberger JE, Koos R, et al. Multislice spiral computed tomography of the heart: technique, current applications and perspective. Cardiovasc Intervent Radiol 2005;28:388e99. Prasad SR, Wittram C, Shepard JA, et al. Standard-dose and 50%reduced-dose chest CT: comparing the effect on image quality. AJR Am J Roentgenol 2002;179:461e5.
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
Contents lists available at ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection G. Wang, J. Gao*, S. Zhao, X. Sun, X. Chen, X. Cui Department of Radiology, The General Hospital of Chinese People’s Armed Police Forces, Beijing, China
art icl e i nformat ion Article history: Received 17 November 2013 Received in revised form 13 April 2014 Accepted 16 April 2014
AIM: To develop a quantitative body mass index (BMI)-dependent tube voltage and tube current selection method for obtaining consistent image quality and overall dose reduction in computed tomography coronary angiography (CTCA). METHODS AND MATERIALS: The images of 190 consecutive patients (group A) who underwent CTCA with fixed protocols (100 kV/193 mAs for 100 patients with a BMI of <27 and 120 kV/175 mAs for 90 patients with a BMI of >27) were retrospectively analysed and reconstructed with an adaptive statistical iterative reconstruction (ASIR) algorithm at 50% blending. Image noise was measured and the relationship to BMI was studied to establish BMIdependent tube current for obtaining CTCA images with user-specified image noise. One hundred additional cardiac patients (group B) were examined using prospective triggering with the BMI-dependent tube voltage/current. CTCA image-quality score, image noise, and effective dose from groups B and C (subgroup of A of 100 patients examined with prospective triggering only) were obtained and compared. RESULTS: There was a linear relationship between image noise and BMI in group A. Using a BMI-dependent tube current in group B, an average CTCA image noise of 27.7 HU (target 28 HU) and 31.7 HU (target 33 HU) was obtained for the subgroups of patients with BMIs of >27 and of <27, respectively, and was independent of patient BMI. There was no difference between image-quality scores between groups B and C (4.52 versus 4.60, p > 0.05). The average effective dose for group B (2.56 mSv) was 42% lower than group C (4.38 mSv; p < 0.01). CONCLUSION: BMI-dependent tube voltage/current selection in CTCA provides an individualized protocol that generates consistent image quality and helps to reduce overall patient radiation dose. Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction Computed tomography (CT) coronary angiography (CTCA) has become a mainstream clinical application with * Guarantor and correspondent: J. Gao, Department of Radiology, The General Hospital of Chinese People’s Armed Police Forces, No. 69, Yongding Road, Beijing 100039, China. Tel.: þ86 10 57976744; fax: þ86 10 67873782. E-mail address: [email protected] (J. Gao).
the introduction of multidetector row CT (MDCT) with fast rotation speed and wide coverage that can cover the whole heart in <5 s.1e4 The need to reduce the x-ray dose to cardiac patients has also grown considerably due to the use of high spatial resolution, submillimetre section thicknesses, and small helical pitches in cardiac helical CT to satisfy the requirement of complete coverage of all cardiac phases for high-resolution coronary artery imaging and functional measurement.3e5 Many techniques have been developed in
0009-9260/$ e see front matter Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.crad.2014.04.016
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
2
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
cardiac CT applications to effectively reduce the x-ray dose required to obtain the desired image quality. These techniques have included both hardware and software approaches.6e9 Scan mode change can greatly reduce CTCA radiation dose. Recently, a prospectively triggered CTCA technique has been shown to reduce the dose by up to 83% compared to the retrospectively gated helical mode by minimizing scan overlap.4,10 Imaging parameter optimization provides a way to balance both dose and image quality. In CTCA, the tube current can be selected based on patientspecific information, such as patient size, weight, body mass index (BMI), or based on other more sophisticated methods with phantom calibration, scout scan attenuation measurement, and timing bolus analysis.11e16 Using lower tube voltages and statistical reconstruction algorithms for slim patients has also shown advantages in improving the signal-to-noise ratio (SNR) for contrast-enhanced CT.17,18 As the BMI is a standard parameter measured on every patient undergoing CT, the present authors have proposed a quantitative method to select both the tube voltage and current based on BMI values for obtaining pre-defined image noise levels for CTCA. The purpose of the present study was to evaluate the robustness of this quantitative method for obtaining consistent image quality and assess its dosereduction potential in MDCT coronary artery imaging.
Materials and methods The institutional review board waived the requirement for approval and informed consent for this study. The investigation was divided into two parts: (1) the establishment of a quantitative relationship between patient BMI and tube current required for obtaining pre-defined image noise levels, and (2) the evaluation of the robustness of the quantitative method for obtaining consistent image quality and optimizing radiation exposure. Patients with atrial fibrillation that caused irregular heartbeats; heart rates >70 beats/min after 50 mg oral metoprolol; liver or renal insufficiency; known allergy to iodinated contrast media; potential pregnancy or breastfeeding; or could not breathhold for 10 s were excluded from the study.
Patient selection One hundred and ninety consecutive patients with controlled heart rates of <70 beats/min and suspected of having coronary artery disease [group A, 138 men, 52 women, aged 27e80 years with a mean of 57.3 years, BMI values from 20e34.4 with a mean of 25.97, 100 patients with a BMI of <27 (mean 23.56), and 90 patients with a BMI of >27 (mean 28.6)] who underwent standard cardiac examinations at The General Hospital of Chinese People’s Armed Police Forces, from January 2012 to April 2012 were included to establish the BMI-based tube current selection method. A second set of 100 consecutive cardiac patients recruited from May 2012 to August 2012 (group B: 68 men, 32 women, aged 32e82 years with a mean of 55.1 years; BMI values from 17.9e32.1 with a mean of 25.34) was scanned with an individually adjusted tube current
established from group A to evaluate the robustness and clinical value of the method. Before imaging, the heart rates of all patients were controlled to <70 beats/min using oral metoprolol at 25e50 mg.
CT protocol All patients underwent CT coronary artery imaging using a Discovery CT750 HDCT machine (GE Healthcare, Waukesha, WI, USA) in the supine position using either retrospective gating or prospective triggering scanning mode in group A, and prospective triggering-only in group B. A single scout scan was performed to determine the scan range. Timing bolus scans were performed before CTCA to determine the scan start-delay time: a total of 15 ml contrast media (350 mg iodine/ml Ioversol) at the speed of 4.5 ml/s was injected intravenously. Cine scans were used to monitor the contrast enhancement in the aorta. A region of interest (ROI) was placed on the ascending aorta and a contrast-enhancement curve as function of time was measured to determine the blood circulation time between the vein of the elbow and the aorta. This circulation time was then used to determine the optimal scan delay time for CTCA. CTCA was then performed with the predetermined scan-delay time, and 70 ml contrast media was injected at a speed of 4.5 ml/s intravenously followed by 30 ml saline at the same injection rate. For patients in group A, standard low-dose imaging techniques were used for the CTCA examinations: retrospective gating with electrocardiography (ECG) tube current modulation (peak tube current for cardiac phases from 40% to 80%) for patients with heart rates between 65 and 70 beats/min; prospective triggering with padding time of 50 ms centred at 75% cardiac phase for patients with heart rates <65 beats/min. Four steps were performed to cover the whole heart from tracheal bifurcation to the diaphragm in the prospective triggering mode. Tube voltages and currents were 100 kV/550 mA and 120 kV/500 mA for patients with BMIs of <27 and >27, respectively. Other imaging parameters included 0.35 s rotation speed, 64 sections with 0.625 mm section thickness, small bowtie (25 cm display field of view). For patients in group B, there were three major changes from group A for the CTCA protocol: (1) patient heart rates were better controlled to <65 beats/min so that all CTCA examinations were performed using the prospective triggering mode, and the padding time was reduced to 0 ms after carefully analysing the cardiac phase selection in group A; (2) the number of scans was further reduced from four to three scans when possible; and (3) the tube current was dependent on the patients’ BMI values using the formula determined in group A after image noise analysis.
Image evaluation and statistical analysis All CTCA images were reconstructed with a section thickness of 0.625 mm using the adaptive statistical iterative reconstruction (ASIR) algorithm at 50% blending (50% ASIR). CTCA images were transferred to a standalone
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
3
Figure 1 Exemplary images for scoring CPR image quality. (a) Score of 5: excellent image quality with clear vessel wall definition and limited perceptible lumen noise, fully acceptable for diagnosis; (b) score of 4: good image quality with minimal lumen noise and well maintained vessel wall definition, fully acceptable for diagnosis; (c) score of 3: adequate image quality with moderate lumen noise and minimal limitation of vessel wall definition, acceptable for diagnosis; (d) score of 2: reduced image quality with high lumen noise and poor vessel wall definition or low lumen attenuation, diagnosis acceptable only under limited conditions; and (e) score of 1: impaired image quality limited by excessive lumen noise and poor vessel wall definition or poor lumen attenuation, unacceptable for diagnosis.
advanced workstation (AW4.5, GE Healthcare) for analysis and three-dimensional reconstruction and display. Image noise was measured in an ROI of 100 mm2 placed at the centre of the aortic root for all images. The average value of the standard deviations of three consecutive sections at this level was used to represent image noise to reduce measurement variation. A subset of 100 patients from group A that underwent the prospective triggering CTCA were selected to form group C (BMI values from 20.1e32.5 with a mean of 25.76) for image quality and radiation dose comparison with group B. Two experienced radiologists quantitatively evaluated the curved-plane reformat image quality blindly according to the diagnostic acceptability and Likert score9 in terms of image noise, overall image quality, and vessel sharpness using scores of 1e5 for images in group B and group C. The five-point scale was assigned based on the worst scored artery >2 mm in diameter: 1 ¼ impaired image quality limited by excessive lumen noise and poor vessel wall definition or poor lumen attenuation, unacceptable for diagnosis; 2 ¼ reduced image quality with high lumen noise and poor vessel wall definition or low lumen attenuation, diagnosis acceptable only under limited conditions; 3 ¼ adequate image quality with moderate lumen noise and minimal limitation of vessel wall definition, acceptable for diagnosis; 4 ¼ good image quality with minimal lumen noise and well maintained vessel wall definition, fully acceptable for diagnosis; and 5 ¼ excellent image quality with clear vessel wall definition and limited perceptible lumen noise, fully acceptable for diagnosis. The readers were instructed to ignore motion artefact and stair-step artefact. Final scores were obtained in consensus between the two reviewers and sample images are shown in Fig 1. The CT machine recorded the volume-weighted CT dose index (CTDIvol) and doseelength product (DLP) automatically for each patient when the examination was performed. The effective dose (E) was calculated using the formula E ¼ DLP 0.028 with the newest conversion factor of 0.028, which is specific for cardiac CT.19
Statistical analyses were performed on the patient BMI values, image-quality scores (IQS), image noise measurements, and the effective dose measurements from the two sets using SPSS version 10.0 software with p < 0.05 being statistically significant for differences between the two groups.
Desired image noise selection The desired CTCA image noise for group B was preselected based on image noise analysis from 500 previous cardiac examinations acquired earlier at The General Hospital of Chinese People’s Armed Police Forces.16 An image noise level of 28 HU was selected for patients scanned using 120 kV. As 100 kV enhances the contrast for the contrast medium in the vessel, a more relaxed image noise level of 33 HU was selected for patients scanned at a tube voltage of 100 kV to maintain a similar SNR for all patients.
Results CTCA image noise from group A was plotted against patient BMI, and fitted with a linear equation (Fig 2): SDðmAF Þ ¼ a*BMI þ b
(1)
where SD(mAF) is the CTCA image noise with a fixed tube current (mAF ¼ 550 mA with 100 kV and mAF ¼ 500 mA with 120 kV), and parameters a and b are obtained using the regression analysis. Good correlations were observed between image noise and patient BMI value with constant tube current for the two subgroups in group A (R ¼ 0.62 and R ¼ 0.67 for subgroups with BMIs of <27 and >27, respectively). Using the basic relationship between image noise and tube current, the following equation was obtained: ½SDðmAF Þ=SDðmAX Þ2 ¼ ½mAX =mAF
(2)
where SD(mAx) is a pre-defined or desired CTCA image noise, and mAx is the required tube current to obtain this desired image noise.
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
4
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
Figure 2 Linear relationship was obtained between image noise and patient BMI with constant tube current.
Inserting Eq. (1) into Eq. (2) we obtained the formula to automatically calculate tube current based on patient BMI for obtaining consistent image quality: mAX ¼ mAF ½ða*BMI þ bÞ=SDðmAX Þ2
(3)
If an acceptable image noise is pre-defined, the required tube current can be calculated for any patient with a given BMI using this relationship. Specifically, the following two expressions were obtained for patients with BMIs of <27 (with 100 kV) and >27 (with 120 kV) on the Discovery CT750 HD with 0.35 s gantry rotation speed, respectively: . (4) mAX ¼ 550 ½ð1:5457*BMI 6:6464Þ SDðmAX Þ2 and . mAX ¼ 500 ½ð1:1111*BMI 3:9861Þ SDðmAX Þ2
(5)
For ease of clinical application, a look-up table was established relating the required tube current to patient BMI for obtaining image noise levels of 33 and 28 HU, as demonstrated in Table 1. The patient’s BMI, IQSs, and image noise measurements, and effective dose calculations for groups B and C are listed
in Table 2. The average BMI value and IQS for groups B and C were 25.34, 4.52, 25.76, and 4.60, respectively. Statistical analysis indicated that there was no difference between the corresponding parameters for the two groups. The average CTCA image noise in group B with BMI-dependent tube current selection was 27.7 HU for the BMI >27 subgroup and 31.7 HU for the BMI <27 subgroup. These values were within 5% of the presets. Fig 3 shows the scatter plots of the image noise as a function of patient BMI for groups B and C. The overall image noise and IQS in group B was statistically the same as that in group C (p ¼ 0.10 and p ¼ 0.42; Table 2). Adequate SNR values of 17 and 14.5 were obtained for the 100 kV subgroup (BMI <27) and 120 kV subgroup (BMI >27) in group B, respectively. In addition, there was no difference in the overall SNR values between groups B and C (16.07 versus 16.71; Table 2). The CTDIw distributions for patients in groups B and C are shown in Fig 4. The average effective dose values for patients in groups B and C were 2.56 0.88 and 4.38 0.82 mSv, respectively. A 42% dose reduction was achieved on average and a dose reduction of approximately 69% (1.14 versus 3.7 mSv) was achieved for the smallest patient in the present cohort with the use of an adaptive tube current (Fig 5).
Discussion Table 1 Tube current values required to obtain an image noise of 33 HU [for a body mass index (BMI) of <27 with 100 kVp tube voltage] and 28 HU (for a BMI of >27 with 120 kVp tube voltage) at computed tomography coronary angiography (CTCA) based on patient BMI with a 0.35 s gantry rotation speed and 0.625 mm image section thickness. BMI (kg/m2)
Required tube voltage/ current (kV/mA) for an image noise of 33 HU
BMI (kg/m2)
Required tube voltage/current (kV/mA) for an image noise of 28 HU
18 19 20 21 22 23 24 24.5 25 26 26.5
100/225 100/260 100/295 100/335 100/380 100/420 100/470 100/490 100/515 100/570 100/595
27 27.5 28 28.5 29 30 31 32 33 34 35.5
120/430 120/450 120/470 120/490 120/505 120/550 120/590 120/635 120/680 120/730 120/800
In the present study, a quantitative low-dose imaging protocol for selecting tube voltage and current based on patient BMI in CTCA applications was proposed. The study demonstrated that by effectively combining many dosereduction techniques20 and adaptively adjusting tube voltage and current based on the patient’s BMI value, one Table 2 Statistical analysis for parameters in group B [with body mass index (BMI)dependent tube current values] and Group C (with fixed tube current values). Group B BMI Image-quality score Image noise Signal-to-noise ratio Doseelength product (mGy$cm) Effective dose (mSv)
25.34 4.52 30.23 16.07 91.43
Group C 2.85 0.59 2.79 2.75 31.43
2.56 0.88
25.76 4.6 29.01 16.71 156.43
p-Value 2.82 0.55 3.67 3.55 29.29
4.38 0.82
0.3 0.42 0.1 0.16 0 0
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
5
Figure 3 Image noise variation as function of patient BMI in group B [(aeb) with BMI-dependent tube current values] and group C [(ced) with fixed tube current values].
could optimize dose delivery to patients and obtain consistent and desired image quality. The use of a prospective triggering scan mode provides the initial huge dose reduction compared with the conventional retrospective-gated helical scan mode,4,10 which requires the use of sufficient b-blockers to control heart rate and heart rate variability.20 For group B, patient heart rates were controlled <65 beats/min, which allowed the use of the prospective-triggering mode for all patients in this group. Furthermore, due to the low and steady heart rates, the padding time for the prospective triggering could be reduced from the initial 50 ms to 0 ms for the study group. The reduction in padding time translated to a 30% dose reduction for the patients in the study group. Scan range could also be optimized to further reduce the radiation dose to patients. In the present study, the number of steps was reduced from four to three for approximately 16% of patients by tightly controlling the scan range, which yielded an additional 4% dose reduction to the study population.
Scan protocol optimization does not necessary mean dose reduction in all cases; in some cases a higher dose was required. However, it does optimize the patient dose distribution, and for smaller patients, especially for those with BMI <20, the tube current should be greatly reduced.21,22 In fact, for patients in group B with patient-dependent tube current selection, a wide range of tube currents was required. For some large patients, a high radiation dose (approximately 5 mSv) was required using the retrospective scan mode to produce the desired image quality. On the other hand, for smaller patients, the dose reduction could be very dramatic. An effective dose of only 1.14 mSv was achieved for the patient with a BMI of 17.1 kg/m2 in the present cohort with acceptable image quality, resulting in an almost 70% dose reduction. As demonstrated in Figs 3 and 4 for group C where a constant tube current was used, the effective dose was independent of patient BMI but the image noise had strong dependency on patient BMI values. Image noise ranged from 22e38 HU. This would
Figure 4 CTDIw values as function of patient BMI in group B (left with BMI-dependent tube current values) and group C (right with fixed tube current values). Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
6
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
Figure 5 An example of patient-dependent tube current/voltage adjustment that enabled excellent image quality for a large patient (BMI ¼ 32.1 kg/m2, CTDIw ¼ 12.2 mGy) and substantial dose reduction with equally good image quality for a smaller patient (BMI ¼ 19.5 kg/m2, CTDIw ¼ 3.39 mGy).
indicate over-exposure for smaller patients and possible under-exposure for large patients. On the other hand, for group B with the proposed adaptive tube current selection, image noise had a narrower range of 24e35 HU. The line fitted to the image noise and patient BMI for group B had a slope of close to zero indicating image noise in this group was independent of patient BMI values. The image noise level selection does have a very big impact on the overall dose delivery to patient population. Lower noise levels means higher overall doses and vice versa. On the other hand, clinical practice indicated that the incremental impact of image noise to diagnostic ability is limited when image noises change within a certain range.23 This provided us with the ability to moderately decrease the image noise requirement without negatively impacting diagnostic outcomes. Finding the highest image noise (lowest x-ray dose) for acceptable diagnostic quality would be desirable and would require systematic studies. Such systematic studies would be facilitated by the quantitative tube currenteBMI formula proposed in this study, which enabled a desired noise level to be achieved more consistently across the patient population. The correlation between cardiac image noise and patient BMI is only moderate. This is because BMI is only an average value. It does not specifically reflect the fat and muscle composition of the chest where the heart is located. Even with same BMI value, different patients could have different shapes and different fat and muscle distributions, resulting in cardiac image noise deviations from the pre-selected noise value. In the future, the noise prediction accuracy could be improved by incorporating individual information such as gender and age. In addition, more robust methods for selecting the tube voltage and current could be evaluated to obtain more consistent cardiac images. Another limitation is that the values determined in the present study are manufacturer (and been probably machine) specific given the proprietary ASIR, percent image blending, etc.,
which influence image noise; although the method could potentially be used by practitioners using other vendors equipment. In conclusion, the present study describes a quantitative and effective method for tube voltage and tube current selection based on patient BMI value for CTCA. This method provides an individualized protocol to obtain consistent image quality and optimized dose delivery across the patient population.
References 1. Taguchi K, Anno H. High temporal resolution for multislice helical computed tomography. Med Phys 2000;27:861e72. 2. Herzog C, Abolmaali N, Balzer JO, et al. Heart-rate-adapted image reconstruction in multidetector-row cardiac CT: influence of physiological and technical prerequisite on image quality. Eur Radiol 2002;12: 2670e8. 3. Choi HS, Choi BW, Choe KO, et al. Pitfalls, artifacts, and remedies in multi-detector row CT coronary angiography. RadioGraphics 2004;24: 787e800. 4. Hsieh J, Londt J, Vass M, et al. Step-and-shoot data acquisition and reconstruction for cardiac x-ray computed tomography. Med Phys 2006;33:4236e48. 5. Hurwitz LM, Reiman RE, Yoshizumi TT, et al. Radiation dose from contemporary cardiothoracic multidetector CT protocols with an anthropomorphic female phantom: implications for cancer induction. Radiology 2007;245:742e50. 6. Avinash G. Method and apparatus for enhancing discrete pixel images. US Patent, http://patft.uspto.gov/netacgi/ - h0http://patft.uspto.gov/ netacgi/ - h2; 2001. 6,208,763. 7. Jakobs TF, Becker CR, Ohnesorge B, et al. Multislice helical CT of the heart with retrospective ECG gating: reduction of radiation exposure by ECGcontrolled tube current modulation. Eur Radiol 2002;12:1081e6. 8. McCollough C, Bruesewitz M, Kofler Jr J. CT dose reduction and dose management tools: overview of available options. RadioGraphics 2006;26:503e12. 9. Leipsic J, Labounty TM, Heilbron B, et al. Adaptive statistical iterative reconstruction: assessment of image noise and image quality in coronary CT angiography. AJR Am J Roentgenol 2010;195:649e54. 10. Earls JP, Berman EL, Urban BA, et al. Prospectively gated transverse coronary CT angiography versus retrospectively gated helical technique:
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016
G. Wang et al. / Clinical Radiology xxx (2014) 1e7
11.
12.
13.
14.
15.
16. 17.
improved image quality and reduced radiation dose. Radiology 2008;246:742e53. Mahnken AH, Wildberger JE, Simon J, et al. Detection of coronary calcifications: feasibility of dose reduction with a body weight-adapted examination protocol. AJR Am J Roentgenol 2003;181:533e8. Herzog BA, Husmann L, Valenta I, et al. Determinants of vessel contrast in BMI-adapted low dose CT coronary angiography with prospective ECG-triggering. Int J Cardiovasc Imaging 2009;25:625e30. Das M, Mahnken A, Muhlenbruch G, et al. Individually adapted examination protocols for reduction of radiation exposure for 16-MDCT chest examinations. AJR Am J Roentgenol 2005;184:1437e43. Hur G, Hong S, Kim S, et al. Uniform image quality achieved by tube current modulation using SD of attenuation in coronary CT angiography. AJR Am J Roentgenol 2007;189:188e98. Qi W, Li J, Du X. Method for automatic tube current selection for obtaining a consistent image quality and dose optimization in a cardiac multidetector CT. Korean J Radiol 2009;10:568e74. Gao J, Li J, Earls J, et al. Individualized tube current selection for 64-row cardiac CTA based on analysis of the scout view. Eur J Radiol 2011;79:266e71. Marin D, Nelson RC, Schindera ST, et al. Low-tube-voltage, high-tubecurrent multidetector abdominal CT: improved image quality and
18.
19.
20.
21.
22.
23.
7
decreased radiation dose with adaptive statistical iterative reconstruction algorithm d initial clinical experience. Radiology 2010;254:145e53. Feuchtner GM, Jodocy D, Klauser A, et al. Radiation dose reduction by using 100-kV tube voltage in cardiac 64-slice computed tomography: a comparative study. Eur J Radiol 2010;75:e51e6. Gosling O, Loader R, Venables P, et al. Cardiac CT: are we underestimating the dose? A radiation dose study utilizing the 2007 ICRP tissue weighting factors and a cardiac specific scan volume. Clin Radiol 2010;65:1013e7. Raff GL, Chinnaiyan KM, Share DA, et al. Radiation dose from cardiac computed tomography before and after implementation of radiation dose-reduction techniques. JAMA 2009;301:2340e8. Hunold P, Vogt FM, Schmermund A, et al. Radiation exposure during cardiac CT: effective doses at multi-detector row CT and electron-beam CT. Radiology 2003;226:145e52. Mahnken AH, Wildberger JE, Koos R, et al. Multislice spiral computed tomography of the heart: technique, current applications and perspective. Cardiovasc Intervent Radiol 2005;28:388e99. Prasad SR, Wittram C, Shepard JA, et al. Standard-dose and 50%reduced-dose chest CT: comparing the effect on image quality. AJR Am J Roentgenol 2002;179:461e5.
Please cite this article in press as: Wang G, et al., Achieving consistent image quality and overall radiation dose reduction for coronary CT angiography with body mass index-dependent tube voltage and tube current selection, Clinical Radiology (2014), http://dx.doi.org/10.1016/ j.crad.2014.04.016