Ⅰ. INTRODUCTION
Computed tomography (CT) involves the data obtained through the rotation of the X-ray irradiation device that are reconstructed as a cross-sectional image of the 3D structure of the human body based on computerized calculations. With the advancement of the computer technology, the CT device continuously advanced, and of note is the marked changes in medical science through the development of the multi-detector CT (MDCT) in the late 1990s[1].
Compared to the conventional single-detector CT, the MDCT enabled to reduce the testing time by extending the scope of the data collection at a single rotation of the tube[2]. The reduced testing time in turn decreases the use of enthesis for the movement, and implies the possibility of testing a larger number of patients within limited time. In the U.S., the number of CT tests exponential increased by approximately 2,133% in 26 years, from 3 million cases in 1980 to 67 million cases in 2006[3,4]. The limitation of MDCT, however, is that a quantitative analysis cannot be performed for the lesions as two materials of different atomic numbers but a similar level of attenuation coefficient cannot be differentiated[5]. To overcome this limitation, the spectral CT was developed.
In Spectral CT, the analyses of the HU and keV and of the HU and the contrast agent provide the basic data that are prerequisite to the Spectral CT. The present study thus aims to analyze the correlation between the HU and keV and between the HU and dilution ratio of the contrast agent in Spectral CT, to provide the basic data for future studies regarding Spectral CT.
Ⅱ. MATERIALS AND METHODS
In Spectral CT, the image is obtained through simultaneous but separate receipt of the high-energy and low-energy irradiations of the polyenergy from the dual-layer detector. In addition, the reconstruction of virtual monoenergetic image (VMI) in 40 keV - 200 keV range is possible so that the dose of the contrast agent can be controlled[6]. VMI allows an increase or a decrease in the Hounsfield Unit (HU) according to the unique atomic number of a substance, which makes it easy to distinguish between target substances[7]. It is thus of critical importance to understand the relationship between the HU and the atomic number of substances. It is also fundamentally essential to establish the relationship between the HU and a number of parameters that may alter the attenuation coefficient of tissues. While there may be a variety of such external factors, the most influential parameter is the contrast agent, which can alter the attenuation coefficient of tissues in a target area to create a difference in contrast from the surrounding structures[8]. The contrast agent is thus utilized in many different tests such as angiography and the differentiation between a tumor and an angioma[9-11].
1. Experimental device and phantom
In this study, a CT device (IQon Spectral CT, Philips, Netherlands) was used for imaging analysis, while a 20 cc syringe was used as the phantom. Here, a nonionic iodine contrast agent (350 mg/mL) was used. To set the syringe at the same height as the abdominal aorta, a pillow was used as a prop for support[Fig 1].
Fig. 1. IQon Spectral CT (A) and pillow to match the height of the abdominal aorta (B).
2. Experimental procedures
The contrast agent was mixed with saline in the 20 cc syringe for dilution in the ratios of 8:2, 7:3, 6:4, 5:5, 4:6, and 3:7. Each syringe after dilution was placed on the pillow and the tip of the syringe needle was set to the iso-center by adjusting the position of the table at 44 vertically and -77.5 horizontally. The CT imaging test was performed in the following conditions: tube voltage 120 kVp, tube current 105 mAs, slice thickness 2 mm, scan time 1.93 sec, FOV 50 mm, and length 57 mm, which were identical across all tests[Table 1]. For reliability, the tests were repeated ten times in the identical conditions for each dilution ratio.
Table 1. Scan Parameter
*. ST-Slice Thickness, FOV-Field of View
3. Measurements
The data obtained from the scan were reconstructed for each dilution ratio at monoenergy (MonoE) 40 keV, 45 keV, 50 keV, 55 keV, 60 keV, 65 keV, 70 keV, 75 keV, and 80 keV, using the IQon-Spectral CT V4.7.5 program, while each reconstructed image was set to the ROI of 349.4 mm square (sq) to generate 56 axial sections from each image. The HU was measured at the following three points; the first slice (-103.7 mm), the 28th slice (-77.7 mm), and the 56th slice (-50.7 mm). From the measurement data, 10 times of scan data were obtained for each of the six different dilution ratios, and through the measurements at the three phantom points on the nine reconstructed MonoE images of each test, a total of 1,620 HU values were collected.
Fig. 2. Measurement Data at dilution ratio 8:2 and 40 keV, first slice (A) and 28th slice (B) and 56th slice (C). Arrow is the chosen axial slice image of syringe.
2.4. Statistical analysis
To analyze the mean values of the HU data per dilution ratio and per keV, the two way ANOVA test was performed using a statistical software (SPSS 18.0, IBM, USA). Dunnett test was used for the post-hoc analysis, and multiple regression analysis was performed to verify the correlations. Here, statistical significance was set as p ≤ 0.05.
Ⅲ. RESULT
The measured values of CT number (HU) according to the changes of MonoE (keV) in 20 cc syringes with varying dilution ratios, as obtained using the spectral CT, are presented in [Table 2].
Table 2. The HU values according to keV and dilution rate changes
* Post hoc test with Dunnett showed p<0.05 in all variables
1. Comparison of HU according to dilution ratio per MonoE
At 40 keV, the highest HU and the lowest HU were 3040.62±4.86 at dilution ratio 8:2 and 2948.91±4.37 at dilution ratio 3:7, respectively (p<0.05). At 45 keV, the highest HU and the lowest HU were 3024.50±4.58 at dilution ratio 8:2 and 2912.66±4.23 at dilution ratio 3:7, respectively (p<0.05). At 50 keV, the highest HU and the lowest HU were 3004.54±4.49 at dilution ratio 8:2 and 2872.24±4.14 at dilution ratio 3:7, respectively (p<0.05). At 55 keV, the highest HU and the lowest HU were 2982.74±4.44 at dilution ratio 8:2 and 2827.35±4.06 at dilution ratio 3:7, respectively (p<0.05). At 60 keV, the highest HU and the lowest HU were 2959.46±4.48 at dilution ratio 8:2 and 2776.00±3.92 at dilution ratio 3:7, respectively (p<0.05). At 65 keV, the highest HU and the lowest HU were 2935.16±4.42 at dilution ratio 8:2 and 2704.61±8.32 at dilution ratio 3:7, respectively (p<0.05). At 70 keV, the highest HU and the lowest HU were 2910.03±4.46 at dilution ratio 8:2 and 2361.67±29.31 at dilution ratio 3:7, respectively (p<0.05). At 75 keV, the highest HU and the lowest HU were 2853.44±11.50 at dilution ratio 8:2 and 2002.68±24.06 at dilution ratio 3:7, respectively (p<0.05), and at 80 keV, the highest HU and the lowest HU were 2521.57±18.23 at dilution ratio 8:2 and 1716.06±19.88 at dilution ratio 3:7, respectively (p<0.05).
2. Comparison of HU according to MonoE per dilution ratio
At dilution ratio 8:2, the highest HU and the lowest HU were 3040.62±4.86 at 40 keV and 2521.57±18.23 at 80 keV, respectively (p<0.05). At dilution ratio 7:3, the highest HU and the lowest HU were 3036.20±3.40 at 40 keV and 2443.00±6.96 at 80 keV, respectively (p<0.05). At dilution ratio 6:4, the highest HU and the lowest HU were 3027.82±2.39 at 40 keV and 2386.10±7.09 at 80 keV, respectively (p<0.05). At dilution ratio 5:5, the highest HU and the lowest HU were 3005.13±4.67 at 40 keV and 2804.25±5.78 at 80 keV, respectively (p<0.05). At dilution ratio 4:6, the highest HU and the lowest HU were 2980.56±4.19 at 40 keV and 2216.68±2.99 at 80 keV, respectively (p<0.05). At dilution ratio 3:7, the highest HU and the lowest HU were 2948.91±4.37 at 40 keV and 1716.06±19.88 at 80 keV, respectively (p<0.05).
3. Correlation analysis for HU per MonoE and per dilution ratio
The multiple regression analysis found a negative correlation of -15.014 ± 0.298 (R2=0.519) for the HU per keV and a negative correlation of -61.372 ± 3.608 (R2=0.152) for the HU per dilution ratio[Table 3].
Table 3. Multiple linear regression analysis of HU values according to keV and dilution rate changes.
*. SE-Standard Errors
Ⅳ. DISCUSSION
In spectral CT, the images are reconstructed in various MonoE (keV), so that the HU varies according to the changes in keV and dilution ratio of the contrast agent. However, there is a general lack of studies analyzing the correlation of the HU with keV or dilution ratio of the contrast agent. The present study thus aimed to use a phantom to analyze the correlation between the HU and keV as well as dilution ratio of the contrast agent.
For the analysis, the images were reconstructed on nine levels of MonoE (40 - 80 keV), while the contrast agent was varied in dilution ratio on six levels from 8:2 to 3:7, and as a result, a total of 1,620 data were obtained. The results showed that the HU had a negative correlation of -15.014±0.298 with the changes in keV and a negative correlation of -61.372±3.608 with the changes in dilution ratio of the contrast agent.
In Ha et al.[12], where abdominal CT was performed, the image using 100 kVp tube voltage was reported to have shown higher HU values for the abdominal organs than the image using 120 kVp tube voltage. In Wintermark et al.[13], the use of the contrast agent in CT was reported to have led to higher HU values at 80 kVp than at 120 kVp.
The results in the studies by Ha et al. and Wintermark et al. collectively indicated that an increase in kVp decreased the HU values. In this study, the finding that an increase in keV led to a fall in the HU may be noteworthy, but since keV is not equivalent to kVp, the study cannot be said to coincide with the previous studies. Nevertheless, as in kVp, the negative correlation between keV and the HU may be significant.
In Yamaychi et al.[14], where the HU values were compared after the reconstruction based on 40 keV and 70 keV in Dual-Energy CT, the reconstructed image at 40 keV was reported to have shown higher HU values than the image at 70 keV. In Rassouli et al.[15], where Spectral CT was used for the reconstruction at 40 keV, 60 keV, 80 keV, and 100 keV, after which the HU values were compared, an increase in keV was reported to have led to a fall in the HU. The two studies agreed with the present study in that the HU values decreased with an increase in MonoE (keV) in Dual Energy CT and in Spectral CT. Nonetheless, the two studies differed from the present study in that the relation with the contrast agent had not been analyzed, and that a smaller range of MonoE was applied; 40 keV and 70 keV in the study by Yamaychi et al. and 40 keV, 60 keV, 80 keV, and 100 keV in the study by Rassouli et al.
Using the contrast agent after dilution decreases the concentration, and reduced concentration causes the viscosity of the contrast agent to fall. Seeliger et al.[16] reported that the viscosity of the contrast agent was an important factor related to Contrast Induced Nephropathy (CIN), one of the side effects of the contrast agent. Bak et al.[17] reported that the use of a low concentration of contrast agent could reduce the exposure dose for patients in the angiography using the Automatic Exposure Control (AEC). The use of the contrast agent after dilution can thus prevent such side effects as CIN and reduce the exposure dose. Hence, it is preferable to use the contrast agent after dilution as it leads to a safer CT test. However, as it is difficult to lower the concentration of the contrast agent to a level that degrades the imaging quality, analyzing the correlation between the Hu and dilution ratio of the contrast agent is crucial.
Kim[18] analyzed the HU values for the peripheral arterial image of MDCT between the original solution of the contrast agent and the contrast agent of 7:3 dilution ratio, to report that the HU showed no significant difference between the original solution and the diluted solution of the contrast agent. Lee et al.[19], in the MDCT test of the carotid artery, also reported no significant difference in the HU between the original solution of the contrast agent and the contrast agent of 9:1 and 8:2 dilution ratios. However, the two studies focused on the correlation between the HU and the contrast agent in MDCT, while there is still a general lack of studies analyzing the correlation between the HU and the contrast agent in spectral CT. It is thus anticipated that the analysis of the HU according to dilution ratio of the contrast agent per MonoE in spectral CT in this study would provide highly significant data.
The limitations in this study are that the phantom size could not be varied and that only 350 mg/ml iodine content in the contrast agent was used so that the analysis could not be performed according to the changes in iodine content. The unified testing condition of 120 kVp and 105 mAs was another limitation. Nevertheless, the significance of this study lies in the quantitative analysis of the correlation between the HU and keV and between the HU and dilution ratio of the contrast agent, through the use of spectral CT.
Ⅴ. CONCLUSION
In the spectral CT analysis, the HU and keV showed a negative correlation, where an increase in keV led to a fall in the HU, while the HU and dilution ratio of the contrast agent also showed a negative correlation, where an increase in dilution ratio of the contrast agent led to a fall in the HU. To conclude, the findings in this study are anticipated to serve as the basic data in studies analyzing the HU using spectral CT, according to the changes in the phantom size, the changes in the content of the contrast agent, and the changes in the testing conditions.
ACKNOWLEDGEMENTS
This research was supported by 2020 eulji university, University Innovation Support Project grant funded
References
- J. T. Hathcock, R. L. Stickle, "Principles and concepts of computed tomography", Veterinary Clinics of North America: Small Animal Practice, Vol. 23, No. 2, pp. 399-415, 1993. http://dx.doi.org/10.1016/S0195-5616(93)50034-7
- M. Prokop, "General principles of MDCT", European Journal of Radiology, Vol. 45, pp. 4-10, 2003. http://dx.doi.org/10.1016/S0720-048X(02)00358-3
- F. A. Jr Mettler, P. W. Wiest, J. A. Locken, C. A. Kelsey, "CT scanning: patterns of use and dose", Journal of Radiological Protection, Vol. 20, No. 4, pp. 353-359, 2000. http://dx.doi.org/10.1088/0952-4746/20/4/301
- NRCP, "Report no. 160: ionizing radiation exposure of the population of the United States", 2009.
- C. H. McCollough, S. Leng, L. Yu, J. G. Fletcher, "Dual- and Multi-Energy CT: Principles, Technical Approaches, and Clinical Applications", Radiology, Vol. 276, No. 3, pp. 637-653, 2015. http://dx.doi.org/10.1148/radiol.2015142631
- A. C. Silva, B. G. Morse, A. K. Hara, R. G. Paden, N. Hongo, W. Pavilicek, "Dual-Energy (Spectral) CT: Applications in Abdominal Imaging", Radiographics, Vol. 31, No. 4, pp. 1031-1046, 2011. http://dx.doi.org/10.1148/rg.314105159
- L. Yu, S. Leng, C. H. McCollough, "Dual-energy CT -based monochromatic imaging", American Journal of Roentgenology, Vol. 199, No. 5, pp. 9-15, 2012. http://dx.doi.org/10.2214/AJR.12.9121
- K. R. Thomson, D. K. Varma, "Safe use of radiographic contrast media", Australian Prescriber, Vol. 33, No. 1, pp. 19-21, 2010. http://dx.doi.org/10.18773/austprescr.2010.006
- D. Fleischmann, G. D. Rubin, A. A. Bankier, K. Hittmair, "Improved uniformity of aortic enhancement with customized contrast medium injection protocols at CT angiography", Radiology, Vol. 214, No, 2, pp. 363-371, 2000. http://dx.doi.org/10.1148/radiology.214.2.r00fe18363
- A. F. Kopp, "Angio-CT: heart and coronary arteries", European Journal of Radiology, Vol. 45, pp. 32-36, 2003. http://dx.doi.org/10.1016/S0720-048X(02)00360-1
- A. Laghi, R. Iannaccone, P. Rossi, I. Carbone, R. Ferrari, F. Mangiapane, I. Nofroni, R. Passariello, "Hepatocellular carcinoma: detection with triple-phase multi-detector row helical CT in patients with chronic hepatitis", Radiology, Vol. 226, No. 2, pp. 543-549, 2003. http://dx.doi.org/10.1148/radiol.2262012043
- H. I. Ha, S. S. Hong, M. J. Kim, K. Lee, "100 kVp Low-Tube Voltage Abdominal CT in Adults: Radiation Dose Reduction and Image Quality Comparison of 120 kVp Abdominal CT", Journal of the Korean Society of Radiology, Vol. 75, No. 4, pp. 285-295, 2016. http://dx.doi.org/10.3348/jksr.2016.75.4.285
- M. Wintermark, P. Maeder, F. R. Verdun, "Using 80 kVp versus 120 kVp in perfusion CT measurement of regional cerebral blood flow", American Journal of Neuroradiology, Vol. 21, No. 10, pp. 1881-1884, 2000.
- H. Yamauchi, M. Buehler, M. M. Goodsitt, N. Keshavarzi, A. Srinivasan, "Dual-Energy CT-Based Differentiation of Benign Posttreatment Changes From Primary or Recurrent Malignancy of the Head and Neck: Comparison of Spectral Hounsfield Units at 40 and 70 keV and Iodine Concentration", American Journal of Roentgenology, Vol. 206, No. 3, pp. 580-587, 2016. http://dx.doi.org/10.2214/AJR.15.14896
- N. Rassouli, M. Etesami, A. Dhanantwari, P. Rajiah, "Detector-based spectral CT with a novel dual-layer technology: principles and applications", Insights into Imaging, Vol. 8, No. 6, pp. 589-598, 2017. http://dx.doi.org/10.1007/s13244-017-0571-4
- E. Seeliger, B. Flemming, T. Wronski, M. Ladwig, K. Arakelyan, M. Godes, M. Möckel, P. B. Persson. "Viscosity of Contrast Media Perturbs Renal Hemodynamics", Journal of the American Society of Nephrology, Vol. 18, No. 11, pp. 2912-2920, 2007. http://dx.doi.org/10.1681/ASN.2006111216
- H. Bak, J. S. Jeon, Y. W. Kim, S. J. Jang "Dose assessment according to Differences in the Content of Iodine in Contrast Media used in Interventional Procedure", The Journal of the Korea Contents Association, Vol. 14, No. 3, pp. 337-345, 2014. http://dx.doi.org/10.5392/JKCA.2014.14.03.337
- S. H. Kim, "A Convergence Study on effectiveness of contrast agent reduction by normal saline solution dilution in the computed tomography of arteries of lower limb", Journal of Digital Convergence, 2015, Vol. 13, No. 9, pp. 431-437, 2015. http://dx.doi.org/10.14400/JDC.2015.13.9.431
- S. J. Lee, et al. "A Study on the usefulness of Dilution of Contrast Media in CT Examinations", Journal of Korean Society of Computed Tomographic Technology, Vol. 13, No. 2, pp. 161-167, 2011.