Performance comparison of dual-layer detector CT parameters from different blood vessels in the detection of anemia

Introduction

Anemia is characterized by a reduction in red blood cell count or hemoglobin concentration, leading to diminished oxygen-carrying capacity and an inability to meet the body’s physiological oxygen requirements (1). Globally, anemia affects approximately 40% of preschool children and 30% of women (2). Patients with anemia often have higher hospitalization rates, longer stays, increased disability, and higher medical costs, and hemoglobin concentration significantly affects the quality of life and prognosis of patients (3-6). Anemia is typically diagnosed by evaluating hemoglobin concentration or hematocrit in peripheral blood samples. However, this procedure may not be applied to patients with primary asymptomatic anemia. In addition, in certain clinical situations (e.g., when the blood routine results of critically ill patients are delayed or unavailable), an additional diagnostic tool for anemia could be essential.

Computed tomography (CT), a non-invasive diagnostic technique, has been used to detect anemia, even without the use of contrast agents (7-9). The “aortic ring sign” and “interventricular septum sign” act as important CT indicators in the diagnosis of anemia (10-12). However, these CT signs are subjective and dependent on the reader’s experience (13). Additionally, differences in CT equipment and blood vessel location may lead to variations in the CT value, which in turn affect the diagnostic accuracy. Through material decomposition, dual-layer computed tomography (DLCT), a dual-energy CT method, can provide additional quantitative parameters beyond the CT value (14,15), and can thus characterize blood components and diagnose anemia better than conventional CT. Unlike other dual-energy CT methods, DLCT is always in “dual-energy mode” (16,17), which facilitates the retrospective collection of spectral analysis data. Recently, dual-energy CT research has used parameters, including the CT value and electron density, to detect anemia based on in vitro blood samples (18). However, to date, no research appears to have explored the diagnostic value of the DLCT parameters for anemia in vivo.

Thus, the present study sought investigate the diagnostic performance of DLCT parameters in detecting anemia in vivo. It also sought to determine which major blood vessels were the most suitable for the diagnosis of anemia when using DLCT. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-24-1671/rc).

Methods

Study population

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by Hunan Cancer Hospital Ethics Committee (No. 2025KYKS44) and the requirement of individual consent for this retrospective analysis was waived. In total, 240 patients who met specific inclusion and exclusion criteria were enrolled in the study between February and March 2024. To be eligible for inclusion in the study, the patients had to meet the following inclusion criteria: (I) be aged >18 years; (II) have undergone chest and abdominal unenhanced CT scan; and (III) have an interval between the CT scan and peripheral blood sampling of <24 hours. Patients were excluded from the study if they met any of the following exclusion criteria: (I) had poor quality CT images; (II) were pregnant; (III) had incomplete blood test data; and/or (IV) had undergone a procedure that could potentially affect blood composition between the CT scan and blood sample collection, such as blood product transfusion or antitumor therapy.

CT examination and data measurement

All the CT scans were performed using a DLCT detector scanner (IQon Spectral CT; Philips Healthcare, Best, the Netherlands) without the administration of contrast agents. The factory default chest and abdomen scan protocol was employed and the scanning parameters adopted were as follows: tube voltage: 120 kV; adaptive tube current; pitch: 0.5; gantry rotation time: 0.5 s; section collimation: 64×0.625 mm. After completing the scanning process, the acquired data were projected for spatial-spectral reconstruction (spectral level 4), with a thickness and an image spacing of 0.992 mm. A conventional CT image, effective atomic number map, and electron density map were automatically generated from the spectral base image data.

A radiologist with over 5 years of experience in chest and abdomen imaging diagnosis measured the DLCT parameters using a post-processing workstation (IntelliSpace Portal, Version 10.0, Philips Healthcare). In each DLCT parameter map, circular regions of interest (ROIs) were placed in the following blood vessels: the aortic arch, pulmonary trunk bifurcation, and hepatic portal vein (Figure 1). To minimize variability due to individual differences in vascular lumen size, the ROI area for the aortic arch and pulmonary artery was standardized to 2 cm², and that for the portal vein was standardized to 0.5 cm². The drawn ROIs were carefully positioned to avoid contact with the vessel walls. The CT value [Hounsfield units (HU)], effective atomic number, and electron density (%) were investigated for each ROI. The measurement of these DLCT parameter values was repeated three times per section, and the average results were recorded for subsequent analysis.

Figure 1 Example of the measurement of DLCT parameters for a 42-year-old man with a hemoglobin concentration of 132 g/L. The CT value, electron density and effective atomic number were 40.7 HU, 104.0% and 7.33 for aortic arch (A-C), respectively; and 42.2 HU, 104.1% and 7.33 for pulmonary artery (D-F), respectively; and 42.6 HU, 104.3% and 7.24 for portal vein (G-I), respectively. The three columns from left to right represent the parameter maps of CT value, electron density and effective atomic number, respectively. Ar, average area; Av, average value; CT, computed tomography; DLCT, dual-layer computed tomography; HU, Hounsfield unit; SD, standard deviation; Perim, perimeter; Z effective, effective atomic number.

Anemia diagnostic criteria

Under the World Health Organization’s diagnostic criteria (19), anemia is defined as follows: (I) males: a hemoglobin concentration <130 g/L; (II) non-pregnant females: a hemoglobin concentration <120 g/L (20).

Statistical analysis

The required sample size was calculated using a prospective power analysis, with a power of 0.95, a type I error rate of 0.05, and a ratio of sample sizes in the negative-to-positive groups of 1.0. Based on these assumptions, the required sample size was estimated to be approximately 176. Descriptive statistics were used to present the clinical characteristics and DLCT parameter values. To assess the differences in the clinical data and DLCT parameter values between the anemia and normal groups, the t-test or rank-sum test was conducted based on the normality test results. Pearson or Spearman correlation analysis was used to examine the relationships between the DLCT parameter and hemoglobin concentration. One-way analysis of variance or the Kruskal-Wallis test was used to compare the differences in the DLCT parameter values among different blood vessels. When the P value was <0.05, post-hoc tests were subsequently performed for pairwise comparisons among those blood vessels. A receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic performance of DLCT parameters. The optimal cut-off values for diagnosing anemia were determined based on the maximum Youden index. Scatter plots and simple linear regression analysis were employed to compute hemoglobin concentration. To investigate the associations between DLCT parameters of various blood vessels and the occurrence of anemia, logistic regression analysis was employed following univariate analysis. Statistical analyses were performed with significance set at P<0.05.

Results

Differences in the clinical characteristics and DLCT parameters between anemia and normal groups

This retrospective study included 240 patients (110 males, 130 females) with a mean age of 58.1±11.6 years. Hemoglobin concentrations of peripheral blood ranged from 65.00 to 181.00 g/L (mean 125.60±20.01 g/L) across the total study cohort. The hemoglobin concentrations of the female cohort were significantly lower than those of male (P<0.001). There were significant differences in hemoglobin concentrations between the anemia and normal groups (138.40±12.64 vs. 107.96±13.95 g/L, P<0.001). Whether in the total, male, or female cohort, the CT and electron density values of the three vessels in the anemia group were significantly lower than those in the normal group (all P<0.05, Table 1). However, no significant differences in effective atomic number were observed between the two groups (all P>0.05, Table 1).

Differences in the DLCT parameters among different blood vessels

Whether in the total, male, or female cohort, the CT value and effective atomic number showed significant differences among the three blood vessels (all P<0.001), while electron density didn’t (all P>0.05) (Table 2). Post hoc tests indicated that CT value and effective atomic number of the portal vein were significantly lower than that of the aortic arch and pulmonary artery (all P<0.05, Table 2).

Correlation between DLCT parameters and hemoglobin concentration

hemoglobin concentration showed positive correlations with CT value and electron density, with r values ranging from 0.435 to 0.583 and 0.570 to 0.639, respectively. The correlation of the CT value of the portal vein was significantly lower than that of the aortic arch (0.435 vs. 0.583, P=0.029) and slightly lower than that of the pulmonary artery (0.435 vs. 0.527, P=0.192). Regarding the correlation of electron density with hemoglobin concentration, there are no statistically significant differences among the three blood vessels (all P>0.05). No significant correlation was found between hemoglobin concentration and effective atomic number (all P>0.05, Table 3).

Diagnostic performance of DLCT parameters of individual blood vessel on anemia

Due to the lack of significant differences in effective atomic number value between the anemia and normal groups, only CT value and electron density were included in the ROC analysis for diagnosing anemia. Regarding the CT value, aortic arch had an area under the curve (AUC) value of 0.79, significantly higher than portal vein (P=0.008) and slightly higher than pulmonary artery (P=0.126). As to electron density, aortic arch had an AUC value of 0.81, slightly higher than portal vein (P=0.239) and pulmonary artery (P=0.095). Among the six CT predictors, the CT value of the portal vein showed the lowest AUC value of 0.68, significantly lower than the CT value of the aortic arch (P=0.008), electron density of the aortic arch (P=0.002) and electron density of the portal vein (P=0.007) (Figure 2, Table 4).

Figure 2 Diagnostic performance of CT value and electron density on anemia. AA, aortic arch; AUC, area under the curve; CT, computed tomography; ED, electron density; LR, logistic regression; PA, pulmonary artery; PV, portal vein.

Based on the optimal diagnostic performance in detecting anemia, scatter plots, and simple linear regression analyses were further conducted for CT value and electron density of the aortic arch (Figure 3). The fitted regression equation was as follows: (I) hemoglobin concentration (g/L) = 35.99 + 2.16 * CT value (HU); (II) hemoglobin concentration (g/L) = − 2671.4 + 26.91 * electron density (%).

Figure 3 Scatter plot and simple linear regression analysis on the relation of hemoglobin concentration with CT value (A) and electron density (B) in the aortic arch. CT, computed tomography; HU, Hounsfield unit.

Diagnostic performance of logistic regression model

We further performed logistic regression analysis to examine the relationship between the DLCT-derived parameters of multiple vessels and the incidence of anemia. Univariate analysis showed that all parameters were statistically significant (all P<0.05; Table 5). Multivariable logistic regression indicates that CT value of the aortic arch, electron density of the pulmonary artery, and electron density of the portal vein are independent predictors of anemia (Table 5). The logistic regression model constructed using the above three CT indicators demonstrated the best predictive performance for anemia (AUC =0.85), outperforming any single CT predictor of an individual vessel (all P<0.05, Table 4). The regression equations are as follows: ln(p/(1−p)) = 289.52914 − 0.17550 * HU (aortic arch) − 0.93861 * electron density (pulmonary artery) − 1.78095* electron density (portal vein).

Discussion

In this study, we investigated the correlation of DLCT parameters with hemoglobin concentration, as well as the diagnostic performance of DLCT parameters of three blood vessels in the detection of anemia. To the best of our knowledge, this study might be the first in vivo report on the utility of dual-energy CT in anemia, as well as the first to explore the role of CT parameters derived from portal vein blood in diagnosing anemia. Whether in aortic arch, pulmonary artery, or portal vein, both the CT value and electron density were found to be significantly correlated with hemoglobin concentration and showed significant differences between the anemia and normal groups. Among the three vessels, the DLCT parameters of the aortic arch demonstrated relatively higher diagnostic performance. The predictive performance of the logistic model constructed by integrating different DLCT parameters of the three vessels is superior to that of a single vessel. In this study, a positive correlation was found between the CT value and hemoglobin concentration, with correlation coefficients ranging from 0.435 to 0.583. Blood has complex components, mainly including plasma and various blood cell types. The iron-containing Hemoglobin in red blood cells is the main factor causing X-ray attenuation (8), leading to the observed correlation between hemoglobin concentration and CT value in our and previous reports (8,9,21,22). In this study, the correlation between hemoglobin concentration and CT value varied with vascular sites. Among the three blood vessels, the strongest correlation was exhibited for aortic arch (r=0.583), in line with previous findings (r=0.600) (22). In contrast, the weakest correlation was found for portal vein, which may be attributed to the influx of fat components absorbed by the gastrointestinal tract into portal vein. This interpretation can be partially supported by our finding that the CT value of the portal vein is significantly lower than that of the aortic arch and pulmonary artery. Differences in fasting duration or the absorption of fat components among individuals before a CT scan may lead to variation in the content of fat components in portal vein, and thereby negatively influence the correlation between hemoglobin concentration and CT value. Similar findings were also reported in a prior CT study, in which the inferior vena cava showed a weaker correlation with hemoglobin concentration (R²=0.457) compared to the abdominal aorta (R²=0.622) (8). This interesting observation regarding the inferior vena cava might be due to the partial absorption of fatty acids by the rectum during food digestion (23), which subsequently refluxes into the inferior vena cava (8,24). Interestingly, this study also found gender differences in the correlation between CT value and hemoglobin concentration, in accordance with previous studies (13,25,26). Specifically, compared to females, males exhibited a closer relationship. Regrettably, these interesting findings have not yet been well understood.

Derived from projection data collected by DLCT through material decomposition, both electron density and effective atomic number are related to X-ray attenuation (14,15). Previous studies have confirmed the ability of electron density and effective atomic number to distinguish materials and provide information on material compositions (14,15). Regarding blood, hemoglobin is the primary substance affecting X-ray attenuation (8), which supports the positive correlation between electron density and hemoglobin concentration in the present study. Compared to the in vitro findings of Schulz et al. (18), which reported a weaker correlation between electron density and hemoglobin concentration (r=0.37–0.49), our results indicated a stronger correlation (r=0.570–0.639). The difference in correlation between these two studies might result from the differences in CT scanners and the use of anticoagulants. The single-source dual-energy CT utilized in Schulz et al.’s research (18) shows inferior performance on spectral separation, as confirmed by a modeling study (27). In contrast, our study employed the DLCT scanner, which exhibits better performance on spectral separation and high-precision measurement of electron density, according to a previous modeling experiment (14). The administration of anticoagulants on blood samples is necessary for in vitro studies, which might have a negative impact on the investigation of the correlation between hemoglobin concentration and DLCT parameters.

Effective atomic number reflects the average atomic number of all substances in the tissue (14,15,28), theoretically indicating its ability to reflect blood components at the atomic level. No significant difference in hemoglobin concentration was observed between the anemia and normal groups. Additionally, our study found no significant correlation between effective atomic number and hemoglobin concentration. A recent study also didn’t reveal an obvious correlation between effective atomic number and hemoglobin concentration (18). These consistent findings could be attributed to the fact that, despite variations in hemoglobin concentration between the two groups, these changes did not significantly influence the overall elemental compositions of the blood at the atomic level, as detectable by effective atomic number. Therefore, effective atomic number seems to lack the ability to accurately detect anemia, since blood is a complex mixture (18). Similar to the finding in CT value, a significantly lower effective atomic number value was observed in portal vein compared to aortic arch and pulmonary artery in our study, which may also be related to the presence of fat composition within portal vein, as the effective atomic number of fat components is typically significantly lower than that of water and soft tissue components (29).

This study, along with previous research on conventional CT (7-9,26), has confirmed that CT value has a good diagnostic performance for anemia. To date, there are limited reports on the utility of dual-energy CT in diagnosing anemia, as well as on the role of CT parameters derived from portal vein blood in the diagnosis of this condition. As for electron density, a quantitative parameter produced by DLCT, its AUC values in diagnosing anemia ranged from 0.75 to 0.81 in our study, which is higher than those of a recent in vitro report (AUC: 0.70–0.76) (18). This difference may be related to the type of blood samples used in the two studies mentioned above, with one using in vivo samples and the other using ex vivo samples. For example, even with sufficient agitation and reasonable storage of blood samples, blood sedimentation is inevitable. The inhomogeneity caused by blood sedimentation may affect the measurement accuracy of DLCT parameter values. In contrast, our study directly measured the DLCT parameter values of blood in vivo, which may more accurately reflect the actual physiological state of blood and promote clinical applications. Additionally, the use of anticoagulants may also interfere with the diagnostic efficacy of DLCT parameters for anemia, as previously discussed.

To facilitate the clinical application of DLCT in diagnosing anemia, aortic arch, pulmonary artery, and portal vein were selected as the target blood vessels in our study, aiming to explore whether there are differences in diagnostic performance among different vascular sites and whether these differences may be potentially influenced by some physiological factors, such as the oxygenated state of hemoglobin and blood components other than hemoglobin (for example fat components). Similar to our observation, Abbasi et al. (26) also found differences in diagnostic performances on anemia among different large blood vessels, with the highest and lowest AUC values observed in aortic arch (AUC =0.816) and inferior vena cava (AUC =0.755), respectively. Among the individual one of three vessels investigated in the present study, the diagnostic performance of DLCT parameters is the best for aortic arch, whether in terms of CT value or electron density. When using the CT value, the diagnostic performance of the portal vein is significantly lower than that of the aortic arch and pulmonary artery, which may be related to the presence of fat in the portal vein, as discussed above. Nevertheless, the diagnostic performance of the pulmonary artery is slightly lower than that of the aortic arch, but the difference (AUC: 0.73 vs. 0.79) is not statistically significant. This may indicate that the oxygenation level of Hemoglobin does not significantly impact the diagnostic efficacy of the CT value, as theoretically, the main difference between the blood in aortic arch and pulmonary artery is the level of hemoglobin oxygenation. When utilizing electron density, there was no significant difference in diagnostic performance among different blood vessels. Additionally, there was no significant difference among the three vessels in terms of the correlation between hemoglobin concentration and electron density. Therefore, we cautiously suggest that electron density is more suitable for the detection of anemia than the CT value.

To further improve the predictive performance for anemia and fully utilize CT data, we conducted a logistic regression model based on the CT and electron density values from different blood vessels. The results indicate that incorporating CT values or electron density from multiple blood vessels significantly improves anemia diagnosis, compared to using a single vessel. Moreover, CT value of the aortic arch, electron density of the pulmonary artery, and electron density of the portal vein serve as independent predictive factors for the model, indicating that all three vessels contribute to the detection of anemia when using CT. Thus, in our opinion, it is beneficial to use both HU and electron density indicators simultaneously for more accurate prediction of anemia in clinical practice.

This study undoubtedly has some limitations. Firstly, this is a single-center, retrospective study. It is highly necessary to conduct larger-scale, multicenter studies to validate our results. Secondly, the study utilized a specific CT scanner, and the parameter values obtained for diagnosing anemia may not apply to other dual-energy CT scanners. Further exploration should be conducted on scanning device models from multiple manufacturers and different scanning settings to verify the accuracy of this study. Finally, the study did not conduct a stratified analysis based on the severity and etiology of anemia, which is our direction for future research.

Conclusions

In summary, both the CT value and electron density of the aortic arch, pulmonary artery, and portal vein contribute to the diagnosis of anemia, suggesting that DLCT parameters may assist in the non-invasive detection of anemia. The DLCT parameters of the aortic arch demonstrated relatively higher diagnostic performance than those of the pulmonary artery and portal vein in detecting anemia. Additionally, integrating different DLCT parameters (CT value and electron density) of multiple blood vessels may provide improved diagnostic performance.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-24-1671/rc

Funding: This study was supported by Hunan Provincial Natural Science Foundation of China (No. 2025JJ80823).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-24-1671/coif). Z.H. is an employee of Philips Healthcare Company. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by Hunan Cancer Hospital Ethics Committee (No. 2025KYKS44) and the requirement for individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Addo OY, Yu EX, Williams AM, Young MF, Sharma AJ, Mei Z, Kassebaum NJ, Jefferds MED, Suchdev PS. Evaluation of Hemoglobin Cutoff Levels to Define Anemia Among Healthy Individuals. JAMA Netw Open 2021;4:e2119123. [Crossref] [PubMed]
2. Stevens GA, Paciorek CJ, Flores-Urrutia MC, Borghi E, Namaste S, Wirth JP, Suchdev PS, Ezzati M, Rohner F, Flaxman SR, Rogers LM. National, regional, and global estimates of anaemia by severity in women and children for 2000-19: a pooled analysis of population-representative data. Lancet Glob Health 2022;10:e627-39. [Crossref] [PubMed]
3. Riva E, Colombo R, Moreo G, Mandelli S, Franchi C, Pasina L, Tettamanti M, Lucca U, Mannucci PM, Nobili A. Prognostic value of degree and types of anaemia on clinical outcomes for hospitalised older patients. Arch Gerontol Geriatr 2017;69:21-30. [Crossref] [PubMed]
4. Nakavachara P, Petchkul J, Jeerawongpanich K, Kiattisakthavee P, Manpayak T, Netsakulnee P, Chaichanwattanakul K, Pooliam J, Srichairatanakool S, Viprakasit V. Prevalence of low bone mass among adolescents with nontransfusion-dependent hemoglobin E/β-thalassemia and its relationship with anemia severity. Pediatr Blood Cancer 2018;
5. Yılmaz E, Yılmaz Z, Çakmak B, Gültekin İB, Çekmez Y, Mahmutoğlu S, Küçüközkan T. Relationship between anemia and depressive mood in the last trimester of pregnancy. J Matern Fetal Neonatal Med 2017;30:977-82. [Crossref] [PubMed]
6. Faraoni D, DiNardo JA, Goobie SM. Relationship Between Preoperative Anemia and In-Hospital Mortality in Children Undergoing Noncardiac Surgery. Anesth Analg 2016;123:1582-7. [Crossref] [PubMed]
7. Zhou QQ, Yu YS, Chen YC, Ding BB, Fang SY, Yang X, Zhang B, Zhang H. Optimal threshold for the diagnosis of anemia severity on unenhanced thoracic CT: A preliminary study. Eur J Radiol 2018;108:236-41. [Crossref] [PubMed]
8. Feng X, Kong Y, Yu H. An Exploratory Investigation into Optimal Indicators and Thresholds for Anemia Diagnosis Through Abdominal CT Plain Scans. Altern Ther Health Med 2023;29:738-43.
9. Zhang J, Wu M, Huang J, Li S, Ye Z. Comparison between thoracic low-dose computed tomography and conventional-dose computed tomography in evaluating anemia: A preliminary study in a Chinese screening cohort. Front Cardiovasc Med 2022;9:987753. [Crossref] [PubMed]
10. Desaint P, Bernard C, Contou D. Diagnosing Anemia with Chest CT-Scan: The Aortic Ring Sign. Am J Med Sci 2022;363:e17. [Crossref] [PubMed]
11. Corcoran HL, Cook DE, Proto AV. Diagnosis of anemia on computed tomography scans of the thorax. J Comput Tomogr 1988;12:116-21. [Crossref] [PubMed]
12. Iuga AI, Pennig L, Caldeira LL, Maintz D, Hickethier T, Doerner J. Prediction of anemia on enhanced computed tomography of the thorax using virtual non-contrast reconstructions. Medicine (Baltimore) 2021;100:e28014. [Crossref] [PubMed]
13. Title RS, Harper K, Nelson E, Evans T, Tello R. Observer performance in assessing anemia on thoracic CT. AJR Am J Roentgenol 2005;185:1240-4. [Crossref] [PubMed]
14. Hua CH, Shapira N, Merchant TE, Klahr P, Yagil Y. Accuracy of electron density, effective atomic number, and iodine concentration determination with a dual-layer dual-energy computed tomography system. Med Phys 2018;45:2486-97. [Crossref] [PubMed]
15. Rajiah P, Parakh A, Kay F, Baruah D, Kambadakone AR, Leng S. Update on Multienergy CT: Physics, Principles, and Applications. Radiographics 2020;40:1284-308. [Crossref] [PubMed]
16. García-Figueiras R, Oleaga L, Broncano J, Tardáguila G, Fernández-Pérez G, Vañó E, Santos-Armentia E, Méndez R, Luna A, Baleato-González S. What to Expect (and What Not) from Dual-Energy CT Imaging Now and in the Future? J Imaging 2024;10:154. [Crossref] [PubMed]
17. Borges AP, Antunes C, Curvo-Semedo L. Pros and Cons of Dual-Energy CT Systems: "One Does Not Fit All". Tomography 2023;9:195-216. [Crossref] [PubMed]
18. Schulz B, Euler A, Schmid HR, Kubik-Huch RA, Thali M, Niemann T. In vitro blood sample assessment: investigating correlation of laboratory hemoglobin and spectral properties of dual-energy CT measurements (ρ/Z). Eur Radiol 2024;34:7934-43. [Crossref] [PubMed]
19. WHO. Haemoglobin concentrations for the diagnosis of anaemia and assessment of severity. Vitamin and Mineral Nutrition Information System. World Health Organization, Geneva. 2011. Available online: https://www.who.int/publications/i/item/WHO-NMH-NHD-MNM-11.1. Accessed 05/07/2024.
20. McLean E, Cogswell M, Egli I, Wojdyla D, de Benoist B. Worldwide prevalence of anaemia, WHO Vitamin and Mineral Nutrition Information System, 1993-2005. Public Health Nutr 2009;12:444-54. [Crossref] [PubMed]
21. Jung C, Groth M, Bley TA, Henes FO, Treszl A, Adam G, Bannas P. Assessment of anemia during CT pulmonary angiography. Eur J Radiol 2012;81:4196-202. [Crossref] [PubMed]
22. Kamel EM, Rizzo E, Duchosal MA, Duran R, Goncalves-Matoso V, Schnyder P, Qanadli SD. Radiological profile of anemia on unenhanced MDCT of the thorax. Eur Radiol 2008;18:1863-8. [Crossref] [PubMed]
23. Cummings JH, Pomare EW, Branch WJ, Naylor CP, Macfarlane GT. Short chain fatty acids in human large intestine, portal, hepatic and venous blood. Gut 1987;28:1221-7. [Crossref] [PubMed]
24. Netter FH. Atlas of Human Anatomy Professional Edition. 6th ed. Philadelphia, PA: Elsevier; 2014.
25. Lan H, Nishihara S, Nishitani H. Accuracy of computed tomography attenuation measurements for diagnosing anemia. Jpn J Radiol 2010;28:53-7. [Crossref] [PubMed]
26. Abbasi B, Seyed Hosseini M, Moodi Ghalibaf A, Akhavan R, Emadzadeh M, Bolvardi E. Evaluating anemia on non-contrast thoracic computed tomography. Sci Rep 2022;12:21380. [Crossref] [PubMed]
27. Almeida IP, Schyns LE, Öllers MC, van Elmpt W, Parodi K, Landry G, Verhaegen F. Dual-energy CT quantitative imaging: a comparison study between twin-beam and dual-source CT scanners. Med Phys 2017;44:171-9. [Crossref] [PubMed]
28. Kalender WA, Perman WH, Vetter JR, Klotz E. Evaluation of a prototype dual-energy computed tomographic apparatus. I. Phantom studies. Med Phys 1986;13:334-9. [Crossref] [PubMed]
29. Vrbaški S, Arana Peña LM, Brombal L, Donato S, Taibi A, Contillo A, Longo R. Characterization of breast tissues in density and effective atomic number basis via spectral X-ray computed tomography. Phys Med Biol 2023; [Crossref]