|Year : 2021 | Volume
| Issue : 5 | Page : 1157-1164
Effects of CT images at different reconstruction energies on radiotherapy planning of patients diagnosed with nonsmall cell lung cancer
Tian Xiufang1, Liu Kun1, Wang Jing2, Li Cuihua1, Zhang Jiandong1, Yong Hou1, 3
1 Department of Radiotherapy, The First Affiliated Hospital of Shandong First Medical University, Jinan, Shandong, China
2 Department of Oncology, The First People's Hospital of Xiangyang City, Hubei Medical College, Xiangyang, Hubei, China
|Date of Submission||28-Jun-2021|
|Date of Acceptance||06-Sep-2021|
|Date of Web Publication||27-Nov-2021|
Department of Radiotherapy, The First Affiliated Hospital of Shandong First Medical University, Jinan, Shandong, 250014
Source of Support: None, Conflict of Interest: None
Objective and Aims: We conducted this study to explore the influence of spectral computer tomography (CT) images at different reconstruction energies on the radiotherapy plan of patients with nonsmall-cell lung cancer (NSCLC).
Subject and Methods: Here, 38 NSCLC patients were selected to undergo energy spectral scanning. All energy spectral images obtained were then transferred to the Discover™ CT postprocessing workstation to generate 40k eV, 60 keV, 80keV, 100keV, 120keV, and different 140keV single-energy images. Subsequently, the images were imported to the Eclipse planning system, after which an oncologist contoured the target area and organs at risk (OARs) on these single-energy images described above. Furthermore, a physicist then designed radiotherapy plans to conduct statistical analysis on the tissue CT value and target volume of each single-energy image, to compare the dosimetry of different plans about the OARs and the target area.
Results: The CT values of gross tumor volumes (GTV), heart, lung, and spinal cord samples subjected to different energy CT images were statistically different (P < 0.05). Among them, the CT value of each tissue obtained in the 40 keV group was the largest and decreased with the increase in energy. As shown, no statistically significant differences were observed in the homogeneity index and conformity index, including the maximum, minimum, and average doses of GTV delineated on the CT images of different energies (P > 0.05), as well as the OARs.
Conclusions: When CT images of different energies obtained from the energy spectral CT scans were used in the design of radiotherapy planning, no significant differences were observed in the target area outlines and in the doses caused by energy factors. However, the differences in tissue CT values had statistical significance.
Keywords: Dosimetry, energy spectrum computed tomography, nonsmall cell lung cancer, radiotherapy
|How to cite this article:|
Xiufang T, Kun L, Jing W, Cuihua L, Jiandong Z, Hou Y. Effects of CT images at different reconstruction energies on radiotherapy planning of patients diagnosed with nonsmall cell lung cancer. J Can Res Ther 2021;17:1157-64
|How to cite this URL:|
Xiufang T, Kun L, Jing W, Cuihua L, Jiandong Z, Hou Y. Effects of CT images at different reconstruction energies on radiotherapy planning of patients diagnosed with nonsmall cell lung cancer. J Can Res Ther [serial online] 2021 [cited 2022 Jan 24];17:1157-64. Available from: https://www.cancerjournal.net/text.asp?2021/17/5/1157/331299
| > Introduction|| |
With the development of computed tomography (CT) technology, spectral CT has been widely used recently, which was included in security checks, calculi ingredient analysis, diagnosis of gout and bone marrow edema, early diagnosis of tumors, etc.,,,,,, Compared with traditional CT, spectral CT has various advantages, including the ability to generate energy decay curves, iodine-based maps, water-based maps, and virtual-plain scans. These abilities lay broad application prospects for qualitative diagnosis and quantitative analysis of lesions. At present, spectral CT is mainly used in the chest and abdomen examinations. The advantage of virtual-plain scans can not only reduce the patient's exposure dose and trim down errors caused by the inconsistency of measuring the area of interest before and after the scan, these scans also conduct a differential analysis on the nature of these lesions., Energy spectrum CT uses a single X-ray tube to realize dual-energy imaging through single-source instantaneous kV-switching technology. Based on these two energy data, the attenuation coefficient of the voxel in the energy range of 40–140 keV is determined, and 101 single-energy images are further obtained. This relatively pure single-energy image can therefore greatly reduce the impact of hardening artifacts and produce relatively purely CT value images, which make CT values more consistent and reliable regardless of the position, scan under the entire visual field, or patient. Furthermore, during image processing, different substances are decomposed into matrix pairs based on photoelectric effects, electron densities, effective atomic numbers, and matrix distribution densities. These factors are then calculated to provide more diagnostic information for the clinic, which expands the diagnostic applications of CT. The development of spectral CT technology also poses new challenges to radiotherapy simulation positioning technologies on the basis of how to apply these spectral CT technologies to improve the accuracy of CT value-tissue parameter conversions, thereby improving the accuracy of radiotherapy.
Before radiotherapy, the patient should perform a positioning scan on a CT simulator to obtain CT images containing the patient's anatomy and position information. Then, the images are imported into the radiotherapy-planning system. Subsequently, the treatment planning system obtains the conversion curve of the CT value obtained and the relative electron density of the tissue from the CT images. Furthermore, the inhomogeneous tissue would then be calibrated based on the relative electron density of the tissue, after which dose calculations are conducted.,, To obtain clear CT images, the patient's scanning conditions will be different during simulated positioning scans, such as their scan voltages, scan currents, and scan layer thickness. These will then affect the CT value obtained and the relative electron density conversion curves in addition to the patient's dose calculation of the radiotherapy plan. Therefore, this study used single-energy images obtained by spectral CT to study the effects of CT images of different energies on the radiotherapy plan of patients with nonsmall-cell lung cancer (NSCLC). We propose that this study will guide the setting of scanning conditions during radiotherapy simulation positioning This study will also reduce the influencing factors of dose calculations when devising a radiotherapy plan, which will improve the accuracy of radiotherapy.
| > Subject and Methods|| |
Here, 38 patients diagnosed with NSCLC as confirmed by pathological biopsy were selected, including 23 cases with lesions on the left and 15 cases with lesions on the right lung. The male-to-female ratio was 21:17. Furthermore, the age range of selected patients was 55–75 years, with an average age of 61.2 years.
Equipment and software
All inspections were conducted using the gemstone (Discover™ CT 750HD, GE Healthcare, USA) energy spectrum-scanning mode. The planning system followed was that of the Eclipse 10 radiotherapy-planning system.
Positioning and scanning
The patients took a supine position with their heads advanced. The scanning range was from the thorax entrance to the diaphragm level. The energy spectrum scan of the entire lung was also taken. Scanning conditions were as follows: Tube voltage was instantaneously switched from 40 kVp to 140 kVp, tube current (550 mA), rotation time (0.8 S), layer thickness (5 mm), while the spacing was 5 mm. Furthermore, the reconstruction interval was set at 1.25 mm, whereas, the pitch was 1.375 mm.
Postprocessing of computed tomography images
All energy spectrum imaging data were transferred to the postprocessing workstation (AW4.5) and reconstructed to generate single-energy images of 40keV, 60 keV, 80 keV, 100 keV, 120 keV, and 140 keV, respectively. Then, we transferred the single-energy images to the Eclipse 10 treatment planning system.
Contouring of target areas and organs at risk
According to the ICRU83 report and the RTOG1106 consensus, senior clinical physicians contoured the target areas and organs at risk (OARs) on single-energy CT images as described above. Gross tumor volumes (GTV) included the primary tumor and metastatic lymph nodes in the mediastinum and supraclavicular regions as determined by the images. However, the clinical target volume (CTV) was 3 mm at the upper and lower ends of GTV, and 5 mm at the axial extension. Subsequently, we generated PTV values by expanding at 8 mm in all directions from the CTV. The OARs, such as the lungs, spinal cord, and heart were then contoured, respectively.
Plans were generated using the Eclipse10 (Varian Inc.,) treatment planning system. All treatment plans were delivered using a 6-MV photon beam. Subsequently, 7-Field intensity-modulated radiation therapy plans were completed by senior physicists, and physicians with deputy directors or more senior officials reviewed and evaluated the uniformity of the target dose and organ-endangered doses using dosimetry-related indicators. The PTV dose selected for all patients was 60 Gy/30 fractions, and the prescribed dose was proposed to cover 95% of the target volume. The chosen maximum dose of the planning target volume was 120%. Furthermore, the plans of all different energy CT images of the same patient used the same field energy, field direction, optimization parameters, and dose calculation algorithm (AAA). Therefore, we first planned to meet prescription requirements and OAR limits on a single-energy image. We also used this plan as a template to design plans for other energy groups. For the OARs, maximal dose values for the spinal cord were limited to 45 Gy. However, V40 <30%, V30 <40%, and Dmean <25 Gy values were selected for the heart. V5 <60%, V20 <30%, and V30 <20% values were also selected for both lungs. The planning objectives for the OARs were uniform for all patients and were established following the strictest constraint recommendations to assure that the lowest possible dose was delivered to normal tissues.
The target HI area as referred to in the ICRU83 report. It was defined as follows:
HI = (D2% − D98%)/D50% (1)
In which, D2% represents the irradiation dose received by 2% of the target tumor volume, D98% represents the irradiation dose received by 98% of the target tumor volume, and D50% represents the irradiation dose received by 50% of the target tumor volume.
However, the CI was:
CI = (Vt, ref/Vt) × (Vt, ref/Vref) (2)
Where Vt is the volume of the target area, Vt, ref is the volume of the target area surrounded by the reference isodose line surface, and Vref is the volume of all areas surrounded by the reference isodose lines. The closer the CI value was to 1, the better the conformal coverage was.
However, comparing the CT value of tissues and GTV volumes with different energy CT images, we also evaluated HI, CI, maximum dose (Dmax), minimum dose (Dmin), and average dose (Dmean) of the target areas. For the OARs, the maximal dose of the spinal cord was 45 Gy. Furthermore, we evaluated the V40, V30, and Dmean for the heart, and V5, V20, and V30 for both lungs, respectively.
The data were divided into six groups based on different energies, (40 keV, 60 keV, 80 keV, 100 keV, 120 keV, and 140 keV). SPSS 20.0 (IBM, Stanford university, American) was then used to analyze the measurement data. All measurement data were first tested for normality. When the measurement data obeyed the normal distribution, and the variance was homogeneous, one-way analysis of variance (ANOVA) was subsequently conducted. Otherwise, when the variance was not uniform, Brown–Forsythe and Welch tests were used. P < 0.05 was considered statistically significant.
| > Results|| |
On the basis of the images obtained, the study subjects were divided into six groups of different energies from the 38 patients diagnosed with nonsmall-cell lung carcinoma. For the 228 plans obtained, CT values of tissues, including their target volumes in addition to their target and organ-endangered doses were included as comparative indicators for the analysis. All measurement data conformed to the normal distribution. The results were then expressed as the mean plus or minus standard deviation (x̄ ± s). Two significant digits after the decimal point and three significant digits after the decimal point was used to report the mean of HI and CI, respectively.
Comparison of the computed tomography values of different energy obtained from the computed tomography images
After transmitting the CT images to the planning system, the physician contoured their GTV, CTV, PTV, and OAR (for the left lung, right lung, heart, and spinal cord). Then, five points were randomly selected within each contoured structure. The planning system then read the CT value of the corresponding points. Subsequently, the arithmetic average of the CT values of these five points was used as the CT value of the tissue. In this study, the CT values for GTV, heart, lung, and spinal cord were then selected for univariate ANOVA. Results are shown in [Table 1]:
|Table 1: Computer tomography values obtained from the different energy groups|
Click here to view
[As can be seen from the table above, the P values for the CT images of the four tissues obtained were all <0.05, and the differences in CT values of similar patients at different energy levels were statistically significant. [Figure 1]] shows that the CT values of the four tissues decreased with an increase in energy. Furthermore, as shown, the range of CT values for the heart and lung varied widely, whereas, that of the spinal cord was small.
|Figure 1: Graphs of computed tomography values for different tissues under different energies|
Click here to view
Comparison of differences in gross tumor volumes contouring
On the planning system, the GTV volumes contoured by the physician were obtained. Results are shown in [Table 2]. As shown, the P value was >0.05. Hence, the difference was not statistically significant.
|Table 2: The volume of gross tumor volumes obtained from different energies|
Click here to view
Six pairs of data as shown above were compared pairwisely. Among them, the 40keV energy group had the largest GTV volume, reaching 105.38 ± 140.98 cm3, and the 120keV energy group had the lowest average volume. Subsequently, (Vmax-Vmin)/Vmax × 100% was used to indicate the difference between the maximum and minimum values. The value for GTV obtained was 4.94%. Furthermore, the smallest difference obtained for the GTV volume was between 60 keV and 80 keV, which was only 0.14%.
The CT images of different energies having different display details are as shown in [Figure 2].
|Figure 2: Computed tomography images of different energies at the same level (W/L: 400/40)|
Click here to view
Note: The energy levels represented by serial numbers 1–6 are 40 keV, 60 keV, 80 keV, 100 keV, 120 keV, and 140 keV, respectively.
Dosimetric comparison of planning target areas
All plans took PTV as their target areas. PTV information, including Dmax, Dmean, Dmin, D2%, D98%, D50%, VPTV100%, VPTV, and Vt of all 228 plans was obtained to get their CI and HI values. Subsequently, one-way ANOVA was conducted. Data are shown in [Table 3].
From [Table 3], we can see that the maximum HI value was 0.166, whereas, the minimum value was 0.131. When compared between groups, P > 0.05. Hence, the difference was not statistically significant. Furthermore, the maximum value of CI was 0.721 when the minimum value was 0.686. Considering the comparison between groups, P > 0.05, the difference was also not statistically significant. In summary, when planned using different energies, the uniformity and conformity of the target area were not affected by energy.
When compared between groups, it can be shown from the table above that the P value was >0.05, and the maximum group of Dmax, Dmean, and Dmin were all 120 keV (6929.90 ± 159.26, 6367.44 ± 80.49, and 5142.29 ± 391.38, respectively). However, both Dmax and Dmean minimum values appeared in the 40 keV group (6775.31 ± 100.36 and 6772.15 ± 45.29), which all met the requirements of the plan. Furthermore, the maximum and minimum values of Dmax were different by 4.49%, Dmean by 1.49%, and Dmin by 3.58%, all within 5%.
Dosimetric comparison of the lung, heart, and spinal cord tissues
Lung dose comparison
The lungs were divided into the ipsilateral and healthy lungs for dose comparison. For both lungs, V5, V20, V30, and Dmean were evaluated, respectively. The results are shown in [Table 4] and [Table 5].
Comparing the doses of the six groups from the ipsilateral lung, P > 0.05, which means that the difference was not statistically significant. Among them, the doses of V5, V20, V30, and Dmean on the affected side of the 60 keV group were the largest. However, the differences from the minimum values were 6.89%, 8.26%, 11.57%, and 5.87%, respectively. The minimum doses of V20, V30, and Dmean all appeared in the 80 keV group, whereas, that of the minimum doses of V5 was 40 keV.
As shown in [Table 5], similar to the ipsilateral lung, the dose data of the six groups from the healthy lung were compared, and the result were P > 0.05. Hence, the difference was not statistically significant. Among them, the ipsilateral lungs in the 40 keV group had the largest V20 and V30, whereas, V5 and Dmean were the smallest. In the 60 keV group, the ipsilateral lung had the largest V5 and Dmean. However, the minimum values of V20 and V30 appeared in the 140 keV and 100 keV groups, respectively. Furthermore, the differences in the minimum values were 6.88%, 7.00%, 23.53%, and 3.97%. The dose curve of the results obtained from the ipsilateral and healthy lungs is shown in [Figure 3]a, [Figure 3]b, [Figure 3]c, [Figure 3]d.
|Figure 3: Doses of organs at risk. (a) V5 of the lungs, (b) V20 of the lungs, (c) V30 of the lungs, (d) Mean dose of the lungs, (e) V30 and V40of the heart dose, (f) Mean dose of the heart and maximum dose of the spinal cord|
Click here to view
Comparison of cardiac and spinal doses
We also compared the V30, V40, and Dmean doses of the heart and those of the maximum dose (Dmax) of the spinal cord, respectively. Data are shown in [Table 6]. It can be seen that for the spinal dose of the six groups, P > 0.05, and for the V30, V40, and Dmean of the heart dose, P > 0.05. This result showed that no statistically significant difference was observed between the maximum dose of the spinal cord and that of the heart.
From [Table 6], we can see that the maximum cardiac dose was in the 40keV group, and the differences from the minimum were 33.46%, 42.92%, and 5.66%, respectively. However, the minimum value of V30 was in the 60 keV group, and that of the minimum value of V40 and Dmean was in the 80 keV group. Furthermore, the maximum value of the spinal cord dose was 3941.00 ± 823.45, which appeared in the 100keV group and that of the minimum value was 3739.28 ± 656.60, and it appeared in the 60 keV group. Furthermore, the difference between the maximum and minimum values was 5.12%. Cardiac and spinal cord dose curves are shown in [Figure 3]e and [Figure 3]f.
In summary, slight differences existed in single-energy images of the different energy results obtained from spectral CT scans when they were used in the design of radiotherapy plans. As reported above, although these differences observed in the CT values of images were obtained from different energies, no significant impact was noticed on physicians' target delineation. In terms of dose, when the radiotherapy plan was conducted on CT images of different energies, the difference between HI and CI values of the target area was not statistically significant. The differences between the various dose indicators were not significant as well. For the comparison between groups, the target area dose difference was the smallest (maximum 4.49%). Nevertheless, the difference between the lung, heart, and spinal doses was within 3.97%–42.92%, which was such a difference is large. However, these differences were still statistically insignificant.
| > Discussion|| |
The interactions between X-rays and substances mainly involve photoelectric, Compton, and electron pair effects. The commonly used diagnostic X-ray energy used in clinics is between 30 and 200keV. In this energy band, the photoelectric effect is dominant, and the mode of interaction depends on the characteristics of the substance itself and the X-ray energy. Therefore, the lower the energy, the stronger the ability to distinguish substances. Based on this characteristic, the traditional CT imaging method is attenuation imaging. However, the current CT image reconstruction algorithms include; filtered back-projection algorithm, iterative reconstruction algorithm, and direct Fourier back-projection algorithm. Dual-energy CT (DECT) reconstruction methods are also involved, which mainly include preprocessing, postprocessing, and iterative method. The image reconstruction method is an influencing factor of image error, which makes this study important.
Jie et al. studied the diagnosis of chest diseases using dual-source CT virtual-plain scan technology. Their results showed that the image quality of the virtual-plain scan was not statistically different from that of the conventional plain scan. However, for the display of partial calcifications and small lymph nodes near the vena cava, the plain scan group was superior to that of the virtual-plain scan. Zhaoxian also studied the image quality of the head and neck dual-energy low-dose radiation and found that compared with conventional doses, the signal-to-noise ratio and image quality scores of the low-dose group images were not statistically significant. Likewise, Xintang et al. compared the enhanced images of lung cancer obtained by dual-energy scanning and conventional scanning. He found that the images obtained using the two scanning methods were not different from that obtained in the detection rate of the lesion as well as that obtained during the detailed display of the lobulation sign, burr sign, cavity, necrosis, and so on. The studies above also compared the differences between ordinary scanning and energy spectrum-scanning imaging techniques. Our research refined the difference in the target area delineation caused by different reconstruction energies of the energy spectrum scanning. The results showed that GTV volume differences obtained from the images of six groups of energy was statistically insignificant.
Mingying et al. believed that the lower the X-ray energy, the greater the absorption coefficient. The more X-ray attenuation, the better the CT value at 40 keV would reflect of the substance. The CT value of tumor tissues of lung adenocarcinoma and squamous cell carcinoma also decreased with an increase in energy. This trend was consistent with the results of this study. Furthermore, Zhaoxian believed that no difference existed in the CT values of blood vessels obtained using different doses of CT imaging. Likewise, Goodsitt et al. used phantom scanning and theoretical calculations to find that as the energy changed between 40 keV and 120keV, the CT root-mean-square error also varied between 6 HU and 248 HU, especially for the combination between the low energy and high atomic number. This difference was more evident. In this study, the maximum CT value difference was within the GTV of the 40 keV group, reaching 231HU. In addition, Zhenyu et al. showed that the scanning voltage, CT bed surface, and geometric position of the phantom were all influencing factors of CT value. In this study, the CT values of GTV, heart and spinal cord tissues decreased with increasing energy. The significant difference in CT values did not also have a significant impact on the volume of the GTV outlined by the physician. Previous studies reported that this result would be related to the recognition ability of the human eyes. Therefore, under the condition of specific window width and level, if the gray difference between adjacent areas was less than IMax/16, it was difficult for the human eye to distinguish. However, with the development of artificial intelligence, automatic contouring technology has received extensive attention. Thus, with contouring technology based on CT values, inevitably higher demands will be placed on the accuracy of the CT values of the images.
The dose calculation of the radiotherapy-planning system depends on the CT value-electron density curve. The CT value obtained from the phantom scan under specific scanning conditions and electronic density values provided by the manufacturer are input during data configuration. Liquan et al. used different CT-electron density conversion curves for their study. The difference in MU values obtained using different plans was all <20%, of which the intensity modulation plan was larger. Recently, scholars have conducted many studies to identify the relationship that exists between CT values and electron density curves.,,,, Therefore, Jiaojiao et al. proposed a spectral CT-electron-density image reconstruction technology based on multi-material decomposition algorithms. Numerical simulation experiment results also showed that the electron density image reconstructed by this algorithm can easily and intuitively distinguish between four substances; water, ethanol, glycerol, and plexiglass. However, that of the internal structure information of the image was more abundant. This result showed that DECT has great potential in both diagnosis and radiation therapy.
| > Conclusions|| |
In this study, under different reconstruction energies, the CT value of the same tissue was different, which in turn caused differences in radiotherapy doses. Since our study used the same CT value-electron density curves, the accuracy of the calculation used for each energy dose requires further improvement. In subsequent studies, future research in this area can be refined to provide a reference for selecting scanning conditions for precision radiotherapy.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Hounsfield GN. Computerized transverse axial scanning (tomography). 1. Description of system. Br J Radiol 1973;46:1016-22.
Qi W, Zhiqiang C, Xiaoping W, Xue-wu W, Li Z, Ke-jun K. Overview of X-ray safety inspection technology [J]. CT Theory and Application Research 2004;13: 32-7.
Xinxin H, Yun Z, Wei L, Chen L, Xuejun Z. Application of dual-energy CT in diagnosis and treatment of gout [J].Chinese Journal of Clinical Immunology and Allergy 2017;11:370-4.
Primak AN, Fletcher JG, Vrtiska TJ, Dzyubak OP, Lieske JC, Jackson ME, et al.
Noninvasive differentiation of uric acid versus non-uric acid kidney stones using dual-energy CT. Acad Radiol 2007;14:1441-7.
Foti G, Catania M, Caia S, Romano L, Beltramello A, Zorzi C, et al.
Identification of bone marrow edema of the ankle: Diagnostic accuracy of dual-energy CT in comparison with MRI. Radiol Med 2019;124:1028-36.
Xiaozhu L, Weixia L, Yanbo Z, Haipeng D, Peijie L, Fei M, et al
. Preliminary Application of Gem Energy Spectrum CT in Tumor Diagnosis [J]. Journal of Diagnostics Theory and Practice 2010;9: 155-60.
Shifeng T, Ailian L, Heqing W, Lihua C, Xijia D, Yijun L, et al
. Diagnostic value of single-source dual-energy CT iodine base map for early gastric cancer [J].Chinese Journal of Clinicians (Electronic Edition) 2016;10:3674-7.
Chae EJ, Song JW, Seo JB, Krauss B, Jang YM, Song KS. Clinical utility of dual-energy CT in the evaluation of solitary pulmonary nodules: Initial experience. Radiology 2008;249:671-81.
Chae EJ, Song JW, Krauss B, Song KS, Lee CW, Lee HJ, et al.
Dual-energy computed tomography characterization of solitary pulmonary nodules. J Thorac Imaging 2010;25:301-10.
Qingguo R, Yanqing H, Jianying L. The basic principle and clinical application of CT energy spectrum imaging. Int J Med Radiol 2011;34:559-63.
Johnson TR. Dual-energy CT: General principles. Am J Roentgenol 2012;199 Suppl 5:S3-8.
Kry SF, Bednarz B, Howell RM, Dauer L, Followill D, Klein E, et al
. AAPM TG158: measurement and calculation of doses outside the treated volume from external-beam radiation therapy[J]. Med Phys 2017;44:e391-e429.
Xiangkun D, Yunlai W, Linchun F, Wei Y. Research on MVCT images of spiral tomography accelerator used for head and neck dose calculation [J].Chinese Journal of Medical Devices 2014;38:141-4.
Dingyi Z, Dazhen J, Hong Q, Gang L, Hui L, Zhi-rong B, et al
. The influence of four-dimensional CT scanning parameters and scanning modes on CT values [J].Chinese Journal of Medical Physics 2016;33:919-23.
Grégoire V, Mackie TR. State of the art on dose prescription, reporting and recording in Intensity-Modulated Radiation Therapy (ICRU report No. 83)[J]. Cancer Radiother 2011;15:555-9.
Tsunoo T, Torikoshi M, Sasaki M, Endo, M, Yagi, N, Uesugi, K. Distribution of Electron Density Using Dual-Energy X-Ray CT[J]. IEEE Transactions on Nuclear Science 2003;50:1678-82.
Yimin H. Radiation Physics. Beijing: Atomic Energy Press; 1999.
Peng Z, Jia H, Yuxiang X. Theoretical error analysis of dual-energy CT imaging base material decomposition method [J]. CT Theory and Application Research 2011;2:3-12.
Jie J, Bo H, Jun Z, Dan H. Application value of dual-source CT dual-energy chest virtual flat scan [J]. Chinese Medical Imaging Technology 2012;4:185-9.
Zhaoxian Y. Research on dual-energy CTA low radiation dose imaging of head and neck [D]. Guangzhou University of Chinese Medicine 2015;1:1-20.
Xintang Z, Jin S, Dan H. The clinical value of dual-source CT dual-energy technology in the diagnosis of lung cancer. J Pract Radiol 2013;29:1245-8,57.
Mingying L, Guangli W, Dingye C, Chengqi Z. Preliminary study on classification of lung cancer by energy spectrum CT imaging [J]. Journal of Shandong University (Medical Edition) 2012;50:73-6.
Goodsitt MM, Christodoulou EG, Larson SC. Accuracies of the synthesized monochromatic CT numbers and effective atomic numbers obtained with a rapid kVp switching dual energy CT scanner. Med Phys 2011;38:2222-32.
Zhenyu Q, Shaomin H, Xiaowu D. Calibration detection of CT value of radiotherapy plan and analysis of its influencing factors [J]. Cancer 2006;01;110-4.
Shuqian L, Guohong Z. Medical Image Processing and Analysis. Beijing: Science Press; 2010.
Wenjian Q. Research on key issues of medical image segmentation based on machine learning and its application in tumor diagnosis and treatment [D]. University of Chinese Academy of Sciences (Shenzhen Advanced Technology Research Institute of Chinese Academy of Sciences) 2019;2:9-26.
Liquan G, Xiaozhe S, Zhihui L. Acquisition and correction of CT-electron density conversion curve and its effect on dose calculation of treatment plan. Med Health Equip 2016;37:103-5.
Jiaojiao Q. Research on the application of dual-energy CT (DECT) electron density image reconstruction technology [D]. Shandong University 2018;1:1-36.
Saito M. Potential of dual-energy subtraction for converting CT numbers to electron density based on a single linear relationship. Med Phys 2012;39:2021-30.
Hünemohr N, Krauss B, Tremmel C, Ackermann B, Jäkel O, Greilich S, et al.
Experimental verification of ion stopping power prediction from dual energy CT data in tissue surrogates. Phys Med Biol 2014;59:83-96.
Bourque AE, Carrier JF, Bouchard H. A stoichiometric calibration method for dual energy computed tomography. Phys Med Biol 2014;59:2059-88.
van Abbema JK, van Goethem MJ, Greuter MJ, van der Schaaf A, Brandenburg S, van der Graaf ER. Relative electron density determination using a physics based parameterization of photon interactions in medical DECT. Phys Med Biol 2015;60:3825-46.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]