Introduction
Lung cancer is one of the most common cancers and has the highest age-standardized rate of all cancers [
1]. Lung cancer affects 22.5 patients per 100 000 people, and it remains a leading cause of death among cancer patients [
2]. Lung cancer is generally classified as small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). NSCLC accounts for 80–85% of lung cancers [
3].
Clinical treatment strategies for lung cancer differ based on histological type, and the choice of treatment strategy has a direct impact on outcome. Two parameters commonly used in clinical practice are size and density. Selection of appropriate size and density measurements is clinically important for follow-up observation and qualitative and quantitative diagnosis of nodules [
4]. Computed tomography (CT) imaging is a standard-of-care imaging modality used in the cancer treatment process that plays a key role in lung cancer screening and treatment response assessment [
5]. Clinical response criteria based on CT images, such as the Response Evaluation Criteria in Solid Tumors (RECIST) and modified versions, have been developed and widely used [
6,
7]. These criteria use changes in tumor size over time to monitor the tumor response. Tumor size measurement in CT images is objectively influenced by image quality. The quality of CT images results from a combination of various factors in the scanning protocol, including tube voltage, tube current, slice thickness, field of view and reconstruction methods, among which tube voltage and tube current are the most important.
The determination of optimal scanning protocols has long been investigated in order to reduce radiation dose and improve tumor recognition and discrimination. New technologies such as low-dose CT and ultra-dose CT have opened up more possibilities for lung screening [
8]. Christe et al. evaluated the optimal dose level in screening chest CT and concluded that 100 kVp and 25 mAs can provide satisfactory detection of solid nodules and ground glass nodules in lung cancer [
9]. Du et al. tested the screening capability for small nodules with phantom scanning and found that low-dose CT results agreed with results obtained with conventional standard chest CT [
10]. Jin et al. compared lung nodule detection results under high-definition and standard-definition CT and claimed that no significant differences in image quality were noted between the two scanning protocols [
11]. The studies mentioned above aim at preferred image quality for lung cancer screening but did not quantitatively calculate the influence on tumor size.
Several studies [
12‐
14] have demonstrated that tumor density, which can be quantified by the HU value, provides extra assessment information as well. Criteria for lung nodule assessment define response patterns by changes in tumor size without specifying scanning parameters. Despite the recommendation to use standard chest CT, studies of lung nodule assessment have used a variety of scanning parameters [
15]. Strauch et al. performed a meta-analysis of tumor response to cancer in dynamically enhanced CT and summarized the scanning protocols applied in the study: kVp ranged from 80 to 120 and mAs ranged from 36 to 200. Different scanning parameters result in different image quality and potentially different conclusions for response assessment [
15].
The aim of this study is to evaluate the impact of CT scanning parameters (tube voltage and tube current) on tumor size and density measurement and investigate the optimal energy and mA image quality for lung nodule size and density based on a phantom study.
Discussion
GGNs are vague and increased shadows of lung nodules that do not cover bronchial and pulmonary vascular structures, which are often the imaging manifestations of early lung adenocarcinoma, while solid nodules are lung nodules with high density that cannot be seen through the nodules in the lung texture. With the wide clinical application of CT, the detection rate of lung nodules has been increasing, and its diagnosis and treatment have received more and more attention. Currently, the management of lung cancer screening results and the diagnostic and therapeutic evaluation of lung cancer staging are mainly based on nodule size and type.
The results of this study showed that the maximum diameter of solid nodules measured at 80 kVp and 140 mA was closer to the true size, and the maximum diameter of ground-glass nodules measured at 100 kVp and 100 mA was closer to the true size, but these differences were not statistically significant. However, for nodal CT values (HU), CT values of GGNs and solid nodules were closest to ground truth when measured at 80 kVp and 100 kVp, respectively. The focus of the study was to evaluate the effect of different tube voltage and tube current conditions on nodule size and density, and to find out the energy and mA values for obtaining the best image quality for different types of lung nodules.
The accuracy of nodule measurements is important in interpreting the possibility of whether it is a tumor, and commonly used tumor response criteria, such as RECIST 1.1 [
6] and iRECIST [
3], use the maximum diameter as an indicator of tumor size to monitor changes over time. On the other hand, changes in density of tumor areas before and after treatment were detected in CT images in the form of changes in HU values [
16], which could enhance the injection of density changes as additional functional information into the tumor assessment criteria to further improve the assessment accuracy. This phantom study revealed that the combination of scans required for more accurate assessment of lung nodules by evaluating both aspects. It provided more accurate diagnostic information to improve the clinical management of lung nodules.
In our study, the combination of 80 kVp and 140 mA scan was preferred for solid nodule scans and 100 kVp and 100 mA scan was preferred for GGNs. The difference can be derived from each component during image acquisition. In addition to tube current and tube voltage differences, variations in field of view and slice thickness can affect image quality and subsequent measurements. Differences in scanners can have a significant impact, as vendors are equipped with different technologies based on mechanical and reconstruction methods to obtain good image quality. In clinical applications, it is important to consider how emerging reconstruction methods (e.g., ASiR) compare with classical filtered projection back (FPB) in terms of image quality [
22,
23], but the latest deep learning image reconstruction (DLIR) techniques in CT will gradually be applied in clinical practice, providing more choices of “optimal” scanning parameters. Jiang et al. demonstrated that DLIR reduced image noise, improved nodule detection and measurement accuracy on ultra-low-dose chest CT images compared to adaptive statistical iterative reconstruction-V [
24].
The present study has the following limitations. First, in this study, body models were used for experimental purposes. So the conclusion needs to be verified by further clinical applications. The body model used in this study was based on 70 kg adult males. Therefore, further studies are needed to determine whether this body model is suitable for other body sizes and body types. Second, in this study, only one CT scanner was used to acquire images. Therefore, further validation is needed to determine the feasibility of other types of CT scanners as well as other computer-aided design software. Third, in this study, the diameters of the simulated pulmonary nodules were 8, 10, and 12 mm. Although these diameters simulated CT Hounsfield unit values of − 100 HU and − 800 HU (tube voltage: 120 kVp), these diameters do not fully simulated lung nodules encountered in clinical work, considering the significant differences in size, shape, CT attenuation values, and other aspects of the lung nodules [
25,
26]. Therefore, further in-depth studies are needed to validate the findings of this study. Finally, only three energies were included in the study. In the standard scanning procedure, the images contained both photoelectric scattering and Compton scattering. More energies with separation between these two phenomena will improve the image quality and are therefore desired for optimizing the images used for tumor response assessment. The dual energy technique is suitable for this situation and we will follow this direction.
Conclusion
A LUNGMAN N1 body model multifunctional anthropomorphic chest model with two types of artificial lung nodules (diameters: 8, 10, and 12 mm; CT values: − 100, − 630, and − 800 HU) was used in the present study and demonstrated that a combination of 80 kVp and 140 mA scans was preferred for measuring the size of the solid nodules, and a combination of 100 kVp and 100 mA scans was preferred for measuring the size of the GGNs. However, when measuring the CT values of GGNs and solid nodules, 80 kVp and 100 kVp were preferred, respectively, and the CT values were closest to the true CT values of the nodules. Therefore, the combination of scanning parameters should be selected for different types of nodules to obtain more accurate nodal data.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.