Radiation dose reduction and image quality in pediatric abdominal CT with kVp and mAs modulation and an iterative reconstruction technique

    Radiation dose reduction and image quality in pediatric abdominal CT with kVp and mAs modulation and an iterative reconstruction technique Jun-Hwee Kim, Myung-Joon Kim, Ha Yan Kim, Mi-Jung Lee PII: DOI: Reference:

S0899-7071(14)00143-0 doi: 10.1016/j.clinimag.2014.05.008 JCT 7630

To appear in:

Journal of Clinical Imaging

Received date: Revised date: Accepted date:

16 January 2014 15 May 2014 19 May 2014

Please cite this article as: Kim Jun-Hwee, Kim Myung-Joon, Kim Ha Yan, Lee Mi-Jung, Radiation dose reduction and image quality in pediatric abdominal CT with kVp and mAs modulation and an iterative reconstruction technique, Journal of Clinical Imaging (2014), doi: 10.1016/j.clinimag.2014.05.008

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Radiation dose reduction and image quality in pediatric abdominal CT with kVp and

T

mAs modulation and an iterative reconstruction technique

RI P

Jun-Hwee Kim, M.D.1, Myung-Joon Kim, M.D., Ph.D.1, Ha Yan Kim, M.S.2, Mi-Jung Lee, M.D., Ph.D.1

of Radiology and Research Institute of Radiological Science, Severance

SC

1Department

Children's Hospital, Yonsei University College of Medicine, Seoul, Korea Collaboration Unit, Yonsei University, Severance Hospital

MA NU

2Biostatistics

Abbreviated title: SAFIRE in pediatric abdominal CT Corresponding author: Mi-Jung Lee, M.D., Ph.D.

Department of Radiology and Research Institute of Radiological Science Severance Children’s Hospital, Yonsei University College of Medicine

ED

50-1, Yonsei-ro, Seodaemun-gu, 120-752 Seoul, Korea

FAX: 82-2-393-3035

PT

TEL: 82-2- 2228-7400

AC

CE

E-mail: [email protected]

1

ACCEPTED MANUSCRIPT Abstract Objective: The objective of this study was to compare the radiation dose and image quality of

T

pediatric abdominal CT using a protocol reconstructed with filtered back projection (FBP)

RI P

and a protocol with both kVp and mAs modulation and sinogram–affirmed iterative reconstruction (SAFIRE).

SC

Materials and methods: We retrospectively reviewed pediatric abdominal CT examinations

MA NU

performed with both kVp and mAs modulation. These raw data were reconstructed with SAFIRE at different strengths from 2 to 4 (SAFIRE group 2-4). Another set of age/sexmatched pediatric abdominal CT examinations were also reviewed, which were performed during the same period with only mAs modulation and FBP (control group). The radiation

ED

dose and image quality were compared between groups. The image quality was objectively evaluated as the noise measured in the liver, aorta, and spleen at the level of the main portal

PT

vein and the image quality was subjectively reviewed by two radiologists for diagnostic

CE

acceptability using a four-point scale (0: unacceptable; 1: worse than the control group, but acceptable; 2: comparable with the control group; and 3: better than the control group). An

AC

independent t-test was used in order to compare the radiation dose. An independent t-test with Bonferroni correction and generalized estimating equations were used for the comparison of the objective and subjective image quality, respectively. Results: Twenty-nine patients (M:F = 19:10, mean age 10.0) were enrolled in each group. The SAFIRE group, using the size-specific dose estimates calculation method showed a 64.2% radiation dose reduction (from 8.1 to 2.9 mGy, p<0.05), compared with the results of the control group. The objective image noise of the SAFIRE groups 2 and 3 was comparable to that of the control group. The subjective image quality was the best in SAFIRE group 3 (odds ratio [OR] 3.015, p<0.001 when comparing to SAFIRE group 0; and OR 1.513, p<0.001 2

ACCEPTED MANUSCRIPT when comparing to SAFIRE group 2). Conclusions: Image acquisition with both kVp and mAs modulation and iterative

T

reconstruction using SAFIRE with strength 3 can preserve the objective and subjective image

RI P

quality of pediatric abdominal CT scans with less than half the radiation dose.

MA NU

SC

Index Terms: Pediatric, Computed tomography, Radiation dose, Iterative reconstruction

Introduction

There are increased concerns about radiation exposure from computed tomography (CT)

ED

especially in pediatric patients [1, 2]. It is well known that children are more susceptible to

of malignancy [3].

PT

radiation than adults, because of their rapid cellular proliferation and increased lifetime risk

CE

One of the best methods for reducing radiation exposure is avoiding unnecessary CT evaluation and minimizing multi-phase scanning. However, CT is still a useful diagnostic tool

AC

in children when considering its non-invasiveness and accuracy with short acquisition time. In addition, recent advances in CT technology have suggested several methods for reducing radiation exposure. The automated modulation of tube voltage and current according to a patient’s body size has been suggested as one of the radiation dose reduction methods [4, 5]. However, lowering the tube voltage and current may increase the image noise, thus lowering the image quality and diagnostic accuracy [2]. Recently, several iterative reconstruction techniques have been suggested as post image acquisition processing methods to decrease the image noise in low-dose CT scans [6, 7]. These techniques can enhance the image quality by decreasing the image noise by the 3

ACCEPTED MANUSCRIPT introduction of multiple correction loops into the reconstruction process. One of these techniques is sinogram-affirmed iterative reconstruction (SAFIRE). This technique utilizes

T

both projection space data and image space data with two different correction loops. The

RI P

corrected image is compared to the original with a number of iterations in each space dependent on the needs of a specific scan. Thanks to advanced hardware and software, these

SC

complicated imaging processes can be handled faster and easier nowadays and we can use

MA NU

these clinically.

There have been many recent studies using variable iterative reconstruction techniques. However, a majority of these studies have been performed in adults and little is known about these techniques in children even though radiation dose reduction has greater clinical

ED

importance in children. In this study, we retrospectively compared the radiation dose and image quality in pediatric abdominal CT scans using a protocol with only mAs modulation

PT

and reconstruction with filtered back projection (FBP) and a protocol with both kVp and mAs

AC

CE

modulation and reconstruction with SAFIRE.

Materials and Methods Patient selection This single-center retrospective study was approved by our hospital institutional review board and informed consent was waived. We retrospectively reviewed abdominal CT images performed in pediatric patients with a protocol using both tube current and voltage modulation and iterative reconstruction using SAFIRE (SAFIRE group) in our hospital from January to February 2012. Other age/sex-matched pediatric abdominal CT examinations were also reviewed which were performed during the same time period with a protocol using only 4

ACCEPTED MANUSCRIPT tube current modulation and FBP reconstruction (control group) in the emergency room. Only

RI P

Scanning technique and radiation dose measurements

T

single phase (portal venous phase) examinations were included in this study.

Our hospital has a 64 channel CT scanner (Sensation 64; Siemens Healthcare, Forchheim,

SC

Germany) with only the tube current modulation program (CARE Dose4D) and no iterative

MA NU

reconstruction in the emergency room. This tube current modulation program automatically adapts radiation dose to the size and shape of the patient based on the topogram, defined reference setting, and real-time angular dose modulation. For the reference setting, this technique uses the term of a quality reference mAs as a defined value that we would use for a

ED

standard sized adult patient weighing 75 kg. We used a quality reference mAs of 100mAs for abdominal CT in children. We scanned pediatric abdominal CTs with 80 kVp for children less

PT

than 10 kg, 100 kVp for children between 10-40 kg, and 120 kVp for children more than 40

CE

kg according to body weight. The acquisition parameters were 64 x 0.6-mm detector collimation, 1.4 pitch factor, and 0.33-sec gantry rotation. The slice thickness and increment

AC

were 2 mm each.

Our hospital also has a dual source CT scanner (Somatom Definition Flash; Siemens Healthcare), which has automated tube voltage and current modulation program (CARE kV and CARE Dose4D; Siemens Healthcare), and the iterative reconstruction program named SAFIRE (Siemens Healthcare). The tube voltage modulation program uses information gathered by the topogram and automatically recommends the optimally low tube voltage for the entire length of the scan while keeping the contrast-to-noise ratio. A single tube voltage (kV) is selected for each exam. It also uses a reference value as a quality reference kV set for a standard patient of 75 kg. We have used an abdominal CT protocol with a quality reference 5

ACCEPTED MANUSCRIPT kV of 100 kVp and a quality reference mAs of 100mAs for children since January 2012. The acquisition parameters were 64 x 0.6-mm detector collimation, 1.4 pitch factor, and 0.28-sec

RI P

standard soft tissue kernel (B30f) for image reconstruction.

T

gantry rotation. The slice thickness and increment were 2 mm each. Both CT scanners used

The protocols of the contrast media injection and patient preparation were the same when

SC

performing these two CT examinations. The abdominal CT scans were performed with

MA NU

300mg iodine/ml concentration intravenous contrast iobitridol (Xenetix, Laboratories Guerbet, Roissy, France). For contrast enhancement, 2 ml/kg of contrast was injected through an upper extremity peripheral intravenous line followed by a saline chaser of 0.5 ml/kg. The scanning was performed 40 seconds after enhancement in the abdominal aorta reached 100

ED

Hounsfield units (HU) for the portal venous phase only.

The volume CT dose indices (CTDIvol) and dose length product (DLP) values were recorded

PT

on the dose page for each of the studies. We multiplied the CTDIvol values by the conversion

CE

factors for size-specific dose estimates (SSDE) for the pediatric patients according to the

AC

American Association of Physicists in Medicine Report 204 [8].

Iterative reconstruction The SAFIRE iteration strength can be changed according to the strength of the noise reduction. The iteration strength can be ordered with an integral scale of 1 to 5, with 1 being the weakest and 5 being the strongest iteration strength. Two pediatric radiologists (M.J.K. and M.J.L.) with more than 30 and 10 years of experience with pediatric abdominal CT preliminarily reviewed several images with all 1 to 5 SAFIRE strengths and subjectively determined that images with SAFIRE strengths of 1 and 5 are too noisy or too smooth, respectively. Therefore, we selected only SAFIRE strengths 2, 3, and 4 for the analyses 6

ACCEPTED MANUSCRIPT (SAFIRE groups 2-4).

T

Objective and subjective image quality analysis

RI P

We reviewed each study of the control group and the SAFIRE groups 2-4 using the picture archiving and communication systems of our institution (Centricity Radiology RA1000, GE

SC

Medical Systems, Milwaukee, WI). For quantitative image quality analysis, we obtained

MA NU

noise measurements for each study. A single circular region of interest (ROI) that measured at least 3cm2 was drawn in the homogenous regions of the right hepatic lobe, abdominal aorta, and spleen at the level of the main portal vein. We measured the noise using standard deviations (SDs) of the attenuation value of the ROI [9].

ED

For subjective image quality analysis, two radiologists (M.J.L. and J.H.K.) with 10 and 4 years of experience with pediatric abdominal CT reviewed all images in the study of each

PT

patient directly and randomly without technical or personal medical information. We used the

CE

standard soft tissue window settings (400HU of window width and 40HU of window level) and carefully checked the anatomic structures of the abdominal organs and vessels as

AC

previously suggested [10]. Each radiologist independently graded each study of the raw data from the dose modulation protocol without SAFIRE (SAFIRE group 0) and with SAFIRE strengths 2, 3, and 4 (SAFIRE groups 2-4) for diagnostic acceptability using a four-point scale in comparison with the age/sex-matched control group. Grade 0 was unacceptable; grade 1 was worse than the control group, but acceptable; grade 2 was comparable with the control group; and grade 3 was better than the control group. We reviewed the images of each SAFIRE group from one patient together in one monitor compared with the images of the control group.

7

ACCEPTED MANUSCRIPT Statistical analysis The data were analyzed using the Statistical Analysis System (version 9.2, SAS Institute Inc.,

T

Cary, NC, USA). The patients’ height, weight, body mass index (BMI), and radiation dose

RI P

were compared using an independent t-test. The objective image noise was compared using an independent t-test with Bonferroni correction. For subjective image quality analysis, the

SC

generalized estimating equation was used in order to compare SAFIRE groups 2-4 with

MA NU

SAFIRE group 0 and SAFIRE group 2 with SAFIRE group 3. The p-value was adjusted for age, sex, and reviewer with the Bonferroni correction. A p-value of less than 0.05 was

Results

PT

Patients and radiation dose

ED

considered statistically significant.

CE

There were 29 pediatric patients enrolled in the SAFIRE group. In addition, another 29 age/sex-matched patients who underwent single-phase abdominal CT scans in the emergency

AC

room in 2011 were enrolled in the control group. Therefore, a total of 58 pediatric patients were enrolled in this study. There were 19 boys and 10 girls in each group. The mean age of each group was 10.0 years (range: 0-16 years). The indications for an abdominal CT scan in the SAFIRE group were the evaluation of abdominal pain and/or fever in 17 patients, an abdominal mass in seven patients, postoperative evaluation in three patients, and the preoperative evaluation of congenital anomalies in two patients. In the control group, 28 patients received an abdominal CT scan for evaluation of abdominal pain and one patient received a scan due to minor injury after a traffic accident. Table 1 demonstrates the results of the demographics and radiation dose in the SAFIRE group 8

ACCEPTED MANUSCRIPT and the control group. Height, weight, and BMI did not differ between the two groups. The radiation dose was significantly lower in the SAFIRE group in comparison with the control

T

group (p<0.001). The average tube voltage of the SAFIRE group was 100 kVp compared to

RI P

111.7 kVp in the control group, showing a 10.5% reduction by using CARE kV. The average SSDE of the SAFIRE group was 2.9 mGy compared to 8.1 mGy in the control group,

SC

resulting in a 64.2% dose reduction. The average DLP of the SAFIRE group was 137.8 mGy

MA NU

x cm compared to 389.0 mGy x cm in the control group, showing a 64.6% reduction in the SAFIRE group.

Objective image quality

ED

The image noise was higher in the SAFIRE group 0 compared to that in the control group (Table 2). And the noise showed a tendency to decrease according to the increased SAFIRE

PT

strength. When compared with the control group, SAFIRE groups 2 and 3 showed

CE

comparable image noise in almost all of the areas except in the spleen in SAFIRE group 3.

group.

AC

There was less image noise in the spleen in SAFIRE group 3 compared with the control

Subjective image quality The results of the subjective image quality assessments are summarized in Table 3. When compared to the SAFIRE group 0, SAFIRE groups 2 (Odds ratio [OR] 1.993, p<0.001) and 3 (OR 3.015, p<0.001) showed better image quality in diagnostic acceptability. However, SAFIRE group 4 showed no significant improvement in the image quality in comparison with SAFIRE group 0 (OR 1.109, p=0.082) (Figure 1). When comparing between SAFIRE group 2 and 3, SAFIRE group 3 was better in subjective image quality (OR 1.513, p<0.001). 9

ACCEPTED MANUSCRIPT Discussion Making the image quality acceptable while reducing the radiation exposure is one of the most

T

important issues in regards to CT examinations. Recent iterative reconstruction technologies,

RI P

including SAFIRE, can lead to novel post-processing techniques in order to reduce radiation dose and improve image quality. However, there are limited studies of the use of these

SC

techniques in children. In addition, there is no definite guideline of the optimal iterative

MA NU

reconstruction strength for the best image quality. Our study showed that this technique is also useful in pediatric patients with the effect of a radiation dose reduction of more than 60%. We also found that iterative reconstruction strength 3 can be the best for both the objective and subjective image quality for pediatric abdominal CTs.

ED

Among the variable methods to reduce radiation dose, automatic kVp and mAs modulation can both be options. The use of mAs modulation automatically adapts the tube current in both

PT

the angular and longitudinal directions according to the patient’s size and thickness in order

CE

to maintain predefined image noise or image quality characteristics, and it is an important technique for reducing radiation dose [11, 12]. In addition, the use of the optimal tube

AC

potential can be another important technique, even though there have only been a few studies on the effects of kVp modulation on radiation dose reduction [13, 14]. While the use of kVp modulation resulted in a 10.5% reduction of the mean tube voltage, its overall effect on radiation dose cannot be determined due to the simultaneous use of mAs modulation. Further studies to clarify the effect of automatic kVp modulation technique on dose reduction and the evaluation for the possibility of additional dose reduction and dose protocol optimization using this technique are needed. Iterative reconstruction techniques can be another method used to reduce the radiation dose. There have been several studies using these techniques even in pediatric CT scans. A majority 10

ACCEPTED MANUSCRIPT of these studies were performed using Adaptive Statistical Iterative Reconstruction (ASIR, GE healthcare). Vorona et al. [15] reported a 33% dose reduction in pediatric abdominal CTs

T

and Singh et al. [16] reported 46% and 38% dose reductions in pediatric chest and abdominal

RI P

CT scans, respectively, using this technique. However, only a few studies have been performed using SAFIRE in children for cardiac CTs [17, 18].

SC

There have been several recent studies focusing on the use of SAFIRE in adults. However,

MA NU

these studies were primarily performed for coronary CTs [19, 20] and only a few studies were performed for abdominal CTs [21-23]. Kalra et al. [21] and Fletcher et al. [22, 23] demonstrated that SAFIRE can reduce the radiation dose by approximately 50% for abdominal CT or CT colonography in adult patients without a significant loss of image

ED

quality. In addition, there is no consensus about the optimal iteration strength using this technique. Fletcher et al. [22] mentioned that the diagnostic confidence and image quality

PT

scores can depend on the reader and they did not evaluate the optimal iteration strength. Kalra

CE

et al. [21] also found that there was a minor blocky, pixelated appearance in the reconstructed images with SAFIRE strengths 3 and 4 without suggesting the optimal reconstruction

AC

strength. In our study, there was also a minor blocky, pixelated appearance in the SAFIRE group 3 images. However, the reviewers determined that SAFIRE group 3 was better in subjective image quality compared with SAFIRE group 0 and group 2. To the best of our knowledge, this study represents the first clinical evaluation of SAFIRE with the suggestion of the optimal reconstruction strength in pediatric abdominal CT. We evaluated both the objective and subjective image quality in order to determine the optimal SAFIRE strength in pediatric abdominal CT and concluded that SAFIRE strength 3 is the most optimal. In the objective image quality analyses, the image noise decreased according to the increased SAFIRE strength. And the noise of the images with SAFIRE strengths 2 and 3 11

ACCEPTED MANUSCRIPT was comparable with that of the control group. In the subjective image quality analyses, we found that SAFIRE strength 3 can also provide the most acceptable images. However,

T

diagnostic acceptability is not the same as diagnostic accuracy. There can be criticisms of

RI P

iterative reconstruction which include the decrease of image sharpness and the limited detection of fine anatomical detail. These can affect not only image quality but also

SC

diagnostic accuracy. Therefore, further study about the effects of the iteration strength on

MA NU

diagnostic accuracy is needed.

There were several limitations in our study. First, we retrospectively reviewed the CT studies between two different groups with a small number of subjects and using two different CT scanners. Although the patients’ height, weight, and BMI between the two groups were not

ED

significantly different, this small difference could have affected the inter-individual variability. Moreover, we compared the images of each SAFIRE group from one patient together in one

PT

monitor which can cause bias. The indications were also different between the two groups.

CE

Almost all patients underwent abdominal CT due to abdominal pain in the control group when there were variable indications in the SAFIRE group. It is because we selected the

AC

control CT examinations performed in the emergency room. These different indications also can affect image quality. However, doing repeated CT scans in the same child for only academic purposes is not ethically appropriate. Therefore, our study design was quite acceptable and necessary for the safety of the pediatric patients. The second limitation was the relatively simple scoring system of the subjective image quality analysis. We only used four grades to evaluate the diagnostic acceptability. We did not evaluate the lesion detectability or diagnostic performance. This was due to the variable purposes of the CT scans in our patients. Additional studies with a large number of patients are needed in order to evaluate the diagnostic performance and accuracy. The third limitation of this study was that 12

ACCEPTED MANUSCRIPT we only examined the portal venous phase images of the contrast-enhanced CTs. This is the routine protocol of pediatric abdominal CT scans in our hospital in order to avoid

T

unnecessary multiphase scans. However, the image noise and quality can vary in the pre-

RI P

contrast images or in different phases of post-contrast images. Further studies are also needed

SC

in order to evaluate the effects of iterative reconstruction according to scan phase.

MA NU

In conclusion, both kVp and mAs modulation protocols can significantly reduce the radiation dose and iterative reconstruction using SAFIRE can preserve image quality in pediatric abdominal CT scans. The most optimal SAFIRE strength was 3 for both the objective and

AC

CE

PT

ED

subjective image quality.

13

ACCEPTED MANUSCRIPT References [1]

Miglioretti DL, Johnson E, Williams A, Greenlee RT, Weinmann S, Solberg LI, et al.

[2]

RI P

and estimated cancer risk. JAMA Pediatr 2013;167:700-7.

T

The use of computed tomography in pediatrics and the associated radiation exposure

Brenner DJ, Elliston CD, Hall EJ, Berdon WE. Estimates of the cancer risks from

SC

pediatric CT radiation are not merely theoretical: comment on "point/counterpoint: in

MA NU

x-ray computed tomography, technique factors should be selected appropriate to patient size. against the proposition". Med Phys 2001;28:2387-8. [3]

Brenner DJ, Hall EJ. Computed tomography--an increasing source of radiation exposure. N Engl J Med 2007;357:2277-84.

Heyer CM, Mohr PS, Lemburg SP, Peters SA, Nicolas V. Image quality and radiation

ED

[4]

exposure at pulmonary CT angiography with 100- or 120-kVp protocol: prospective

Kalra MK, Maher MM, Toth TL, Schmidt B, Westerman BL, Morgan HT, et al.

CE

[5]

PT

randomized study. Radiology [Randomized Controlled Trial] 2007;245:577-83.

Techniques and applications of automatic tube current modulation for CT. Radiology

[6]

AC

[Review] 2004;233:649-57. Scheffel H, Stolzmann P, Schlett CL, Engel LC, Major GP, Karolyi M, et al. Coronary artery plaques: cardiac CT with model-based and adaptive-statistical iterative reconstruction technique. Eur J Radiol 2012;81:e363-9. [7]

Utsunomiya D, Weigold WG, Weissman G, Taylor AJ. Effect of hybrid iterative reconstruction technique on quantitative and qualitative image analysis at 256-slice prospective gating cardiac CT. Eur Radiol 2012;22:1287-94.

[8]

Boone JM, Strauss KJ, Cody DD, McCollough CH, McNitt-Gray MF, Toth TL. Sizespecific dose estimates (SSDE) in pediatric and adult body CT examinations. Report 14

ACCEPTED MANUSCRIPT of AAPM Task Group 204, 2011. [9]

Lee SH, Kim MJ, Yoon CS, Lee MJ. Radiation dose reduction with the adaptive

[10]

RI P

individual comparison. Eur J Radiol 2012;81:e938-43.

Group. ES. European guidelines on quality criteria for computed tomography: EUR

SC

16262.

Kalra MK, Maher MM, Toth TL, Hamberg LM, Blake MA, Shepard JA, et al.

MA NU

[11]

T

statistical iterative reconstruction (ASIR) technique for chest CT in children: an intra-

Strategies for CT radiation dose optimization. Radiology 2004;230:619-28. [12]

Yu L, Bruesewitz MR, Thomas KB, Fletcher JG, Kofler JM, McCollough CH. Optimal tube potential for radiation dose reduction in pediatric CT: principles, clinical

[13]

ED

implementations, and pitfalls. Radiographics 2011;31:835-48. Niemann T, Henry S, Faivre JB, Yasunaga K, Bendaoud S, Simeone A, et al. Clinical

PT

evaluation of automatic tube voltage selection in chest CT angiography. Eur Radiol

[14]

CE

2013;23:2643-51.

Brady SL, Moore BM, Yee BS, Kaufman RA. Pediatric CT: Implementation of ASIR

AC

for Substantial Radiation Dose Reduction While Maintaining Pre-ASIR Image Noise. Radiology 2014;270:223-31. [15]

Vorona GA, Ceschin RC, Clayton BL, Sutcavage T, Tadros SS, Panigrahy A. Reducing abdominal CT radiation dose with the adaptive statistical iterative reconstruction technique in children: a feasibility study. Pediatr Radiol 2011;41:117482.

[16]

Singh S, Kalra MK, Shenoy-Bhangle AS, Saini A, Gervais DA, Westra SJ, et al. Radiation dose reduction with hybrid iterative reconstruction for pediatric CT. Radiology 2012;263:537-46. 15

ACCEPTED MANUSCRIPT [17]

Han BK, Grant KL, Garberich R, Sedlmair M, Lindberg J, Lesser JR. Assessment of

CT datasets. J Cardiovasc Comput Tomogr 2012;6:200-4.

Tricarico F, Hlavacek AM, Schoepf UJ, Ebersberger U, Nance JW, Jr., Vliegenthart

RI P

[18]

T

an iterative reconstruction algorithm (SAFIRE) on image quality in pediatric cardiac

R, et al. Cardiovascular CT angiography in neonates and children: image quality and

SC

potential for radiation dose reduction with iterative image reconstruction techniques.

[19]

MA NU

Eur Radiol 2013;23:1306-15.

Wang R, Schoepf UJ, Wu R, Reddy RP, Zhang C, Yu W, et al. Image quality and radiation dose of low dose coronary CT angiography in obese patients: sinogram affirmed iterative reconstruction versus filtered back projection. Eur J Radiol

[20]

ED

2012;81:3141-5.

Ebersberger U, Tricarico F, Schoepf UJ, Blanke P, Spears JR, Rowe GW, et al. CT

PT

evaluation of coronary artery stents with

iterative image reconstruction:

CE

improvements in image quality and potential for radiation dose reduction. Eur Radiol 2013;23:125-32.

Kalra MK, Woisetschlager M, Dahlstrom N, Singh S, Lindblom M, Choy G, et al.

AC

[21]

Radiation dose reduction with Sinogram Affirmed Iterative Reconstruction technique for abdominal computed tomography. J Comput Assist Tomogr 2012;36:339-46. [22]

Fletcher JG, Krueger WR, Hough DM, Huprich JE, Fidler JL, Wang J, et al. Pilot study of detection, radiologist confidence and image quality with sinogram-affirmed iterative reconstruction at half-routine dose level. J Comput Assist Tomogr 2013;37:203-11.

[23]

Fletcher JG, Grant KL, Fidler JL, Shiung M, Yu L, Wang J, et al. Validation of dualsource single-tube reconstruction as a method to obtain half-dose images to evaluate 16

ACCEPTED MANUSCRIPT radiation dose and noise reduction: phantom and human assessment using CT colonography and sinogram-affirmed iterative reconstruction (SAFIRE). J Comput

AC

CE

PT

ED

MA NU

SC

RI P

T

Assist Tomogr 2012;36:560-9.

17

ACCEPTED MANUSCRIPT Table 1. Demographics and radiation dose of the control group and the SAFIRE group. Control group

*p-value

Height (cm)

138.3 ± 24.6

137.7 ± 23.8

0.925

Weight (kg)

35.6 ± 14.3

38.3 ± 17.7

17.7 ± 2.1

19.5 ± 7.3

0.207

100.0± 7.6

mass

RI P

Body

index

0.522

SC

(kg/m2)

<0.001

2.9 ± 0.8

8.1 ± 4.6

<0.001

137.8 ± 54.5

389.0 ± 259.2

<0.001

111.7± 10.0

MA NU

Tube voltage (kVp) SSDE(mGy) DLP (mGy x cm)

T

SAFIRE group

ED

Note: data are shown in mean ± standard deviation

SAFIRE: sinogram–affirmed iterative reconstruction

PT

SSDE: size-specific dose estimates

CE

DLP: dose length products

AC

*from an independent t-test

18

ACCEPTED MANUSCRIPT

Table 2. Comparison of the image noise between the control group and each SAFIRE group. Data represents mean value (± standard

Noise

SAFIRE 2

SAFIRE 3

CR I

Noise

SAFIRE 0 *p-value

Noise

*p-value

Noise

NU S

Control

PT

deviation) of the image noise measured as the standard deviation of the CT attenuationin each area. SAFIRE 4

*p-value

Noise

*p-value

12.3 (±2.5) 17.4 (±3.2)

<0.001

13.7 (±2.5)

0.152

11.9 (±2.5)

>0.999

10.2 (±2.3)

0.008

Aorta

15.7 (±3.7) 19.6 (±4.1)

0.001

16.3 (±4.1)

>0.999

13.9 (±3.8)

0.309

12.2 (±3.5)

0.002

Spleen 12.8 (±2.6) 15.6 (±2.7)

0.001

12.9 (±2.3)

>0.999

0.027

9.6 (±1.8)

<0.001

MA

Liver

D

11.2 (±2.0)

TE

SAFIRE 0: CT group using both automatic kVpand mAsmodulation without SAFIRE

CE P

SAFIRE 2, 3, and 4: CT group using both automatic kVp and mAs modulation and reconstructed with SAFIRE strengths 2, 3, and 4

AC

*from an independent t-test with Bonferroni correction

19

ACCEPTED MANUSCRIPT Table 3. Subjective image quality analysis between the SAFIRE groups *p-value

SAFIRE 2 vs. SAFIRE 0

1.993 (1.769-2.245)

<0.001

SAFIRE 3 vs. SAFIRE 0

3.015 (2.750-3.304)

SAFIRE 4 vs. SAFIRE 0

1.109 (1.012-1.216)

SAFIRE 3 vs. SAFIRE 2

1.513 (1.333-1.717)

T

Odds ratio (95% Confidence interval)

SC

RI P

Comparison of quality grades

<0.001 0.127 <0.001

MA NU

*from a Generalized Estimating Equation with adjusted for age, sex, and reviewer with

AC

CE

PT

ED

Bonferroni correction.

20

ACCEPTED MANUSCRIPT Figure legends Figure 1. Abdominal CT images at the level of the main portal vein of two 11-year-old boys

T

for the evaluation of abdominal pain. A is a routine radiation dose image with filtered back

RI P

projection in the control group. B is an image using both automatic kVp and mAs modulation without SAFIRE reconstruction in the SAFIRE group. C-E are images reconstructed with

AC

CE

PT

ED

MA NU

SC

SAFIRE strength 2 (C), 3 (D), and 4 (E).

21

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Figure 1A

22

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Figure 1B

23

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Figure 1C

24

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Figure 1D

25

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Figure 1E

26