Usefulness of large beam-shaping filters at different tube voltages of newborn chest CT

Background: To investigate optimizing the use of different beam shaping filters (viz. small, medium and large) when using different tube voltages during the newborn chest computed tomography (CT) on a GE Lightspeed VCT scanner. Methods: We used pediatric anthropomorphic phantoms with a 64 detector-row CT scanner while scanning the chest. A real-time skin dosimeter (RD − 1000; Trek Corporation, Kanagawa, Japan) was positioned into the phantom center of the body, the surface of the body back, and the right and left mammary glands. We performed and compared six scan protocols using small, medium, and large beam shaping filters at 80 and 120 kVp protocols. Result: There were no significant differences in the image noise for the chest scan among the different beam shaping filters. By using the large beam shaping filter at 80 kVp, it was possible to reduce the exposure dose by 5% in comparison with the small beam shaping filter, and by 10% in comparison with the medium beam shaping filter. By using the large beam shaping filter at 120 kVp, it was possible to reduce the exposure dose by 15% in comparison with the small beam shaping filter and by 20% in comparison with the medium beam shaping filter (p < 0.01). Conclusion: The large beam shaping filter had the most dose reduction effect during newborn chest CT on a GE Lightspeed VCT scanner. The additional copper filtration being present in the large bowtie filter of the GE Lightspeed CT scanner when using different tube voltages is more effective in reducing radiation exposure in children.

been previously investigated [12,13]. Another method of radiation dose reduction was the beam shaping filter [14]. The use of a beam shaping filter can reduce the dose by about 50% at the surface in comparison to having just a flat filter in the beam [15]. The selection of beam shaping filters has not been verified to be optimal in newborn CT. The usefulness of beam shaping filters when using low tube voltages has also not been verified.
The purpose of this study was to investigate and optimize the use of small, medium, and large beam shaping filters when using different tube voltages during the newborn chest CT on a GE Lightspeed VCT scanner. We propose that the additional copper filtration being present in the large bowtie filter of the GE Lightspeed CT scanner when using different tube voltages is more effective in reducing radiation exposure in children.

Phantoms
We used a pediatric anthropomorphic phantom (ATOM Phantom, CIRS, Norfolk, Virginia, USA), which represented an average newborn (Fig. 1). The assumed body weight and body height for the newborn were 3.5 kg and 51.0 cm respectively. The phantom was made of radiologically equivalent tissue materials with internal structures, including artificial skeletons, lungs, and soft tissues, formulated for the accurate simulation of clinical exposures.

CT scan
We performed the scan using a 64 detector-row CT scanner (Lightspeed VCT; GE Healthcare, Milwaukee, WI, USA) with a helical scan from the apex of the lung to the subdiaphragm. The scan range for the newborn phantom was 100 mm. The scanning parameters were helical mode, helical pitch 0.986, beam width 40.0 mm, section thickness 5.0 mm, gantry rotation time 0.4 s, 64 × 0.625 mm detector collimation, settings for the small scan field of view were 100 mm, matrix size 512 × 512 mode for full mode, and standard reconstruction kernel. The applied tube voltage was 80 kVp and 120 kVp, and the tube current was used with automatic tube current modulation (noise index 10). The CT images were acquired using filtered back projection algorithms under the standard kernel/filter. We performed six scan protocols along with the use of small, medium, and large beam shaping filters at 80 and 120 kVp protocols.

Beam shaping filter
The small and medium beam shaping filters are composed of carbon (graphite) and aluminum. The large beam shaping filter is composed of carbon (graphite), aluminum, and copper.

Dosimeters and dose measurement
A real-time skin dosimeter (RD − 1000; Trek Corporation, Kanagawa, Japan) was positioned into the phantom center of the body, the surface of the body back, and the right and left mammary glands (Fig. 2). The phantom was then scanned 10 times using each protocol. The measured dose values of the RD − 1000 were compared for each protocol.
Comparison of radiation doses from 10-cm ionization chamber and RD-1000.
Since the traceability of the RD-1000 needed to be checked with the CT equipment, the doses of the 10 cm ionization chamber and the RD-1000 were compared using the 16 cm acrylic phantom of EMF 1625 model. The results confirmed that the RD-1000 has good linearity between the 10 cm ionization chamber and the reference dose.

Image noise measurement
We measured the image noise [standard deviation (SD) of the CT number] at the center of the phantom at the level of the mammary glands area within a circumscribed 10.0 mm diameter region of interest. For each scan, we measured the image noise 5 times for a total of 5 marginal and central points. The mean value for the image noise was calculated, and the image noise was compared between each protocol.

Statistical analysis
For the analysis of measured image noise and measured dose values of the RD − 1000 in the different scan protocols, we used the Kruskal-Wallis analysis of variance. When the assumption of homogeneity of variances was not verified, we performed a Steel-Dwass analysis. The statistical significance was set at p < 0.05. Statistical analyses were performed using the free statistical software "R" (R, version 3.2.2; The R Project for Statistical Computing; http:// www.r-project.org/).

Results
There were no significant differences in the image noise for the chest scan among the different beam shaping filters and tube voltages (  Table 2). By using the large beam shaping filter at 80 kVp, it was possible to reduce the exposure dose by 5% in comparison with the small beam Table 1 Image noise for the chest CT among the small, medium, and large beam shaping filters by using the newborn phantom 80 kVp   The large beam shaping filter was useful even by using the low tube voltage during the newborn chest CT, however, the exposure dose reduction rate was low. Since Cu (copper) is used only for the beam shaping filter, the soft X-ray component is cut [20]. Therefore, we believe that the exposure reduction rate at low tube voltage is reduced during the newborn chest CT.
Our study has several limitations. First, we evaluated image quality only for pediatric patients using an anthropomorphic phantom but not on real pediatric patients. Furthermore, the influence of cardiac and respiratory motion or stair-step artifacts on the image quality was not taken into consideration. Second, we evaluated newborn phantoms as the focus in pediatric patients. With increasing age, the results might be varied due to increasing body size. Third, image noise was measured as a surrogate for image quality. However, filter size needs further study as it may affect hounsfield unit values and noise uniformity across the image such as iodine-contrast examinations in this study. Last, we performed the study using a single model of CT scanner from a single vendor. The relation between tube voltage, image noise, radiation dose, reconstruction kernel, slice thickness, tube filtration, detector system, and phantom size may depend, to some degree, on CT scan specifications, and may vary among scanners.

Conclusion
The large beam shaping filter had the most dose reduction effect during newborn chest CT on a GE Lightspeed VCT scanner. The additional copper filtration being present in the large bowtie filter of the GE Lightspeed CT scanner when using different tube voltages is more effective in reducing radiation exposure in children.
Author contributions T.M. contributed to the study design, data collection, algorithm construction, image evaluation, and the writing and editing of the article; Y.F. carried out the data collection, image evaluation, and the reviewing and editing of the article; T.N. performed supervision, project administration, image evaluation, and reviewing and editing of the article. All authors read and approved the final manuscript.
Funding The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Declarations
Competing interests The authors have no relevant financial or nonfinancial interests to disclose.  (1.86-1.90) at large beam shaping filter, respectively (p < 0.01). By using the large beam shaping filter at 120 kVp, it was possible to reduce the exposure dose by 15% in comparison with the small beam shaping filter and by 20% in comparison with the medium beam shaping filter (p < 0.01, Table 2).

Discussion
By using the large beam shaping filter, it was possible to reduce the exposure dose by 5% at 80 kVp and by 15% at 120 kVp in comparison with the small beam shaping filter and by 10% at 80 kVp and by 20% at 120 kVp in comparison with the medium beam shaping filter.
Importantly, the radiation dose differed with the selection for beam shaping filters such as small, medium, and large during newborn chest CT. Because human bodies are thinner at the edge, beam shaping filter can actually attenuate some of that beam. The impact on image noise with a beam shaping filter is not great, but there is a very substantial effect on surface dose in adults. However, in newborns, the dose reduction at the edges is less. It affects the material and thickness of the beam shaping filter. Therefore, large beam shaping filter materials could affect the reduction of the radiation dose.
The impact on image noise with a beam shaping filter is not great during newborn chest CT. Our results did not change with respect to image noise in each bowtie filter. For newborns with a small physique, the small beam shaping filter may not reduce the radiation dose at the margin. The development of a beam shaping filter specialized for newborns is desired for newborn chest CT.
The large beam shaping filter had the most dose reduction effect during newborn chest CT. The CT equipment manufacturer has many beam shaping filters with factors like deference shape and materials. Some other areas are in the design of the detector. Various approaches to managing dose such as pre-treatment bowtie beam shaping filters, beam tracking and x-ray auto exposure control systems are presently being employed [15][16][17][18][19]. Therefore, the radiation dose and image quality might be different depending on each beam shaping filter. We think it is necessary to confirm