Repeatability and reproducibility of hemodynamic measurements by intracranial 4D Flow MRI: a multi‑vendor and multi‑software cross‑over comparison study

Repeatability and reproducibility of hemodynamic measurements by intracranial 4D Flow MRI: a multi‑vendor and multi‑software cross‑over comparison study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Repeatability and reproducibility of hemodynamic measurements by intracranial 4D Flow MRI: a multi‑vendor and multi‑software cross‑over comparison study Mingfang Luo, Lini Liu, Rui Li, Daniel Giese, Binbin Sui, Yuting Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7170747/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 28 Nov, 2025 Read the published version in BMC Medical Imaging → Version 1 posted 10 You are reading this latest preprint version Abstract Background: Multiple protocols for acquiring intracranial four-dimensional flow magnetic resonance (4D Flow MR) data and post-processing software employed to obtain hemodynamic parameters were available. However, insufficient validation exists to quantify the difference in measurements. This study aims to perform a comprehensive cross-comparison analysis of quantitative results from three major vendors and three post-processing software platforms. Methods: Intracranial 4D Flow MR was conducted on 20 healthy volunteers using three different 3.0T MRI scanners. For each subject, nine major intracranial arteries (a total of 180 arteries) were analyzed using three post-processing software platforms: CVI42, MASS, and TS FLOW. Key hemodynamic parameters—including flow rate, maximum velocity (Vmax), average velocity (Vavg), and delineated vessel area—were extracted. The study compared results across different vendors and software, and further assessed both intra-observer and inter-observer consistency. Results: Regardless of the post-processing software used, no significant differences were observed in the flow rate across different scanners (p > 0.05), with ICC = 0.700-0.919. For flow rate and Vmax, the MASS and CVI42 software were equivalent. For Vavg, TS FLOW and CVI42 were tested equivalent. For vessel area, various agreement was shown across software (ICC = 0.083-0.577). Additionally, the TS FLOW software exhibited repeatability and reproducibility (intra-observer ICC = 0.724-0.998; inter-observer ICC = 0.590-0.981). Conclusions: Scanning protocols by different vendors appear to result in smaller variations in hemodynamic measurements compared to using different post-processing software methods. It is recommended to employ single post-processing software or conduct software calibration when using multiple software platforms to ensure consistency in multi-center clinical studies. 4D Flow Phase‐contrast magnetic resonance imaging Intracranial arteries Hemodynamics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background Hemodynamics evaluation is crucial for understanding pathophysiology and clinical evaluation of cardiovascular diseases. 4D Flow MR offers the unique opportunity to gather time-resolved, flow-encoding information in all three spatial directions of the heart and blood vessels to non-invasively characterize blood flow without contrast agent [1]. As an extension of 2D Flow MRI, it also has the advantage of allowing retrospective evaluation of the acquired 3D volume without the necessity of pre-planning or repeated scans and is therefore particularly useful to evaluate complex vascular anatomies and various flow patterns [2]. Due to the small size and tortuous anatomy of intracranial vessels, obtaining sufficient spatial resolution for cerebral arteries by 4D Flow could be technically challenging, which limited the clinical application. In recent years, with the key advances of various post-processing methods and further acceleration in acquisition, 4D Flow has been increasingly applied to visualize and quantitatively analyze the blood flow in a variety of intracranial vascular diseases including aneurysms, arteriovenous malformation, arterial stenosis, etc [3-5]. Encouraging findings have provided novel insights into disease-related hemodynamic mechanisms, such as the inflow-to-outflow ratio could accurately distinguish between non-hemorrhagic and hemorrhagic arteriovenous malformation [6], and transcranial pulse wave velocity were significantly higher in Alzheimer’s disease compared to age-matched controls [7]. However, there is a general lack in standardization among various scanners, post-processing software, as well as the evidence-based merits on benefits of clinical applications of 4D Flow [8]. Studies of repeatability and reproducibility of 4D Flow evaluating heart and aorta have been reported, showing that inter- and intrareader consistency for clinically used parameters was better for some software (Caas and MevisFlow) but worse for some other (Circle), and was better for some parameters (stroke volume, peak flow, and forward flow) but worse for some other (peak maximum velocity) [9, 10]. 4D Flow cardiovascular magnetic resonance consensus statement have been published in 2015 and updated in 2023, emphasizing the importance of clinical quality assurance and validation [1]. For the intracranial application of 4D Flow, it is important to recognize and measure the differences among various MRI scanners and post-processing programs, all of which must be considered when making comparisons or designing multi-center trials. A previous study has shown that using the same post-processing software and same vendor for multi-center repetitive assessment of intracranial 4D Flow, good repeatability in the measurement of blood flow and peak velocity were achieved [11]. However, the reproducibility across different scanners and post-processing platforms remains unclear. Although time-resolved RF-spoiled gradient echo sequence was generally applied by multiple vendors, technical discrepancies may exist due to 1) gradient system performance; 2) receiver coil array configurations; 3) selection of acceleration factors (e.g., SENSE vs. GRAPPA); 4) intravoxel phase dispersion caused by turbulent flow, etc [1, 12]. In addition, different post-processing platforms may include eddy current correction or not, employ manual or semi-automatic vessel segmentation, and have different interaction design for vessel area delineation. From a clinical perspective, it is essential to validate the repeatability and reproducibility of intracranial 4D Flow in healthy subjects before applying to cerebral vascular diseases with hemodynamic abnormalities [13, 14]. Therefore, this study aims to perform a comprehensive cross-comparison analysis of key hemodynamic measurements by intracranial 4D Flow in healthy subjects using 3.0T scanners (i) from three vendors and (ii) by three post-processing software packages. Evaluation of inter- and Intra-observer consistency was also performed. Methods Subjects This cross-sectional study was conducted in accordance with approval from the Ethics Committee of Sichuan Provincial People's Hospital (No. 2023-409-1). Between July and August 2024, we prospectively recruited 20 healthy adults meeting stringent inclusion criteria: absence of cardiovascular/cerebrovascular history, major risk factors (smoking, diabetes, hypertension), and confirmed normal cerebrovascular anatomy through baseline imaging. Image acquisition All subjects were examined on three 3.0T MRI scanners: MRI1 (MAGNETOM VIDA, Siemens Healthineers), MRI2 (SIGNA Architect, GE Healthcare) and MRI3 (uMR 880, United Imaging Healthcare), with a 32-channel head coil. Prior to each scan, the subjects would rest in a resting state for 15-20 minutes to ensure the conditions (such as blood pressure) were close to similar baseline level. The 4D Flow scanning parameters of the three scanners has been standardized as follows: slice thickness is interpolated to 1 mm using zero-padding interpolation, spatial resolution after reconstruction was 1×1×1 mm³, velocity encoding in all three directions was set to 100 cm/s [15], and 48 axial slices were acquired. The intracranial Circle of Willis was covered using 3D TOF-MRA sequence and pulse gating was used for 4D Flow imaging. Acquisition parameters across the three scanners have been carefully adjusted to closely match each other. Detailed parameters were provided in Table 1. Image post-processing The generated three datasets (20 subjects each) were imported into three commercial post-processing software programs: CVI42 (Circle Cardiovascular Imaging), MASS (Version 2024-EXP, Leiden University Medical Center, Leiden, The Netherlands), TS FLOW (Beijing Qingying Huakang Technology Co., LTD.). Key parameters such as flow rate, Vavg, Vmax, and vessel area of the delineated was extracted. Two operators with over five years of experience in clinical practice of radiology independently conducted evaluations using each software. The workflow encompasses conducting background phase offsets correction, anti-aliasing, segmentation, visualization and quantification. Additionally, both TS FLOW software and CVI42 permitted manual adjustment of the segmentation threshold, but TS FLOW and MASS did not support manual correction for background phase offsets and anti-aliasing (Supplementary Materials 1). Nine intracranial arteries were measured, including the including the internal carotid artery (ICA) at the C3-C4 segment, the anterior cerebral artery (ACA) at the A2 segment, the middle cerebral artery (MCA) at the M1 segment, the posterior cerebral artery (PCA) at the P1 segment, and the basilar artery (BA) before the inferior cerebellar artery [16]. To ensure optimal plane selection, each vessel was positioned centrally within its respective cross-section. Figure 1 illustrates the data processing interfaces of the three software programs and the display of the cerebral artery measurement cross-sections. It is important to note that when delineated vessel area, TS FLOW software employs semi-automatic delineation, MASS utilizes manual registration for customized delineation, and CVI42 implements automated registration-based customized delineation (Supplementary Materials 2). Statistical analysis Depending on the data distribution (normal or non-normal), paired t-tests, Wilcoxon signed-rank tests, or Friedman tests were employed to compare the flow rate differences among the three scanners. The flow rate values for intracranial arteries were presented as mean ± SD. Software across comparisons were performed using the Kruskal-Wallis (K-W) test, followed by pairwise comparisons. p < 0.05 were considered for statistical significance. Additionally, Bland-Altman analysis and ICC were used to evaluate variability and equivalence between software. The interpretation of ICC was as follows: >0.9: excellent; 0.75-0.9: good; 0.5-0.75: moderate; <0.5: poor [17]. In the equivalence test, the equivalence limits were determined by the maximum standard deviation obtained from the analysis of the intra-observer variability (Supplementary Materials 3), which was ±1.96 standard deviation (SD). That was, if the deviation between two software systems falls within the 95% range of the deviation produced by a single observer using the same software for repeated measurements, then these two software systems were considered equivalent [18]. All statistical analyses were conducted using SPSS version 22.0 (IBM Corporation). Results Subject characteristics This study included 20 healthy subjects (mean age 38.7±16.88 years, 10 female), none of whom had a history of cardiovascular or cerebrovascular diseases or other significant medical conditions. Baseline characteristics are summarized in Table 2. Variation across three scanners For scanners comparison, as shown in Supplementary Materials 4, there were no statistically significant differences in the flow rate measurements of any intracranial arteries obtained from 20 subjects across three scanners using any post-processing software ( p > 0.05). The results demonstrated moderate consistency (ICC = 0.700-0.919). Variation across three post-processing software Flow rate TS FLOW demonstrated significantly higher flow rates across all arterial segments compared to MASS and CVI42 ( p 0.05; Figure 2). The Bland-Altman analysis revealed that the difference range between MASS and CVI42 was the narrowest (e.g., BA LOA: -42.4 to 23.0 mL/min) except in the LACA (Supplementary Materials 5). The consistency between MASS and CVI42 was moderate to excellent (ICC: 0.541-0.765; Table 3), with highest consistency in the LMCA (ICC = 0.765). Vavg As shown in Figure 3, TS FLOW measured significantly higher Vavg than CVI42 in the RICA ( p = 0.013) and LPCA ( p = 0.034), with no inter-software differences in other arteries ( p > 0.05). In contrast, MASS exhibited systematically lower Vavg compared to both TS FLOW and CVI42 across all arteries (all p < 0.05). Inter-software agreement analysis revealed moderate-to-high consistency between TS FLOW and CVI42 (ICC: 0.533-0.891), peaking in the LMCA (ICC = 0.891), whereas MASS showed poor concordance with CVI42 (ICC < 0.5; Table 3). Bland-Altman analysis (Supplementary Materials 6) confirmed superior measurement consistency between TS FLOW and CVI42, with narrowest limits of agreement in the LMCA (−4.99 to 5.84 cm/s) and widest in the LACA (−7.22 to 9.14 cm/s). Vmax The Vmax measured by TS FLOW was consistently higher than that measured by MASS and CVI42 in all arteries ( p MASS by 71.7%) and minimal in the RICA (TS FLOW > CVI42 by 7.8%). Notably, MASS measured significantly lower Vmax than CVI42 in the RACA ( p = 0.013; Figure 4). Inter-software consistency analysis (Table 3) revealed moderate MASS-CVI42 concordance across intracranial arteries (ICC: 0.510-0.612), whereas poor agreement was shown for TS FLOW-MASS (ICC: 0.105-0.332). TS FLOW-CVI42 comparisons exhibited suboptimal consistency (ICC range: 0.098-0.312) except in the LMCA (ICC = 0.503). Bland-Altman analysis (Supplementary Materials 7) demonstrated superior measurement consistency between MASS and CVI42 with narrowest LOA in the LACA (−9.68 to 19.81 cm/s) and widest in the LPCA (−15.46 to 23.85 cm/s). Delineated Vessel Area The vessel area of TS flow on ICA and BA was larger than that of the other two software ( p < 0.05; Figure 5). In the consistency test, only MASS and CVI42 showed moderate consistency in the vessel area of the LICA (ICC = 0.557; Table 3). Intra- observer Variability as measure of repeatability In the intra-observer variability analysis, both the TS FLOW software and CVI42 demonstrated moderate to excellent consistency (ICC > 0.5) in terms of flow rate, Vavg, Vmax, and vessel area. The MASS software also showed moderate or excellent consistency (ICC > 0.5) in flow rate, Vavg and Vmax, but for 6 out of 9 arteries, the ICC was less than 0.5 in vessel area (Supplementary Materials 8). For CVI42 software, the ICC of flow rate exhibited a significant positive correlation with the ICC of vessel area ( p = 0.047, r = 0.674, Supplementary Materials 9). Inter-observer Consistency as measure of reproducibility In the inter-observer variability analysis, TS FLOW demonstrated moderate to excellent consistency (ICC = 0.590-0.981) in flow rate, Vavg, Vmax, and vessel area. In contrast, MASS exhibited poor interobserver agreement for flow rate in the posterior and anterior cerebral arteries (RPCA/LACA ICC = 0.317/0.179), suboptimal consistency for Vavg (LPCA/RPCA/LACA/RACA ICC = 0.249-0.480), and poor consistency for vessel area measurements (ICC = 0.107-0.650). CVI42 achieved moderate interobserver agreement for Vavg and Vmax (ICC = 0.530-0.968) across all vessels but showed heterogeneous ICC distributions (0.21–0.83) for flow rate and vessel area without clear patterns (Supplementary Materials 10). For CVI42 software, the ICC of flow rate exhibited a significant positive correlation with the ICC of vessel area ( p = 0.029, r = 0.719, Supplementary Materials 11). Equivalence test As for the flow rate, it was observed that MASS and CVI42 were all within the equivalent range (except for LACA). As for Vavg, TS FLOW and CVI42 were within the equivalent range in each intracranial artery. As for Vmax, the differences among the three software programs were all within the equivalent range (Figure 6). Discussion This study comprehensively evaluated the across-vendor and across-software repeatability and reproducibility of multiple hemodynamic measurements of major arteries by intracranial 4D Flow MRI in healthy subjects. Possible confounders were carefully controlled. Subjects were scanned three times on the same day under resting-state baseline conditions. Key parameters of acquisition were similar across vendors, post-processing pipelines were standardized, and variances were quantified. The results revealed no difference in flow rate measurement across different mainstream vendors. The consistency across three post-processing software, however, seems less satisfactory. The flow rate obtained by TS FLOW was higher than that by the other two software programs (18.81% to 49.8% higher). On the other hand, TS FLOW, which employs semi-automatic delineation, exhibited higher reproducibility and repeatability. This study provides a reference for the quantification of intracranial 4D Flow data in various cerebrovascular pathological conditions and also for the acquisition and analysis of future multi-center studies. Focus on scanners In clinical practice, the 4D Flow magnetic resonance sequences provided by different manufacturers exhibit varying degrees of inconsistency [19]. Therefore, in this study, we endeavored to achieve matching for as many key parameters as possible, including flow velocity encoding, acceleration factor, spatial resolution, temporal resolution, and reconstruction phase. The results demonstrate that, in the most stable blood flow rate metric, there is good consistency across manufacturers. Wen et al. evaluated the performance of neurovascular 4D Flow MRI by assessing multiple healthy volunteers using the same scanner model at three distinct locations. Their findings indicate that hemodynamic parameters, such as blood flow rate and peak velocity, exhibit excellent multi-center repeatability, test-retest reliability, and inter-observer agreement [11]. These results basically aligned with ours, suggesting good consistency when using essentially the same technical parameters of acquisition. Focus on post- processing This study found less satisfactory consistency of measurements by different software. The quantitative values of TS FLOW for flow rate and delineated area seemed higher than those of the other two software. We speculate that the differences in flow rate and Vmax may be associated with variations in delineated vessel area. Notably, the three post-processing software packages differ in their ROI delineation methods. TS FLOW employs semi-automatic delineation, MASS uses manual registration with customizable delineation and contour correction, and CVI42 utilizes automatic registration delineation with manual contour correction (Supplementary Materials 2). Furthermore, Vmax is typically determined by the peak value in included voxels, whereas the flow rate is calculated as the sum of velocities across all voxels within the ROI at each time phase, multiplied by the region's area [20]. However, since the semi-automatic segmentation delineation method cannot manually correct the vascular contour of each phase, this might be one of the factors leading to an overestimation of blood flow. This may suggest why the flow rate and Vmax values obtained by TS FLOW software tend to be larger in intracranial vessels. In addition, the differences among various software packages could also be attributable to disparities in post-processing algorithms, such as background phase correction, interpolation techniques and the partial volume effect [21-23]. Therefore, in future multi-center studies, the above-mentioned influencing factors should be fully considered when using different post-processing software to ensure the consistency and comparability of the results. Focus on equivalence test Furthermore, the results of the equivalence test indicated that within the pre-defined clinically acceptable difference range, despite the technical differences among the three software, they still demonstrated clinical interchangeability in terms of hemodynamic core parameters, especially the Vmax. This finding provides empirical evidence for the use of different analysis platforms in multi-center studies. However, it should be noted that due to the complexity of intracranial arterial anatomy and the current limitations of 4D flow MRI sequence resolution, local measurement differences may occur in intracranial stenosis or small vessels (such as LACA). In such cases, rigorous manual verification during the post-processing stage is necessary to ensure data accuracy. Inter- and Intra-observer Variability For the CVI42 software, the consistency of flow rate was poor and no clear pattern is observed. It is worth noting that the intra- and inter- observer ICC of flow rate and that of delineated vessel area were positively correlated, indicating that the repeatability and reproducibility of flow rate measurement by CVI42 may mainly depend on the consistency of vessel area delineation. The reason for this poor consistency may be attributed to the subjectivity inherent in manual segmentation during ROI delineation. Oechtering et al. also confirmed that in the validation of Cardiovascular 4D Flow, CVI42 exhibited significant variation in contour delineation [9]. It is assumed that the measurements could be more influenced by manual delineation by different observers when using MASS and CVI42. Therefore, it is recommended that the selected planes and the size of the ROI should be carefully determined at each stage of blood flow measurement, especially when using the manual delineation method. A previous study has compared the hemodynamic measurements obtained using a fully automated post-processing method based on deep learning with that measured by human observers, showing that this method can effectively reduce the post-processing time for 4D Flow data in the ventricles while minimizing inter-observer differences [24]. However, customizable delineation and contour correction could allow for analysis of more specified target vessels in the setting of various cerebrovascular pathologies (such as bridge artery after bypass surgery), which would require further investigation. Limitations There are several limitations to our study. Firstly, due to the lack of a golden standard for intracranial flow measurements, it is not feasible to determine which software provides more accurate parameter values. Second, this study quantified the differences among the software, but the source of difference cannot be fully determined. Moreover, this is a single center study without validation of test-retest consistency at external centers. Finally, since the three software programs share only a limited set of quantitative values, other more advanced hemodynamic parameters, such as wall shear stress and pressure gradient, were not included into analysis in this study. Conclusions In conclusion, our study reveals that in the 4D Flow analysis of intracranial arteries, scanning by different vendors appear to result in smaller variations in hemodynamic measurements compared to using different post-processing software methods. Software that employs semi-automatic delineation exhibited general better repeatability and reproducibility. This could serve as a reference for the acquisition and analysis of future multi-center studies. Abbreviations 4D Flow MR: Four-dimensional flow magnetic resonance ICA: Internal carotid artery ACA: Anterior cerebral artery MCA: Middle cerebral artery PCA: Posterior cerebral artery BA: Basilar artery Vavg: Average velocity Vmax: Maximal velocity ROI: Region of interest SD: Standard deviation ICC: Intraclass correlation coefficients Declarations Ethics approval and consent to participate This cross-sectional study was conducted in accordance with approval from the Ethics Committee of Sichuan Provincial People's Hospital (No. 2023-409-1). All participants have provided written informed consent. Consent for publication Not Applicable. Availability of data and materials The patient data and images supporting this study are not publicly available due to patient confidentiality restrictions under the ethical approval by the Ethics Committee of Sichuan Provincial People's Hospital (No. 2023-409-1). De-identified data may be made available from the corresponding author upon reasonable request, subject to institutional ethical approval and execution of a data access agreement. Competing interests The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article. Although some co-authors are from Siemens Healthcare Ltd., they contributed mainly by providing WIP (work in progress) sequence, and all authors only have relationships of academic cooperation. Funding This study has received funding by National Natural Science Foundation (Youth Project), grant No. 82202123. Author information Mingfang Luo and Lini Liu contributed equally to this work. Authors' contributions M.L designed the study, analyzed and interpreted the data, and contributor in writing the manuscript. L. L analyzed and interpreted the data and contributed to write the manuscript. R.L conceptualized the study and interpreted the data. D.G contributed mainly by providing WIP (work in progress) sequence. B.S supervised the study and revised the manuscript. Y.W designed, supervised, provided funding and contributor in writing the manuscript. Acknowledgements The authors thank all the volunteers participating, and thank Meining Chen from MR Research Collaboration, Siemens Healthineers Ltd., Liqiang Zhou from the United Imaging Healthcare and Ruzhi Zhang from GE Healthcare for their professional technical support during the implementation of flow imaging sequences. References Bissell M M, Raimondi F, Ait Ali L, et al. 4D Flow cardiovascular magnetic resonance consensus statement: 2023 update. J Cardiovasc Magn Reson. 2023;25:40. Rizk J. 4D flow MRI applications in congenital heart disease. Eur Radiol. 2021;31:1160-1174. Ansari S A, Schnell S, Carroll T, et al. Intracranial 4D flow MRI: toward individualized assessment of arteriovenous malformation hemodynamics and treatment-induced changes. AJNR Am J Neuroradiol. 2013;34:1922-8. Morgan A G, Thrippleton M J, Wardlaw J M, et al. 4D flow MRI for non-invasive measurement of blood flow in the brain: A systematic review. J Cereb Blood Flow Metab. 2021;41:206-218. Peng F, Xia J, Zhang F, et al. Intracranial aneurysm instability prediction model based on 4D-Flow MRI and HR-MRI. Neurotherapeutics. 2025;22:e00505. Takeda Y, Kin T, Sekine T, et al. Hemodynamic Analysis of Cerebral AVMs with 3D Phase-Contrast MR Imaging. AJNR Am J Neuroradiol. 2021;42:2138-2145. Rivera-Rivera L A, Cody K A, Eisenmenger L, et al. Assessment of vascular stiffness in the internal carotid artery proximal to the carotid canal in Alzheimer's disease using pulse wave velocity from low rank reconstructed 4D flow MRI. J Cereb Blood Flow Metab. 2021;41:298-311. Boban M. Editorial for "Traveling Volunteers - A Multi-Vendor, Multi-Center Study on Reproducibility and Comparability of 4D Flow Derived Aortic Hemodynamics in Cardiovascular Magnetic Resonance". J Magn Reson Imaging. 2022;55:223-224. Oechtering T H, Nowak A, Sieren M M, et al. Repeatability and reproducibility of various 4D Flow MRI postprocessing software programs in a multi-software and multi-vendor cross-over comparison study. J Cardiovasc Magn Reson. 2023;25:22. Demir A, Wiesemann S, Erley J, et al. Traveling Volunteers: A Multi-Vendor, Multi-Center Study on Reproducibility and Comparability of 4D Flow Derived Aortic Hemodynamics in Cardiovascular Magnetic Resonance. J Magn Reson Imaging. 2022;55:211-222. Wen B, Tian S, Cheng J, et al. Test-retest multisite reproducibility of neurovascular 4D flow MRI. J Magn Reson Imaging. 2019;49:1543-1552. Soulat G, McCarthy P, Markl M. 4D Flow with MRI. Annu Rev Biomed Eng. 2020;22:103-126. Dyverfeldt P, Bissell M, Barker A J, et al. 4D flow cardiovascular magnetic resonance consensus statement. J Cardiovasc Magn Reson. 2015;17:72. Burkhardt B E U, Kellenberger C J, Callaghan F M, et al. Flow evaluation software for four-dimensional flow MRI: a reliability and validation study. Radiol Med. 2023;128:1225-1235. Wu C, Schnell S, Vakil P, et al. In Vivo Assessment of the Impact of Regional Intracranial Atherosclerotic Lesions on Brain Arterial 3D Hemodynamics. AJNR Am J Neuroradiol. 2017;38:515-522. Zarrinkoob L, Wåhlin A, Ambarki K, et al. Quantification and mapping of cerebral hemodynamics before and after carotid endarterectomy, using four-dimensional flow magnetic resonance imaging. J Vasc Surg. 2021;74:910-920.e1. Koo T K, Li M Y. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med. 2016;15:155-63. Zange L, Muehlberg F, Blaszczyk E, et al. Quantification in cardiovascular magnetic resonance: agreement of software from three different vendors on assessment of left ventricular function, 2D flow and parametric mapping. J Cardiovasc Magn Reson. 2019;21:12. Xie L, Zhang Y, Hong H, et al. Higher intracranial arterial pulsatility is associated with presumed imaging markers of the glymphatic system: An explorative study. Neuroimage. 2024;288:120524. Gatehouse P D, Keegan J, Crowe L A, et al. Applications of phase-contrast flow and velocity imaging in cardiovascular MRI. Eur Radiol. 2005;15:2172-84. Bock J, Frydrychowicz A, Stalder A F, et al. 4D phase contrast MRI at 3 T: effect of standard and blood-pool contrast agents on SNR, PC-MRA, and blood flow visualization. Magn Reson Med. 2010;63:330-8. Stalder A F, Russe M F, Frydrychowicz A, et al. Quantitative 2D and 3D phase contrast MRI: optimized analysis of blood flow and vessel wall parameters. Magn Reson Med. 2008;60:1218-31. Petersson S, Dyverfeldt P, Ebbers T. Assessment of the accuracy of MRI wall shear stress estimation using numerical simulations. J Magn Reson Imaging. 2012;36:128-38. Corrado P A, Wentland A L, Starekova J, et al. Fully automated intracardiac 4D flow MRI post-processing using deep learning for biventricular segmentation. Eur Radiol. 2022;32:5669-5678. Tables Table 1. Scan parameters of three scanners. Siemens VIDA GE SIGNA Architect United imaging uMR 880 Field of view (mm 3 ) 180*168*48 180*180*48 180*165*48 Slice thickness (mm) 1 1 1 Spatial resolution (mm 3 ) 1.36*1.02*1.24 1.0*1.0*1.0 1.02*1.02*1.26 Reconstructed spatial resolution (mm 3 ) 1.02*1.02*1.0 1.0*1.0*1.0 1.02*1.02*1.0 Number of heart phases 9-15 9-15 9-15 Number of reconstructed heart phases 20 20 20 Pulse synchronization Retrospective Retrospective Retrospective Flip Angle 16 15 16 Velocity‐encoding factor 100 100 100 Acceleration mode CS hyperkat ACS Acceleration factor 8.9 6 4 TE (ms) 3.78 minimum 3.73 TR (ms) 52.2 23.2 28 Scan time (min) 4-7 4-7 6-13 Note: TE = echo time, TR = repetition time, CS = Compressed Sensing, hyperkat = Hyper-accelerated k-t Acquisition Technique, and ACS = AI-assisted Compressed Sensing. Table 2. Demographics of included subjects Variables Healthy Subjects (N=20) Age (years) 38.7 ± 16.88 (22-72 years) Gender radio (male: female) 10:10 Heart rate (bpm) 67±9 Weight (kg) 61.35±9.13 Height (cm) 167.1±7.48 Body mass index (m 2 ) 21.89±2.28 Table 3. Variation across three software Flow rate ml/min TS FLOW MASS CVI42 ICC a ICC b ICC c LICA 327.96±62.18 234.51±39.59 237.93±34.23 0.716* 0.275 0.299 BA 202.74±33.91 135.34±21.98 144.94±20.29 0.635* 0.225 0.172 LMCA 168.69±33.27 135.09±26.05 141.98±31.78 0.765** 0.546* 0.410 LPCA 87.53±14.02 60.50±10.78 65.32±15.36 0.541* 0.260 0.139 LACA 79.44±12.25 65.39±14.26 63.49±19.11 0.546* 0.390 0.244 Vavg cm/s LICA 28.40±6.37 23.69±5.95 31.36±8.39 0.295 0.533* 0.537* BA 30.79±4.20 25.32±6.36 33.67±5.55 0.233 0.552* 0.272 LMCA 38.02±6.14 32.06±7.05 37.60±5.52 0.472 0.891** 0.403 LPCA 27.84±3.62 18.89±4.13 25.56±3.88 0.257 0.553* 0.163 LACA 26.72±4.94 19.19±4.98 25.75±4.49 0.110 0.608* 0.269 Vmax cm/s LICA 83.12±20.97 53.27±16.10 62.30±11.16 0.532* 0.108 0.176 BA 78.18±10.74 67.64±9.01 71.45±8.27 0.510* 0.150 0.330 LMCA 87.28±11.70 71.27±10.45 75.84±11.16 0.537* 0.503* 0.332 LPCA 88.65±20.11 51.63±11.34 55.82±9.48 0.511* 0.098 0.105 LACA 55.74±10.07 46.63±8.97 51.69±10.08 0.612* 0.312 0.277 area mm 2 LICA 23.38±4.19 16.98±3.27 15.84±3.70 0.557* 0.194 0.287 BA 12.94±2.50 9.30±2.57 7.35±1.61 0.083 0.127 0.070 LMCA 7.91±0.42 7.30±0.72 6.95±0.49 0.371 0.042 0.137 LPCA 6.19±1.18 5.58±1.44 4.67±1.24 0.271 0.157 0.210 LACA 5.39±0.75 5.70±0.86 4.88±1.03 0.292 0.380 0.096 Note: The left artery was measured representatively for arteries distributed in pairs. ICC a is used to test the software CVI42 vs MASS. ICC b is used to test the software TS FLOW vs CVI42. ICC c is used to test the software TS FLOW vs MASS. Asterisks indicate ***excellent, **good, and *moderate ICC. No asterisk indicates poor agreement. Additional Declarations Competing interest reported. The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article. Although some co-authors are from Siemens Healthcare Ltd., they contributed mainly by providing WIP (work in progress) sequence, and all authors only have relationships of academic cooperation. Supplementary Files supplementarymaterial.docx Cite Share Download PDF Status: Published Journal Publication published 28 Nov, 2025 Read the published version in BMC Medical Imaging → Version 1 posted Editorial decision: Revision requested 01 Sep, 2025 Reviews received at journal 28 Aug, 2025 Reviewers agreed at journal 28 Aug, 2025 Reviewers agreed at journal 26 Aug, 2025 Reviews received at journal 03 Aug, 2025 Reviewers agreed at journal 28 Jul, 2025 Reviewers invited by journal 28 Jul, 2025 Editor assigned by journal 24 Jul, 2025 Submission checks completed at journal 24 Jul, 2025 First submitted to journal 20 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7170747","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":492900491,"identity":"f9b01de1-8bb5-43c3-83d7-5397c6e4e784","order_by":0,"name":"Mingfang Luo","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Mingfang","middleName":"","lastName":"Luo","suffix":""},{"id":492900492,"identity":"409bbee1-5dfc-4b94-bd72-7f769c706207","order_by":1,"name":"Lini Liu","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Lini","middleName":"","lastName":"Liu","suffix":""},{"id":492900493,"identity":"1218c680-481f-41a5-84d6-4de98b71d795","order_by":2,"name":"Rui Li","email":"","orcid":"","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Rui","middleName":"","lastName":"Li","suffix":""},{"id":492900494,"identity":"88f711cc-5dff-4417-8ffa-68c878f24e63","order_by":3,"name":"Daniel Giese","email":"","orcid":"","institution":"Diagnostic Imaging Magnetic Resonance Cardiovascular， Siemens Healthcare GmbH， Erlangen","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Giese","suffix":""},{"id":492900495,"identity":"c3a1d1fd-8635-467e-b4b0-d27e3ed898a8","order_by":4,"name":"Binbin Sui","email":"","orcid":"","institution":"Beijing Hospital","correspondingAuthor":false,"prefix":"","firstName":"Binbin","middleName":"","lastName":"Sui","suffix":""},{"id":492900496,"identity":"fa4c1388-b1dc-4b88-8cd2-1861e24faf5f","order_by":5,"name":"Yuting Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYBADHiBmfvjBwIaHn7+BKB0GIC1sxhIVaTKSMw4QpwVMSvCcOWxj0JBAQO3xs4df87b9kTHnX8BgINl2nseA4QDjh485eLScyUuz5m0z4LGc8YDhQWHbbR5z5gZmyZnbcGsxO5BjZgzSYnDjAMiW2zyWDQfYmHnxaTn/BqFFgrftHI/BgQQCWm7kGD8GaznfAPL+AcJa7G+8MWOcc84YaAs4kJN5JGccbMbrF8n+HOMPb8rk7A3OHwBFpZ09P3/zwQ8f8WgBAjYpUDwySOR/gAowNuBVDwTMH3+AKP4DhBSOglEwCkbBSAUAtZlRqsaEls0AAAAASUVORK5CYII=","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":true,"prefix":"","firstName":"Yuting","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-07-20 15:53:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7170747/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7170747/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12880-025-02039-8","type":"published","date":"2025-11-28T15:57:37+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":88118704,"identity":"217ce85b-4821-438e-8dbe-8dd4440e53f3","added_by":"auto","created_at":"2025-08-01 15:16:21","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":8671660,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(a) \u003c/strong\u003eInterfaces of software A (TS flow v1.0), \u003cstrong\u003e(b)\u003c/strong\u003e software B: MASS (Medis medical imaging systems, Version 2024-EXP), \u003cstrong\u003e(c)\u003c/strong\u003e software C: CVI42 (v5.9.4, Circle Cardiovascular Imaging, Alberta, Calgary, Canada), and \u003cstrong\u003e(d)\u003c/strong\u003edefinition of cut plane location of intracranial arteries.\u003c/p\u003e","description":"","filename":"FIGURE1.png","url":"https://assets-eu.researchsquare.com/files/rs-7170747/v1/862085c5f2737a59ff8a15c1.png"},{"id":88118701,"identity":"e5c12afb-0532-4cae-8c46-26c813045b41","added_by":"auto","created_at":"2025-08-01 15:16:21","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":10264434,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of intracranial arterials flow rate obtained by three post-processing software. The visualization results of the (a) LICA, (b) RICA, (c) BA, (d) LMCA, (e) RMCA, (f) LPCA, (g) RPCA, (h) LACA and (i) RACA respectively. p \u0026lt; 0.05 was considered statistically significant. LICA = left internal carotid artery; RICA = right internal carotid artery; BA = basilar artery; LMCA = left middle cerebral artery; RMCA = right middle cerebral artery; LPCA = left posterior cerebral artery; RPCA = right posterior cerebral artery; LACA = left anterior cerebral artery; RACA = right anterior cerebral artery.\u003c/p\u003e","description":"","filename":"FIGURE2.png","url":"https://assets-eu.researchsquare.com/files/rs-7170747/v1/fc9261e54d8f6eb414152b2b.png"},{"id":88119614,"identity":"f8bb844c-fd1c-476d-9252-c1c8d3a2cb02","added_by":"auto","created_at":"2025-08-01 15:24:21","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":7944054,"visible":true,"origin":"","legend":"\u003cp\u003eThe comparison results of Vavg of (a) LICA, (b) RICA, (c) BA, (d) LMCA, (e) RMCA, (f) LPCA, (g) RPCA, (h) LACA and (i) RACA obtained by three post-processing software. p \u0026lt; 0.05 was considered statistically significant. Vavg = average velocity; LICA = left internal carotid artery; RICA = right internal carotid artery; BA = basilar artery; LMCA = left middle cerebral artery; RMCA = right middle cerebral artery; LPCA = left posterior cerebral artery; RPCA = right posterior cerebral artery; LACA = left anterior cerebral artery; RACA = right anterior cerebral artery.\u003c/p\u003e","description":"","filename":"FIGURE3.png","url":"https://assets-eu.researchsquare.com/files/rs-7170747/v1/1a3e3771e64fa26a628123d7.png"},{"id":88118700,"identity":"c624b7d9-d99a-47e6-822a-e594505b1ed7","added_by":"auto","created_at":"2025-08-01 15:16:21","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":9019656,"visible":true,"origin":"","legend":"\u003cp\u003eThe comparison results of Vmax of (a) LICA, (b) RICA, (c) BA, (d) LMCA, (e) RMCA, (f) LPCA, (g) RPCA, (h) LACA and (i) RACA obtained by three post-processing software. p \u0026lt; 0.05 was considered statistically significant. Vmax = maximum velocity; LICA = left internal carotid artery; RICA = right internal carotid artery; BA = basilar artery; LMCA = left middle cerebral artery; RMCA = right middle cerebral artery; LPCA = left posterior cerebral artery; RPCA = right posterior cerebral artery; LACA = left anterior cerebral artery; RACA = right anterior cerebral artery.\u003c/p\u003e","description":"","filename":"FIGURE4.png","url":"https://assets-eu.researchsquare.com/files/rs-7170747/v1/22f1306f2c03d01335acda90.png"},{"id":88119616,"identity":"741bfd8f-579f-447c-9feb-e7dbcf018f58","added_by":"auto","created_at":"2025-08-01 15:24:22","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":8955879,"visible":true,"origin":"","legend":"\u003cp\u003eThe results of regions of interest area data of each cerebral vessel obtained by various post-processing software. The comparison results of regions of interest (ROI) area data of (a) LICA, (b) RICA, (c) BA, (d) LMCA, (e) RMCA, (f) LPCA, (g) RPCA, (h) LACA and (i) RACA obtained by three post-processing software. p \u0026lt; 0.05 was considered statistically significant. LICA = left internal carotid artery; RICA = right internal carotid artery; BA = basilar artery; LMCA = left middle cerebral artery; RMCA = right middle cerebral artery; LPCA = left posterior cerebral artery; RPCA = right posterior cerebral artery; LACA = left anterior cerebral artery; RACA = right anterior cerebral artery.\u003c/p\u003e","description":"","filename":"FIGURE5.png","url":"https://assets-eu.researchsquare.com/files/rs-7170747/v1/21de08885a4246d222815b78.png"},{"id":88119615,"identity":"aea20d14-d98f-4b79-95f1-3672604d9fc5","added_by":"auto","created_at":"2025-08-01 15:24:21","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":2733413,"visible":true,"origin":"","legend":"\u003cp\u003eEquivalence testing for (a) flow rate assessment, (b) Vavg and (c) Vmax. Equivalence of measurements of two software is shown if the confidence interval for software comparison (indicated as black lines, squares marked upper and lower limits) are contained within the equivalence limits (mark with a dotted line). T/M= TS FLOW vs MASS; T/C= TS FLOW vs CVI42; C/M= CVI42 vs MASS.\u003c/p\u003e","description":"","filename":"FIGURE6.png","url":"https://assets-eu.researchsquare.com/files/rs-7170747/v1/82370c7e6a29f431bbf40d5b.png"},{"id":97178837,"identity":"14fdf132-b1fc-4eaa-8e36-975a3310e12e","added_by":"auto","created_at":"2025-12-01 16:13:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":37866686,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7170747/v1/5358673a-aa54-45f4-b672-8f3b64841384.pdf"},{"id":88118705,"identity":"3d965bd9-1591-4337-b7e9-687228ebb6a6","added_by":"auto","created_at":"2025-08-01 15:16:22","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1700481,"visible":true,"origin":"","legend":"","description":"","filename":"supplementarymaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-7170747/v1/c3e2655a0002c0c9c9011143.docx"}],"financialInterests":"Competing interest reported. The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article. Although some co-authors are from Siemens Healthcare Ltd., they contributed mainly by providing WIP (work in progress) sequence, and all authors only have relationships of academic cooperation.","formattedTitle":"Repeatability and reproducibility of hemodynamic measurements by intracranial 4D Flow MRI: a multi‑vendor and multi‑software cross‑over comparison study","fulltext":[{"header":"Background","content":"\u003cp\u003eHemodynamics evaluation is crucial for understanding pathophysiology and clinical evaluation of cardiovascular diseases. 4D Flow MR offers the unique opportunity to gather time-resolved, flow-encoding information in all three spatial directions of the heart and blood vessels to non-invasively characterize blood flow without contrast agent [1]. As an extension of 2D Flow MRI, it also has the advantage of allowing retrospective evaluation of the acquired 3D volume without the necessity of pre-planning or repeated scans and is therefore particularly useful to evaluate complex vascular anatomies and various flow patterns [2].\u003c/p\u003e\n\u003cp\u003eDue to the small size and tortuous anatomy of intracranial vessels, obtaining sufficient spatial resolution for cerebral arteries by 4D Flow could be technically challenging, which limited the clinical application. In recent years, with the key advances of various post-processing methods and further acceleration in acquisition, 4D Flow has been increasingly applied to visualize and quantitatively analyze the blood flow in a variety of intracranial vascular diseases including aneurysms, arteriovenous malformation, arterial stenosis, etc\u0026nbsp;[3-5]. Encouraging findings have provided novel insights into disease-related hemodynamic mechanisms, such as the inflow-to-outflow ratio could accurately distinguish between non-hemorrhagic and hemorrhagic arteriovenous malformation [6], and transcranial pulse wave velocity were significantly higher in Alzheimer’s disease compared to age-matched controls [7].\u003c/p\u003e\n\u003cp\u003eHowever, there is a general lack in standardization among various scanners, post-processing software, as well as the evidence-based merits on benefits of clinical applications of 4D Flow [8]. Studies of repeatability and reproducibility of 4D Flow evaluating heart and aorta have been reported, showing that inter- and intrareader consistency for clinically used parameters was better for some software (Caas and MevisFlow) but worse for some other (Circle), and was better for some parameters (stroke volume, peak flow, and forward flow) but worse for some other (peak maximum velocity) [9, 10]. 4D Flow cardiovascular magnetic resonance consensus statement have been published in 2015 and updated in 2023, emphasizing the importance of clinical quality assurance and validation [1].\u003c/p\u003e\n\u003cp\u003eFor the intracranial application of 4D Flow, it is important to recognize and measure the differences among various MRI scanners and post-processing programs, all of which must be considered when making comparisons or designing multi-center trials. A previous study has shown that using the same post-processing software and same vendor for multi-center repetitive assessment of intracranial 4D Flow, good repeatability in the measurement of blood flow and peak velocity were achieved [11]. However, the reproducibility across different scanners and post-processing platforms remains unclear. Although time-resolved RF-spoiled gradient echo sequence was generally applied by multiple vendors, technical discrepancies may exist due to 1) gradient system performance; 2) receiver coil array configurations; 3) selection of acceleration factors (e.g., SENSE vs. GRAPPA); 4) intravoxel phase dispersion caused by turbulent flow, etc [1, 12]. In addition, different post-processing platforms may include eddy current correction or not, employ manual or semi-automatic vessel segmentation, and have different interaction design for vessel area delineation. From a clinical perspective, it is essential to validate the repeatability and reproducibility of intracranial 4D Flow in healthy subjects before applying to cerebral vascular diseases with hemodynamic abnormalities [13, 14].\u003c/p\u003e\n\u003cp\u003eTherefore, this study aims to perform a comprehensive cross-comparison analysis of key hemodynamic measurements by intracranial 4D Flow in healthy subjects using 3.0T scanners (i) from three vendors and (ii) by three post-processing software packages. Evaluation of inter- and Intra-observer consistency was also performed.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eSubjects\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis cross-sectional study was conducted in accordance with approval from the Ethics Committee of Sichuan Provincial People's Hospital (No. 2023-409-1). Between July and August 2024, we prospectively recruited 20 healthy adults meeting stringent inclusion criteria: absence of cardiovascular/cerebrovascular history, major risk factors (smoking, diabetes, hypertension), and confirmed normal cerebrovascular anatomy through baseline imaging.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImage acquisition\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll subjects were examined on three 3.0T MRI scanners: MRI1 (MAGNETOM VIDA,\u0026nbsp;Siemens Healthineers), MRI2 (SIGNA Architect, GE Healthcare) and MRI3 (uMR 880, United Imaging Healthcare), with a 32-channel head coil. Prior to each scan, the subjects would rest in a resting state for 15-20 minutes to ensure the conditions (such as blood pressure) were close to similar baseline level. The 4D Flow scanning parameters of the three scanners has been standardized as follows: slice thickness is interpolated to 1 mm using zero-padding interpolation, spatial resolution after reconstruction was 1×1×1 mm³, velocity encoding in all three directions was set to 100 cm/s [15], and 48 axial slices were acquired. The intracranial Circle of Willis was covered using 3D TOF-MRA sequence and pulse gating was used for 4D Flow imaging. Acquisition parameters across the three scanners have been carefully adjusted to closely match each other. Detailed parameters were provided in Table 1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImage post-processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe generated three datasets (20 subjects each) were imported into three commercial post-processing software programs: CVI42 (Circle Cardiovascular Imaging), MASS (Version 2024-EXP, Leiden University Medical Center, Leiden, The Netherlands), TS FLOW (Beijing Qingying Huakang Technology Co., LTD.). Key parameters such as flow rate, Vavg, Vmax, and vessel area of the delineated was extracted. Two operators with over five years of experience in clinical practice of radiology independently conducted evaluations using each software. The workflow encompasses conducting background phase offsets correction, anti-aliasing, segmentation, visualization and quantification. Additionally, both TS FLOW software and CVI42 permitted manual adjustment of the segmentation threshold, but TS FLOW and MASS did not support manual correction for background phase offsets and anti-aliasing (Supplementary Materials 1).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNine intracranial arteries were measured, including the including the internal carotid artery (ICA) at the C3-C4 segment, the anterior cerebral artery (ACA) at the A2 segment, the middle cerebral artery (MCA) at the M1 segment, the posterior cerebral artery (PCA) at the P1 segment, and the basilar artery (BA) before the inferior cerebellar artery [16]. To ensure optimal plane selection, each vessel was positioned centrally within its respective cross-section. Figure 1 illustrates the data processing interfaces of the three software programs and the display of the cerebral artery measurement cross-sections. It is important to note that when delineated vessel area, TS FLOW software employs semi-automatic delineation, MASS utilizes manual registration for customized delineation, and CVI42 implements automated registration-based customized delineation (Supplementary Materials 2).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDepending on the data distribution (normal or non-normal), paired t-tests, Wilcoxon signed-rank tests, or Friedman tests were employed to compare the flow rate differences among the three scanners. The flow rate values for intracranial arteries were presented as mean ± SD.\u0026nbsp;Software across comparisons were performed using the Kruskal-Wallis (K-W) test, followed by pairwise comparisons. \u003cem\u003ep\u0026nbsp;\u003c/em\u003e\u0026lt; 0.05 were considered for statistical significance. Additionally, Bland-Altman analysis and ICC were used to evaluate variability and equivalence between software. The interpretation of ICC was as follows: \u0026gt;0.9: excellent; 0.75-0.9: good; 0.5-0.75: moderate; \u0026lt;0.5: poor [17].\u003c/p\u003e\n\u003cp\u003eIn the equivalence test, the equivalence limits were determined by the maximum standard deviation obtained from the analysis of the intra-observer variability (Supplementary Materials 3), which was ±1.96 standard deviation (SD). That was, if the deviation between two software systems falls within the 95% range of the deviation produced by a single observer using the same software for repeated measurements, then these two software systems were considered equivalent [18]. All statistical analyses were conducted using SPSS version 22.0 (IBM Corporation).\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eSubject characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study included 20 healthy subjects (mean age 38.7±16.88 years, 10 female), none of whom had a history of cardiovascular or cerebrovascular diseases or other significant medical conditions. Baseline characteristics are summarized in Table 2.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVariation across three scanners\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor scanners comparison, as shown in\u0026nbsp;Supplementary Materials\u0026nbsp;4, there were no statistically significant differences in the flow rate measurements of any intracranial arteries obtained from 20 subjects across three scanners using any post-processing software (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e\u0026gt; 0.05). The results demonstrated moderate consistency (ICC = 0.700-0.919).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVariation across\u0026nbsp;three post-processing software\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e\u0026nbsp;Flow rate\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTS FLOW demonstrated significantly higher flow rates across all arterial segments compared to MASS and CVI42 (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05), with the difference ranging from 18.8% to 49.8%. On the other hand, MASS and CVI42 showed no inter-software differences (\u003cem\u003ep\u003c/em\u003e \u0026gt; 0.05; Figure 2). The Bland-Altman analysis revealed that the difference range between MASS and CVI42 was the narrowest (e.g., BA LOA: -42.4 to 23.0 mL/min) except in the LACA (Supplementary Materials 5). The consistency between MASS and CVI42 was moderate to excellent (ICC: 0.541-0.765; Table 3), with highest consistency in the LMCA (ICC = 0.765).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eVavg\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs shown in Figure 3, TS FLOW measured significantly higher Vavg than CVI42 in the RICA (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= 0.013) and LPCA (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= 0.034), with no inter-software differences in other arteries (\u003cem\u003ep\u003c/em\u003e \u0026gt; 0.05). In contrast, MASS exhibited systematically lower Vavg compared to both TS FLOW and CVI42 across all arteries (all\u003cem\u003e\u0026nbsp;p\u003c/em\u003e \u0026lt; 0.05). Inter-software agreement analysis revealed moderate-to-high consistency between TS FLOW and CVI42 (ICC: 0.533-0.891), peaking in the LMCA (ICC = 0.891), whereas MASS showed poor concordance with CVI42 (ICC \u0026lt; 0.5; Table 3). Bland-Altman analysis (Supplementary Materials 6) confirmed superior measurement consistency between TS FLOW and CVI42, with narrowest limits of agreement in the LMCA (−4.99 to 5.84 cm/s) and widest in the LACA (−7.22 to 9.14 cm/s).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eVmax\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Vmax measured by TS FLOW was consistently higher than that measured by MASS and CVI42 in all arteries (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05), except for the LACA (\u003cem\u003ep\u003c/em\u003e = 1.000).\u0026nbsp;The maximal discrepancy in the LPCA (TS FLOW \u0026gt; MASS by 71.7%) and minimal in the RICA (TS FLOW \u0026gt; CVI42 by 7.8%).\u0026nbsp;Notably, MASS measured significantly lower Vmax than CVI42 in the RACA (\u003cem\u003ep\u003c/em\u003e = 0.013; Figure 4). Inter-software consistency analysis (Table 3) revealed moderate MASS-CVI42 concordance across intracranial arteries (ICC: 0.510-0.612), whereas poor agreement was shown for TS FLOW-MASS (ICC: 0.105-0.332). TS FLOW-CVI42 comparisons exhibited suboptimal consistency (ICC range: 0.098-0.312) except in the LMCA (ICC = 0.503). Bland-Altman analysis (Supplementary Materials 7) demonstrated superior measurement consistency between MASS and CVI42 with narrowest LOA in the LACA (−9.68 to 19.81 cm/s) and widest in the LPCA (−15.46 to 23.85 cm/s).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eDelineated Vessel\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cem\u003eArea\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe vessel area of TS flow on ICA and BA was larger than that of the other two software (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; Figure 5). In the consistency test, only MASS and CVI42 showed moderate consistency in the vessel area of the LICA (ICC = 0.557; Table 3).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIntra- observer Variability as measure of repeatability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the intra-observer variability analysis, both the TS FLOW software and CVI42 demonstrated moderate to excellent consistency (ICC \u0026gt; 0.5) in terms of flow rate, Vavg, Vmax, and vessel area. The MASS software also showed moderate or excellent consistency (ICC \u0026gt; 0.5) in flow rate, Vavg and Vmax, but for 6 out of 9 arteries, the ICC was less than 0.5 in vessel area (Supplementary Materials 8). For CVI42 software, the ICC of flow rate exhibited a significant positive correlation with the ICC of vessel area (\u003cem\u003ep\u003c/em\u003e = 0.047, r = 0.674, Supplementary Materials 9).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInter-observer Consistency as measure of reproducibility\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the inter-observer variability analysis, TS FLOW demonstrated moderate to excellent consistency (ICC = 0.590-0.981) in flow rate, Vavg, Vmax, and vessel area. In contrast, MASS exhibited poor interobserver agreement for flow rate in the posterior and anterior cerebral arteries (RPCA/LACA ICC = 0.317/0.179), suboptimal consistency for Vavg (LPCA/RPCA/LACA/RACA ICC = 0.249-0.480), and poor consistency for vessel area measurements (ICC = 0.107-0.650). CVI42 achieved moderate interobserver agreement for Vavg and Vmax (ICC = 0.530-0.968) across all vessels but showed heterogeneous ICC distributions (0.21–0.83) for flow rate and vessel area without clear patterns (Supplementary Materials 10). For CVI42 software, the ICC of flow rate exhibited a significant positive correlation with the ICC of vessel area (\u003cem\u003ep\u003c/em\u003e = 0.029, r = 0.719, Supplementary Materials 11).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEquivalence test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs for the flow rate, it was observed that MASS and CVI42 were all within the equivalent range (except for LACA). As for Vavg, TS FLOW and CVI42 were within the equivalent range in each\u0026nbsp;intracranial\u0026nbsp;artery. As for Vmax, the differences among the three software programs were all within the equivalent range (Figure 6).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study comprehensively evaluated the across-vendor and across-software repeatability and reproducibility of multiple hemodynamic measurements of major arteries by intracranial 4D Flow MRI in healthy subjects. Possible confounders were carefully controlled. Subjects were scanned three times on the same day under resting-state baseline conditions. Key parameters of acquisition were similar across vendors, post-processing pipelines were standardized, and variances were quantified. The results revealed no difference in flow rate measurement across different mainstream vendors. The consistency across three post-processing software, however, seems less satisfactory. The flow rate obtained by TS FLOW was higher than that by the other two software programs (18.81% to 49.8% higher). On the other hand, TS FLOW, which employs semi-automatic delineation, exhibited higher reproducibility and repeatability. This study provides a reference for the quantification of intracranial 4D Flow data in various cerebrovascular pathological conditions and also for the acquisition and analysis of future multi-center studies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFocus on scanners\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn clinical practice, the 4D Flow magnetic resonance sequences provided by different manufacturers exhibit varying degrees of inconsistency\u0026nbsp;[19].\u0026nbsp;Therefore, in this study, we endeavored to achieve matching for as many key parameters as possible, including flow velocity encoding, acceleration factor, spatial resolution, temporal resolution, and reconstruction phase. The results demonstrate that, in the most stable blood flow rate metric, there is good consistency across manufacturers. Wen et al. evaluated the performance of neurovascular 4D Flow MRI by assessing multiple healthy volunteers using the same scanner model at three distinct locations. Their findings indicate that hemodynamic parameters, such as blood flow rate and peak velocity, exhibit excellent multi-center repeatability, test-retest reliability, and inter-observer agreement\u0026nbsp;[11]. These results basically aligned with ours, suggesting good consistency when using essentially the same technical parameters of acquisition.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFocus on post-\u003c/strong\u003e\u003cstrong\u003eprocessing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study found less satisfactory consistency of measurements by different software. The quantitative values of TS FLOW for flow rate and delineated area seemed higher than those of the other two software. We speculate that the differences in flow rate and Vmax may be associated with variations in delineated vessel area. Notably, the three post-processing software packages differ in their ROI delineation methods. TS FLOW employs semi-automatic delineation, MASS uses manual registration with customizable delineation and contour correction, and CVI42 utilizes automatic registration delineation with manual contour correction (Supplementary Materials 2). Furthermore, Vmax is typically determined by the peak value in included voxels, whereas the flow rate is calculated as the sum of velocities across all voxels within the ROI at each time phase, multiplied by the region's area [20]. However, since the semi-automatic segmentation delineation method cannot manually correct the vascular contour of each phase, this might be one of the factors leading to an overestimation of blood flow. This may suggest why the flow rate and Vmax values obtained by TS FLOW software tend to be larger in intracranial vessels. In addition, the differences among various software packages could also be attributable to disparities in post-processing algorithms, such as background phase correction, interpolation techniques and\u0026nbsp;the partial volume effect [21-23]. Therefore, in future multi-center studies, the above-mentioned influencing factors should be fully considered when using different post-processing software to ensure the consistency and comparability of the results.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFocus on equivalence test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFurthermore, the results of the equivalence test indicated that within the pre-defined clinically acceptable difference range, despite the technical differences among the three software, they still demonstrated clinical interchangeability in terms of hemodynamic core parameters, especially the Vmax. This finding provides empirical evidence for the use of different analysis platforms in multi-center studies. However, it should be noted that due to the complexity of intracranial arterial anatomy and the current limitations of 4D flow MRI sequence resolution, local measurement differences may occur in intracranial stenosis or small vessels (such as LACA). In such cases, rigorous manual verification during the post-processing stage is necessary to ensure data accuracy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInter- and Intra-observer Variability\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the CVI42 software, the consistency of flow rate was poor and no clear pattern is observed. It is worth noting that the intra- and inter- observer ICC of flow rate and that of delineated vessel area were positively correlated, indicating that the repeatability and reproducibility of flow rate measurement by CVI42 may mainly depend on the consistency of vessel area delineation. The reason for this poor consistency may be attributed to the subjectivity inherent in manual segmentation during ROI delineation. Oechtering et al. also confirmed that in the validation of Cardiovascular 4D Flow, CVI42 exhibited significant variation in contour delineation [9]. It is assumed that the measurements could be more influenced by manual delineation by different observers when using MASS and CVI42. Therefore, it is recommended that the selected planes and the size of the ROI should be carefully determined at each stage of blood flow measurement, especially when using the manual delineation method.\u003c/p\u003e\n\u003cp\u003eA previous study has compared the hemodynamic measurements obtained using a fully automated post-processing method based on deep learning with that measured by human observers, showing that this method can effectively reduce the post-processing time for 4D Flow data in the ventricles while minimizing inter-observer differences [24]. However, customizable delineation and contour correction could allow for analysis of more specified target vessels in the setting of various cerebrovascular pathologies (such as bridge artery after bypass surgery), which would require further investigation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLimitations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThere are several limitations to our study. Firstly, due to the lack of a golden standard for intracranial flow measurements, it is not feasible to determine which software provides more accurate parameter values. Second, this study quantified the differences among the software, but the source of difference cannot be fully determined. Moreover, this is a single center study without validation of test-retest consistency at external centers. Finally, since the three software programs share only a limited set of quantitative values, other more advanced hemodynamic parameters, such as wall shear stress and pressure gradient, were not included into analysis in this study.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eIn conclusion, our study reveals that in the 4D Flow analysis of intracranial arteries, scanning by different vendors appear to result in smaller variations in hemodynamic measurements compared to using different post-processing software methods. Software that employs semi-automatic delineation exhibited general better repeatability and reproducibility. This could serve as a reference for the acquisition and analysis of future multi-center studies.\u0026nbsp;\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e\u003cstrong\u003e4D Flow MR:\u0026nbsp;\u003c/strong\u003eFour-dimensional flow magnetic resonance\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eICA:\u0026nbsp;\u003c/strong\u003eInternal carotid artery\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eACA:\u0026nbsp;\u003c/strong\u003eAnterior cerebral artery\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMCA:\u003c/strong\u003e Middle cerebral artery\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePCA:\u003c/strong\u003e Posterior cerebral artery\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBA:\u0026nbsp;\u003c/strong\u003eBasilar artery\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVavg:\u0026nbsp;\u003c/strong\u003eAverage velocity\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVmax:\u003c/strong\u003e Maximal velocity\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eROI:\u003c/strong\u003e Region of interest\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSD:\u0026nbsp;\u003c/strong\u003eStandard deviation\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eICC:\u003c/strong\u003e Intraclass correlation coefficients\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eEthics approval and consent to participate\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis cross-sectional study was conducted in accordance with approval from the Ethics Committee of Sichuan Provincial People\u0026apos;s Hospital (No. 2023-409-1). All participants have provided written informed consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eConsent for publication\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAvailability of data and materials\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe patient data and images supporting this study are not publicly available due to patient confidentiality restrictions under the ethical approval by the Ethics Committee of Sichuan Provincial People\u0026apos;s Hospital (No. 2023-409-1). De-identified data may be made available from the corresponding author upon reasonable request, subject to institutional ethical approval and execution of a data access agreement.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCompeting interests\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article. Although some co-authors are from Siemens Healthcare Ltd., they contributed mainly by providing WIP (work in progress) sequence, and all authors only have relationships of academic cooperation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eFunding\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study has received funding by National Natural Science Foundation (Youth Project), grant No. 82202123.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAuthor information\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMingfang Luo and Lini Liu contributed equally to this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAuthors\u0026apos; contributions\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eM.L designed the study, analyzed and interpreted the data, and contributor in writing the manuscript. L. L analyzed and interpreted the data and contributed to write the manuscript. R.L conceptualized the study and interpreted the data. D.G contributed mainly by providing WIP (work in progress) sequence. B.S supervised the study and revised the manuscript. \u0026nbsp;Y.W designed, supervised, provided funding and contributor in writing the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAcknowledgements\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors thank all the volunteers participating, and thank Meining Chen from MR Research Collaboration, Siemens Healthineers Ltd., Liqiang Zhou from the United Imaging Healthcare and Ruzhi Zhang from GE Healthcare for their professional technical support during the implementation of flow imaging sequences.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBissell M M, Raimondi F, Ait Ali L, et al. 4D Flow cardiovascular magnetic resonance consensus statement: 2023 update. J Cardiovasc Magn Reson. 2023;25:40.\u003c/li\u003e\n\u003cli\u003eRizk J. 4D flow MRI applications in congenital heart disease. Eur Radiol. 2021;31:1160-1174.\u003c/li\u003e\n\u003cli\u003eAnsari S A, Schnell S, Carroll T, et al. Intracranial 4D flow MRI: toward individualized assessment of arteriovenous malformation hemodynamics and treatment-induced changes. AJNR Am J Neuroradiol. 2013;34:1922-8.\u003c/li\u003e\n\u003cli\u003eMorgan A G, Thrippleton M J, Wardlaw J M, et al. 4D flow MRI for non-invasive measurement of blood flow in the brain: A systematic review. J Cereb Blood Flow Metab. 2021;41:206-218.\u003c/li\u003e\n\u003cli\u003ePeng F, Xia J, Zhang F, et al. Intracranial aneurysm instability prediction model based on 4D-Flow MRI and HR-MRI. Neurotherapeutics. 2025;22:e00505.\u003c/li\u003e\n\u003cli\u003eTakeda Y, Kin T, Sekine T, et al. Hemodynamic Analysis of Cerebral AVMs with 3D Phase-Contrast MR Imaging. AJNR Am J Neuroradiol. 2021;42:2138-2145.\u003c/li\u003e\n\u003cli\u003eRivera-Rivera L A, Cody K A, Eisenmenger L, et al. Assessment of vascular stiffness in the internal carotid artery proximal to the carotid canal in Alzheimer\u0026apos;s disease using pulse wave velocity from low rank reconstructed 4D flow MRI. J Cereb Blood Flow Metab. 2021;41:298-311.\u003c/li\u003e\n\u003cli\u003eBoban M. Editorial for \u0026quot;Traveling Volunteers - A Multi-Vendor, Multi-Center Study on Reproducibility and Comparability of 4D Flow Derived Aortic Hemodynamics in Cardiovascular Magnetic Resonance\u0026quot;. J Magn Reson Imaging. 2022;55:223-224.\u003c/li\u003e\n\u003cli\u003eOechtering T H, Nowak A, Sieren M M, et al. Repeatability and reproducibility of various 4D Flow MRI postprocessing software programs in a multi-software and multi-vendor cross-over comparison study. J Cardiovasc Magn Reson. 2023;25:22.\u003c/li\u003e\n\u003cli\u003eDemir A, Wiesemann S, Erley J, et al. Traveling Volunteers: A Multi-Vendor, Multi-Center Study on Reproducibility and Comparability of 4D Flow Derived Aortic Hemodynamics in Cardiovascular Magnetic Resonance. J Magn Reson Imaging. 2022;55:211-222.\u003c/li\u003e\n\u003cli\u003eWen B, Tian S, Cheng J, et al. Test-retest multisite reproducibility of neurovascular 4D flow MRI. J Magn Reson Imaging. 2019;49:1543-1552.\u003c/li\u003e\n\u003cli\u003eSoulat G, McCarthy P, Markl M. 4D Flow with MRI. Annu Rev Biomed Eng. 2020;22:103-126.\u003c/li\u003e\n\u003cli\u003eDyverfeldt P, Bissell M, Barker A J, et al. 4D flow cardiovascular magnetic resonance consensus statement. J Cardiovasc Magn Reson. 2015;17:72.\u003c/li\u003e\n\u003cli\u003eBurkhardt B E U, Kellenberger C J, Callaghan F M, et al. Flow evaluation software for four-dimensional flow MRI: a reliability and validation study. Radiol Med. 2023;128:1225-1235.\u003c/li\u003e\n\u003cli\u003eWu C, Schnell S, Vakil P, et al. In Vivo Assessment of the Impact of Regional Intracranial Atherosclerotic Lesions on Brain Arterial 3D Hemodynamics. AJNR Am J Neuroradiol. 2017;38:515-522.\u003c/li\u003e\n\u003cli\u003eZarrinkoob L, W\u0026aring;hlin A, Ambarki K, et al. Quantification and mapping of cerebral hemodynamics before and after carotid endarterectomy, using four-dimensional flow magnetic resonance imaging. J Vasc Surg. 2021;74:910-920.e1.\u003c/li\u003e\n\u003cli\u003eKoo T K, Li M Y. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med. 2016;15:155-63.\u003c/li\u003e\n\u003cli\u003eZange L, Muehlberg F, Blaszczyk E, et al. Quantification in cardiovascular magnetic resonance: agreement of software from three different vendors on assessment of left ventricular function, 2D flow and parametric mapping. J Cardiovasc Magn Reson. 2019;21:12.\u003c/li\u003e\n\u003cli\u003eXie L, Zhang Y, Hong H, et al. Higher intracranial arterial pulsatility is associated with presumed imaging markers of the glymphatic system: An explorative study. Neuroimage. 2024;288:120524.\u003c/li\u003e\n\u003cli\u003eGatehouse P D, Keegan J, Crowe L A, et al. Applications of phase-contrast flow and velocity imaging in cardiovascular MRI. Eur Radiol. 2005;15:2172-84.\u003c/li\u003e\n\u003cli\u003eBock J, Frydrychowicz A, Stalder A F, et al. 4D phase contrast MRI at 3 T: effect of standard and blood-pool contrast agents on SNR, PC-MRA, and blood flow visualization. Magn Reson Med. 2010;63:330-8.\u003c/li\u003e\n\u003cli\u003eStalder A F, Russe M F, Frydrychowicz A, et al. Quantitative 2D and 3D phase contrast MRI: optimized analysis of blood flow and vessel wall parameters. Magn Reson Med. 2008;60:1218-31.\u003c/li\u003e\n\u003cli\u003ePetersson S, Dyverfeldt P, Ebbers T. Assessment of the accuracy of MRI wall shear stress estimation using numerical simulations. J Magn Reson Imaging. 2012;36:128-38.\u003c/li\u003e\n\u003cli\u003eCorrado P A, Wentland A L, Starekova J, et al. Fully automated intracardiac 4D flow MRI post-processing using deep learning for biventricular segmentation. Eur Radiol. 2022;32:5669-5678.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1.\u0026nbsp;\u003c/strong\u003eScan parameters of three scanners.\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"633\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSiemens VIDA\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGE SIGNA Architect\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUnited imaging uMR 880\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eField of view (mm\u003csup\u003e3\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e180*168*48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e180*180*48\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e180*165*48\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eSlice thickness (mm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e1\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e1\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eSpatial resolution (mm\u003csup\u003e3\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e1.36*1.02*1.24\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e1.0*1.0*1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e1.02*1.02*1.26\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eReconstructed spatial resolution (mm\u003csup\u003e3\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e1.02*1.02*1.0\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e1.0*1.0*1.0\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e1.02*1.02*1.0\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eNumber of heart phases\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e9-15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e9-15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e9-15\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eNumber of reconstructed heart phases\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003ePulse synchronization\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003eRetrospective\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003eRetrospective\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003eRetrospective\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eFlip Angle\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eVelocity‐encoding factor\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eAcceleration mode\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003eCS\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003ehyperkat\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003eACS\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eAcceleration factor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e8.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eTE (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e3.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003eminimum\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e3.73\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eTR (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e52.2\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e23.2\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e28\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 35.861%;\"\u003e\n \u003cp\u003eScan time (min)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.8531%;\"\u003e\n \u003cp\u003e4-7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.4313%;\"\u003e\n \u003cp\u003e4-7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 23.8547%;\"\u003e\n \u003cp\u003e6-13\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: TE = echo time, TR = repetition time, CS = Compressed Sensing, hyperkat = Hyper-accelerated k-t Acquisition Technique, and ACS = AI-assisted Compressed Sensing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2.\u0026nbsp;\u003c/strong\u003eDemographics of included subjects\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"365\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43.9891%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56.0109%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHealthy Subjects (N=20)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43.9891%;\"\u003e\n \u003cp\u003eAge (years)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56.0109%;\"\u003e\n \u003cp\u003e38.7 \u0026plusmn; 16.88 (22-72 years)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43.9891%;\"\u003e\n \u003cp\u003eGender radio (male: female)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56.0109%;\"\u003e\n \u003cp\u003e10:10\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43.9891%;\"\u003e\n \u003cp\u003eHeart rate (bpm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56.0109%;\"\u003e\n \u003cp\u003e67\u0026plusmn;9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43.9891%;\"\u003e\n \u003cp\u003eWeight (kg)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56.0109%;\"\u003e\n \u003cp\u003e61.35\u0026plusmn;9.13\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43.9891%;\"\u003e\n \u003cp\u003eHeight (cm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56.0109%;\"\u003e\n \u003cp\u003e167.1\u0026plusmn;7.48\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43.9891%;\"\u003e\n \u003cp\u003eBody mass index (m\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56.0109%;\"\u003e\n \u003cp\u003e21.89\u0026plusmn;2.28\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3.\u0026nbsp;\u003c/strong\u003eVariation across three software\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"626\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFlow rate ml/min\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTS FLOW\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMASS\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCVI42\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eICC\u003csup\u003ea\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eICC\u003csup\u003eb\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eICC\u003csup\u003ec\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLICA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e327.96\u0026plusmn;62.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13.12%;\"\u003e\n \u003cp\u003e234.51\u0026plusmn;39.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16.48%;\"\u003e\n \u003cp\u003e237.93\u0026plusmn;34.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.716*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.275\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.299\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eBA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e202.74\u0026plusmn;33.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13.12%;\"\u003e\n \u003cp\u003e135.34\u0026plusmn;21.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16.48%;\"\u003e\n \u003cp\u003e144.94\u0026plusmn;20.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.635*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.225\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.172\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLMCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e168.69\u0026plusmn;33.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13.12%;\"\u003e\n \u003cp\u003e135.09\u0026plusmn;26.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16.48%;\"\u003e\n \u003cp\u003e141.98\u0026plusmn;31.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.765**\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.546*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.410\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLPCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e87.53\u0026plusmn;14.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13.12%;\"\u003e\n \u003cp\u003e60.50\u0026plusmn;10.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16.48%;\"\u003e\n \u003cp\u003e65.32\u0026plusmn;15.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.541*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.260\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.139\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLACA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e79.44\u0026plusmn;12.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e65.39\u0026plusmn;14.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e63.49\u0026plusmn;19.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.546*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.390\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.244\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVavg cm/s\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLICA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e28.40\u0026plusmn;6.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e23.69\u0026plusmn;5.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e31.36\u0026plusmn;8.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.295\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.533*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.537*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eBA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e30.79\u0026plusmn;4.20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e25.32\u0026plusmn;6.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e33.67\u0026plusmn;5.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.233\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.552*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.272\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLMCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e38.02\u0026plusmn;6.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e32.06\u0026plusmn;7.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e37.60\u0026plusmn;5.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.472\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.891**\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.403\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLPCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e27.84\u0026plusmn;3.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e18.89\u0026plusmn;4.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e25.56\u0026plusmn;3.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.257\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.553*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.163\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLACA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e26.72\u0026plusmn;4.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e19.19\u0026plusmn;4.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e25.75\u0026plusmn;4.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.110\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.608*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.269\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVmax cm/s\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLICA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e83.12\u0026plusmn;20.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e53.27\u0026plusmn;16.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e62.30\u0026plusmn;11.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.532*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.108\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.176\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eBA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e78.18\u0026plusmn;10.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e67.64\u0026plusmn;9.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e71.45\u0026plusmn;8.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.510*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.150\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.330\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLMCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e87.28\u0026plusmn;11.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e71.27\u0026plusmn;10.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e75.84\u0026plusmn;11.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.537*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.503*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.332\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLPCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e88.65\u0026plusmn;20.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e51.63\u0026plusmn;11.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e55.82\u0026plusmn;9.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.511*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.098\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.105\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLACA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e55.74\u0026plusmn;10.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e46.63\u0026plusmn;8.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e51.69\u0026plusmn;10.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.612*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.312\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.277\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003e\u003cstrong\u003earea mm\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLICA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e23.38\u0026plusmn;4.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e16.98\u0026plusmn;3.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e15.84\u0026plusmn;3.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.557*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.194\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.287\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eBA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e12.94\u0026plusmn;2.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e9.30\u0026plusmn;2.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e7.35\u0026plusmn;1.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.083\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.127\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.070\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLMCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e7.91\u0026plusmn;0.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e7.30\u0026plusmn;0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e6.95\u0026plusmn;0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.371\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.042\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.137\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLPCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e6.19\u0026plusmn;1.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e5.58\u0026plusmn;1.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e4.67\u0026plusmn;1.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.271\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.157\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.210\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 18.08%;\"\u003e\n \u003cp\u003eLACA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.96%;\"\u003e\n \u003cp\u003e5.39\u0026plusmn;0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.12%;\"\u003e\n \u003cp\u003e5.70\u0026plusmn;0.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.48%;\"\u003e\n \u003cp\u003e4.88\u0026plusmn;1.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.48%;\"\u003e\n \u003cp\u003e0.292\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9.28%;\"\u003e\n \u003cp\u003e0.380\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.6%;\"\u003e\n \u003cp\u003e0.096\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: The left artery was measured representatively for arteries distributed in pairs. ICC\u003csup\u003ea\u003c/sup\u003e is used to test the software CVI42 vs MASS. ICC\u003csup\u003eb\u003c/sup\u003e is used to test the software TS FLOW vs CVI42. ICC\u003csup\u003ec\u0026nbsp;\u003c/sup\u003eis used to test the software TS FLOW vs MASS. Asterisks indicate ***excellent, **good, and *moderate ICC. No asterisk indicates poor agreement.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-imaging","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmim","sideBox":"Learn more about [BMC Medical Imaging](http://bmcmedimaging.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bmim/default.aspx","title":"BMC Medical Imaging","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"4D Flow, Phase‐contrast magnetic resonance imaging, Intracranial arteries, Hemodynamics","lastPublishedDoi":"10.21203/rs.3.rs-7170747/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7170747/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eMultiple protocols for acquiring intracranial four-dimensional flow magnetic resonance (4D Flow MR) data and post-processing software employed to obtain hemodynamic parameters were available. However, insufficient validation exists to quantify the difference in measurements.\u003cstrong\u003e \u003c/strong\u003eThis study aims to perform a comprehensive cross-comparison analysis of quantitative results from three major vendors and three post-processing software platforms.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eIntracranial 4D Flow MR was conducted on 20 healthy volunteers using three different 3.0T MRI scanners. For each subject, nine major intracranial arteries (a total of 180 arteries) were analyzed using three post-processing software platforms: CVI42, MASS, and TS FLOW. Key hemodynamic parameters—including flow rate, maximum velocity (Vmax), average velocity (Vavg), and delineated vessel area—were extracted. The study compared results across different vendors and software, and further assessed both intra-observer and inter-observer consistency.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eRegardless of the post-processing software used, no significant differences were observed in the flow rate across different scanners (p \u0026gt; 0.05), with ICC = 0.700-0.919. For flow rate and Vmax, the MASS and CVI42 software were equivalent. For Vavg, TS FLOW and CVI42 were tested equivalent. For vessel area, various agreement was shown across software (ICC = 0.083-0.577). Additionally, the TS FLOW software exhibited repeatability and reproducibility (intra-observer ICC = 0.724-0.998; inter-observer ICC = 0.590-0.981).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eScanning protocols by different vendors appear to result in smaller variations in hemodynamic measurements compared to using different post-processing software methods. It is recommended to employ single post-processing software or conduct software calibration when using multiple software platforms to ensure consistency in multi-center clinical studies.\u003c/p\u003e","manuscriptTitle":"Repeatability and reproducibility of hemodynamic measurements by intracranial 4D Flow MRI: a multi‑vendor and multi‑software cross‑over comparison study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-01 15:16:17","doi":"10.21203/rs.3.rs-7170747/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-09-01T05:32:20+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-28T11:41:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"329969179517788412615432303129901709380","date":"2025-08-28T11:35:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"218024327424060555205458806517782159961","date":"2025-08-26T11:22:52+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-03T12:35:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"92870573819734111690914400926962913053","date":"2025-07-29T00:16:35+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-28T18:31:25+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-24T06:58:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-24T06:55:25+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Medical Imaging","date":"2025-07-20T15:43:09+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-imaging","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmim","sideBox":"Learn more about [BMC Medical Imaging](http://bmcmedimaging.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bmim/default.aspx","title":"BMC Medical Imaging","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c38b860e-e86d-4b70-b8ed-f6749d3c04a2","owner":[],"postedDate":"August 1st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-01T16:06:09+00:00","versionOfRecord":{"articleIdentity":"rs-7170747","link":"https://doi.org/10.1186/s12880-025-02039-8","journal":{"identity":"bmc-medical-imaging","isVorOnly":false,"title":"BMC Medical Imaging"},"publishedOn":"2025-11-28 15:57:37","publishedOnDateReadable":"November 28th, 2025"},"versionCreatedAt":"2025-08-01 15:16:17","video":"","vorDoi":"10.1186/s12880-025-02039-8","vorDoiUrl":"https://doi.org/10.1186/s12880-025-02039-8","workflowStages":[]},"version":"v1","identity":"rs-7170747","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7170747","identity":"rs-7170747","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}