Accurate preoperative planning for abdominal aortic aneurysm using fully automated measurement software: A randomized controlled trial

Accurate preoperative planning for abdominal aortic aneurysm using fully automated measurement software: A randomized controlled trial | 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 Article Accurate preoperative planning for abdominal aortic aneurysm using fully automated measurement software: A randomized controlled trial Linlin Guo, Xiaoyu Qi, Ming Yang, Fei Cai, Peng Zhou, Gezheng Chen, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4857239/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Sizing, the first step of endovascular aneurysm repair (EVAR), is essential for a successful procedure. This study evaluated the precision and reproducibility of EVAR sizing facilitated by a novel fully automated software, DetecMicro, in comparison to conventional manual and semi-automatic software. A total of 18 surgeons, consisting of 9 junior residents and 9 vascular surgery specialists, participated in a prospective single-center randomized controlled trial with three parallel arms, stratified based on clinical experience. Each surgeon conducted three repeated measurements for 450 parameters (360 diameter and 90 length parameters). Intra- and inter-observer variability were analyzed using the intraclass correlation coefficient (ICC). Subsequently, the stent size based on the measured results was assessed to determine the impact of measurement errors on stent selection. The reliability of virtual stent implantation (VSI) using DetecMicro was evaluated by comparing it with postoperative models. Compared to PACS and 3mensio, the DetecMicro group exhibited superior accuracy, with 90.39% of diameter measurements and 97.60% of length measurements falling within clinically acceptable ranges, [-2 mm, + 2 mm] and [-5 mm, + 5 mm], respectively. Intra-observer and inter-observer repeatability with DetecMicro demonstrated efficacy, with a mean ICC exceeding 0.9. In the DetecMicro group, clinical experience had a negligible impact on the aforementioned results. VSI, when compared with actual postoperative models, limited errors to within 2 mm. The integration of DetecMicro's measurement and VSI functions holds promise as a reliable tool for preoperative planning in EVAR treatment. Health sciences/Diseases Health sciences/Medical research Physical sciences/Engineering Abdominal aortic aneurysm Fully automatic measurement software Hemodynamics Preoperative planning Virtual stent Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Abdominal aortic aneurysm (AAA) is characterized by the dilation of the three vascular layers of the abdominal aorta, with a diameter of 30 mm or greater [ 1 , 2 ]. Endovascular aneurysm repair (EVAR) has emerged as the main treatment method for AAA [ 3 , 4 ]. Compared with traditional open repair, EVAR significantly reduces intraoperative bleeding, shortens hospital stays, and enhances postoperative quality of life for patients due to its minimally invasive nature [ 4 – 6 ]. The proper preoperative selection of intravascular stent graft is of paramount importance for the success of EVAR, as inaccuracies can lead to graft thrombosis, graft dislocation, and endoleaks [ 7 , 8 ]. The preoperative assessment of abdominal aortic anatomy and the selection of stent graft sizes rely on precise computed tomography combined with angiography (CTA) [ 9 ]. CTA postprocessing software facilitates the determination of relevant abdominal aortic parameters. In the early stages of EVAR's proposal, the preprocedural planning of EVAR has predominantly utilized two-dimensional (2D) imaging modalities such as picture archiving and communication system (PACS) workstations [ 10 , 11 ]. The measurement of PACS workstations relies on manually adjusting the sagittal, coronal, and transverse planes on dual tilt multiplanar reconstruction (MPR) to obtain a direction perpendicular to the aorta [ 12 – 14 ]. However, the accuracy and consistency of PACS measurements are notably suboptimal [ 14 ]. Over the past decade, the semi-automatic measurement software based on the central luminal line, has been employed in preoperative planning of EVAR [ 15 ]. Three-dimensional (3D) reconstructed images contribute to improving consistency between observers and reducing variability in preoperative measurements [ 16 , 17 ]. The fully automated measurement software we have developed, DetecMicro, utilizing mature and reliable graphic image algorithms to analyze digital imaging and communications in medicine (DICOM) images, creates a 3D model of the lesion site. This software employs 3D geometry and morphological algorithms to measure plane size and 3D size, providing morphological parameters, dimensions, angles, and other critical information. However, the accuracy of the sizing method employed by DetecMicro warrants further investigation. Successful EVAR relies on the appropriate selection of stent-grafts and precise positioning to strengthen the aorta [ 18 ]. Many centers still rely solely on the less sophisticated vessel centerline during preoperative planning to measure AAA-related lengths and predict the path for "virtual stenting" [ 19 ]. However, after implanting the vascular stent, the interaction between the stent and the blood vessel causes local folding and bending of the graft, resulting in a shorter stent coverage length than the predicted path [ 20 ]. Additionally, stent-induced remodeling of blood vessels leads to a shortened graft pathway, particularly in curved anatomical structures [ 19 , 21 ]. If the stent model is selected based on the centerline length measurement, it may cause the graft endpoint to be further away, posing a risk of obstructing the internal iliac artery [ 19 ]. Therefore, relying solely on the "centerline only" method for prediction may result in significant deviations. Accurately predicting the end position of an arterial stent after determining the proximal rivet point remains a persistent challenge. The purpose of this study was to assess the accuracy and reproducibility of DetecMicro in preoperative planning for AAA and evaluate the correctness of the path for virtual stent implantation (VSI). Methods CTA database constitution A retrospective identification was conducted for 30 male patients with AAA, each exhibiting a maximum diameter of ≥ 30 mm in any direction. These patients underwent CTA between January 1, 2023, and April 15, 2023. Figure 1 presents a flowchart illustrating the patient selection process. The CTA scans of the 30 patients were organized numerically from 1 to 30. The Ethics Committee of Tongji Medical College Affiliated Union Hospital, Huazhong University of Science and Technology approved this study (approval reference number: UHCT-IEC-SOP-016-03-01), and all patients provided written informed consent. The study design complied with the Declaration of Helsinki. CTA protocol All imaging examinations were performed on a dual-source CT system (Somatom Force; Siemens Healthineers, Forchheim, Germany). The scanning parameters were as follows: slice thickness of 0.75 mm, collimation of 92 × 0.6 mm, tube current time product of 110 mAs, tube voltage of 100 kV, and pitch of 0.8. Imaging was initiated after administering 60 mL of low-osmolar iodinated contrast agent (iopromide 370 mgI/mL; administered at a rate of 3.5 mL/s; Ultravist, Bayer, Wayne, NJ). Soft tissue window parameters of 40 HU center and 400 HU width were applied. Study design and participants The study was a three armed, stratified randomized controlled trial. To evaluate the accuracy and reproducibility of DetecMicro (version V01.03.10.01), PACS, a 2D measurement software, and 3mensio, a 3D semi-automatic software (version 10.3) were used to compare measurement results of the same sample. At the Vascular Surgery Department of Union Hospital affiliated with Huazhong University of Science and Technology, a total of 18 surgeons were recruited, including 9 junior residents (J) and 9 specialists (S) in vascular surgery, with 1–2 years and more than 10 years of experience, respectively. To ensure randomization, a computer-generated list of random numbers was used to assign the surgeons into three groups: PACS group, 3mensio group, and DetecMicro group. Each group included 3 junior residents and 3 specialists, with the allocation being stratified based on clinical experience. Before conducting the measurements, all surgeons received standardized training. They were then required to complete three repeated measurements of CTA for 30 patients, with a 4-week interval between each measurement. Image alalysis As shown in Fig. 2 , DetecMicro was using four steps as follows: Step 1: CT data loading and visualization (Fig. 2 A) The 2D slice views and the smooth volume rendering view were obtained. Step 2: Vessel segmentation (Fig. 2 B) Observers were instructed to select a portion of the vessel without calcification, and the blood vessel segment was automatically extracted. All vessels designated for assessment, such as the external iliac artery, renal artery, and celiac trunk angiography, were included in the segmentation, while unwanted tissues like bones and kidneys were eliminated. Observers could adjust the range of vessel segmentation using cut and add buttons. Step 3: Centerline extraction (Fig. 2 C) To create an automatic bifurcated centerline, four landmarks of the centerline needed to be indicated on the volume rendering: the start point, the bifurcation point, and two endpoints (one on the left and another on the right). It is essential to note that the starting point should be positioned above the celiac artery, while the endpoint should be located below the common iliac artery to evaluate the entire vascular pathway. Step 4: Template measurement ( Fig. 2 D ) The position along the centerline path was represented by a series of points in 3D coordinates. After selecting any point, the system automatically generated an orthogonal image perpendicular to the center-lumen-line and provided the minimum/maximum diameter of the plane. During the diameter measurement process, observers manually drew to mitigate the impact of thrombosis and calcification on the measurement results. When multiple points were selected, the system automatically calculated the distance between them. The AAA template measurement program required clicking on the following 10 points in sequence: the inferior margin of the lower renal artery, aortic bifurcation, left proximal iliac artery, right proximal iliac artery, left distal iliac artery, and right distal iliac artery. Subsequently, all diameters and lengths were automatically generated and displayed in the report within a few seconds (Fig. 2 E). For the fully automatic measurement function, the observer was only required to click the "Automatic calculation" button, and DetecMicro automatically provided all parameters in the template. Customization of Gold Standards Three experienced radiology experts conducted a one-time measurement on 30 samples using three different software. In cases where the analyzed variables exceeded 1 mm, disagreements among the three experts were resolved through reevaluation by a fourth expert. Finally, the gold standard for each quantitative variable was determined as the mean of the measurement results from all four experts. Selection of stenting Based on the aneurysm neck diameter (D1) and anatomical morphology, each observers selected the appropriate Medtronic stent following the principle of a stenting diameter 15–20% larger than D1. Virtual stenting implantation After considering the length of the stent implanted in patients who underwent EVAR, the "segment selection measurement" tool of the DetecMicro software was used to simulate the path of stent implantation. The intersection point between the celiac trunk and the abdominal aorta, as well as the bifurcation point of the common iliac artery, were served as markers. Subsequently, the distance between the end of the stenting and the marked point was compared with the postoperative CTA to assess the accuracy of VSI. Statistical analysis The measured variable errors of three groups of observers were used to evaluate the accuracy of each software. For quantitative variables, the correlation was evaluated using the intraclass coefficient correlation (ICC) with the Statistical Product and Service Software Automatically (SPSSAU). Quantitative results were analyzed using GraphPad Prism (v.7.0.0). One-way analysis of variance (ANOVA) and Student's t-test were used for mean comparisons. A p-value < 0.05 was considered statistically significant. Results 1. Accuracy of three measurement software The accuracy of various software tools was evaluated by examining the measurement deviations of observers from the "Gold Standards". In the APCS and 3mensio groups (Fig. 3 A, B), junior residents showed larger standard deviations and mean measurement errors for most parameters compared to specialists, unlike the DetecMicro group. Statistical analysis was conducted to compare the precision differences in diameter and length errors (Fig. 3 C, D). The DetecMicro groups demonstrated superior accuracy in diameter and length measurements (P < 0.001), especially the DetecMicro-S group (diameter error: 0.49 mm, length error: 1.13 mm), but no significant difference was found between the DetecMicro-S and DetecMicro-J groups (diameter error: 0.54 mm, length error: 1.50 mm) (Fig. 3 C, D, diameter: P = 0.592, length: P = 0.864). 2. Intra-observer and inter-observer reproducibility of measurements The study involved measuring 15 parameters three times by each observer, including 12 diameters and 3 lengths, to assess consistency. Novices in the PACS and 3mensio groups showed significantly poorer intra-observer reproducibility of diameter compared to specialists (Fig. 3 E, APCS: P = 0.049, 3mensio: P = 0.039). DetecMicro significantly improved consistency in diameter measurements for beginners compared to PACS and 3mensio (Fig. 3 E, P < 0.001). Compared with the APCS-J group, DetecMicro-J group exhibited higher consistency in length measurements (Fig. 3 F, P < 0.001). No significant differences in ICC were observed among other intra-observer groups in terms of length measurements (Fig. 3 E, P = 0.052). Subsequently, interobserver agreement was computed, revealing lower ICC values for diameter and length measurements in the PACS-J and 3mensio-J groups compared to their corresponding experienced groups (Fig. 3 G, H, P = 0.035). However, this phenomenon did not occur in the DetecMicro-J group (Fig. 3 G, H, P > 0.999). In addition, DetecMicro-J group also exhibited higher consistency in diameter and length measurements among observers compared to PACS-J and 3mensio-J groups (Fig. 3 G, H, P = 0.011). Among experienced observers, those using DetecMicro showed heightened inter-observer consistency, although not statistically significant compared to the 3mensio-S group (Fig. 3 G, H, diameter: P = 0.244, length: P > 0.999). 3. Influence of measurement error on the selection of stent graft models Diameters and lengths with absolute differences less than 2 mm and 5 mm, respectively, were individually tallied for each group. In DetecMicro-J and DetecMicro-S groups, 90.34% and 90.43% of diameter differences were below 2 mm, with none exceeding 5 mm (Fig. 4 A). The PACS and 3mensio groups had a reduced probability of diameter differences less than 2 mm. Similarly, DetecMicro-J and DetecMicro-S also showed advantages in length measurement, with 96.42% and 98.77% of differences less than 5 mm. To determine whether the measurement error affects stent selection, the stent models implanted in 30 patients were considered as a standard to evaluate the accuracy of stenting selection by other observers. Figure 4 C shows stent selection accuracy, seemingly unaffected by clinical experience (P > 0.999). However, DetecMicro-J and DetecMicro-S significantly improved accuracy compared to PACS and 3mensio, reaching 96.67% and 96.30%, respectively (Fig. 4 C). 4. Learning curve The learning curve assessed the efficiency of each software. PACS-J consistently took more time for the initial sample measurement, gradually decreasing until reaching a plateau. PACS-S reached a plateau early, with an average time of 11.70 mins, which was only 0.77 times that of PACS-J's average time of 15.17 mins (Fig. 5 ). In contrast, both the 3mensio and DetecMicro groups showed flatter learning curves, indicating a significant reduction in the initial sample measurement time. Unlike 3mensio-S, which required an average time of 5.45 mins for a single sample measurement, 3mensio-J required more time, with an average of 8.38 mins (Fig. 5 ). However, the DetecMicro J and S groups demonstrated similar average measurement times of 4.74 and 4.58 mins, respectively (Fig. 5 ). 5. Accuracy and consistency of DetecMicro's fully automated measurement function To assess DetecMicro's fully automatic measurement function, we compared it with semi-automatic measurements conducted by experienced clinical doctors. The results demonstrated that 97.41% of fully automatic diameter measurements had an error of under 2 mm, indicating a 6.98% improvement over the semi-automatic measurements (Fig. 4 A, 6 A). In terms of length, the fully automatic measurements achieved the same level of accuracy as the semi-automatic measurements, with no errors exceeding 5 mm (Fig. 4 B, 6 B). In addition, compared to semi-automatic measurement, fully automatic measurement has significant advantages in the consistency of diameter and length measurement. (Fig. 6 C, D, P = 0.362). The ICC for three consecutive fully automatic measurements exceeded 0.971, indicating strong repeatability (Table 1 ). Table 1 Correlation of three times fully automatic measurements and inter-observer measurement consistency in semi-automatic measurements using DetecMicro. Parameter ICC of semi-automatic ICC of fully automatic D1min 0.995 0.985 D1max 0.99 0.998 D2min 0.968 0.994 D2max 0.962 0.997 D3min 0.98 0.959 D3max 0.973 0.971 D4min 0.974 0.986 D4max 0.944 0.985 D5min 0.98 0.994 D5max 0.986 0.994 D6min 0.952 0.984 D6max 0.929 0.988 L1 0.983 0.999 L2 0.993 0.998 L3 0.989 0.997 6. Accuracy of the virtual stenting implantation The accuracy of VSI was assessed by comparing the distance between the stent's endpoint and the marker point with the actual post-EVAR CTA reconstruction model (CTA model). In the VSI simulation for thoracic endovascular aortic repair (TEVAR), two models were used to represent the stenting length: the centerline-based model (CL model) and greater curvature length-based model (GL model). As shown in Fig. 7 A, the GL model showed a smaller difference in distance (1.42 mm) compared to the CL model (32.01 mm) (Fig. 7 A, B, P = 0.004). Relying solely on CL or GL numerical values for simulating abdominal aortic stent implantation, especially in tortuous anatomical structures of AAAs, seemed unreasonable. The position of the stent bifurcation was determined by the proximal rivet point (inferior margin of the lowermost renal artery). Subsequently, the centerline path was readjusted to minimize the distance between the stent's bifurcation and the proximal ends of the left and right common iliac arteries (Fig. 7 C). Since the main stent body had less curvature, the CL represented the stent body length. Simultaneously, considering the significant curvature of the common iliac artery, which was prone to stent deformation, the GL model was used to represent the length of the iliac branch stenting in addition to the CL model (CL-GL model, Fig. 7 D). Compared to the CL model, the CL-GL model yielded satisfactory results with an error controlled within 1 mm (Fig. 7 E, P = 0.009). Discussion This investigation represented the first use of a three-armed, stratified randomized controlled trial to assess the accuracy and reproducibility of three different AAA preoperative planning software. Importantly, our findings suggested that DetecMicro demonstrated reliability and reproducibility in sizing. When using the PACS workstation for diameter measurement, the orthogonal plane determined by the observer is based on subjective selection to adjust the sagittal and coronal planes perpendicular to the vessel centerline [ 13 , 22 ]. Abada et al. [ 23 ] proposed that when the observer chose the orthogonal planes for measurement, the variability was greater compared to pre-selecting the slice. In the case of length measurement using PACS, it is hard to adjust the sagittal, coronal, and cross-sectional views of vessels with complex anatomical structures to achieve a consistent measurement range at one planes, and this process is significantly influenced by subjective factors [ 14 ]. The 3mensio and DetecMciro generated the centerline of vessels and provided a cross-section perpendicular to the centerline of the lumen. Previous research suggested that software providing a plane perpendicular to the centerline of the lumen improved the accuracy and repeatability of measurements [ 24 – 26 ]. However, surprisingly, 3mensio did not show a significant advantage in diameter measurement compared to PACS, possibly due to its inaccurate automatic identification of the vascular range, which requires manual adjustments (Fig. S1 ). In contrast, DetecMicro has achieved fast and accurate segmentation of target blood vessels by utilizing edge detection algorithms, morphological operations, and other methods. Multiple studies have also confirmed the high accuracy of these algorithms in model extraction [ 27 – 29 ]. Additively, compared with specialists, the accuracy, consistency, and time consumption of PACS and 3mensio measurements performed worse among beginners, but DetecMicro did not show this trend. Our study demonstrated that DetecMicro reduced the impact of clinical experience on measurement outcomes, allowing even young physicians to perform precise AAA preoperative planning. Given the 2–4 mm diameter differences in various Medtronic abdominal aortic stent models used in EVAR, our study evaluated the percentage of diameter errors below 2 mm ( https://www.medtronic.com/ ). Our results revealed that among the three types of software, a higher proportion of measurement errors less than 2 mm appeared in the DetecMicro group. However, it deserves further investigation whether the measurement errors of the three types of software affect the selection of stents. Precise sizing of aortic stent grafts has increasingly been recognized as a major element in achieving optimal early and late outcomes after EVAR [ 30 ]. Multiple studies have shown that device sizes larger than 20% are associated with late aortic neck dilation and subsequent graft migration [ 31 , 32 ]. However, the risk of type I endoleaks was remarkably increased with oversizing of < 10% [ 33 ]. Therefore, currently clinical doctors are more inclined to choose vascular grafts that are 15–20% larger in size to avoid these postoperative complications [ 34 ]. Our results in this study revealed DetecMicro could sensibly improve the accuracy of stent selection by reducing measurement errors. In addition, the normal diameter of the internal iliac artery is 5–6 mm, providing perfusion blood flow to the pelvic cavity [ 35 ]. Previous studies have reported adverse clinical consequences of internal iliac artery occlusion, including colon ischemia, hip limp, erectile dysfunction and spinal cord ischemia [ 36 , 37 ]. Overall, the limits of length error were within the clinically accepted range [− 5 mm, + 5 mm][ 22 ]. Unintentionally, our encouraging results indicated only DetecMciro could control over 99% of length errors within 5 mm, suggesting that this tool has a smaller possibility of excessive stent length covering the internal iliac artery due to the smaller measurement errors in comparison to PACS and 3mensio. Previous studies have highlighted the inaccuracy of using centerline-based measurements for preoperative planning of aortic aneurysms. Therefore, it is crucial to accurately predict the path of stent grafts in the blood vessel [ 38 ]. When stent grafts are deployed in angled arteries, the stent gaps fold or elongate to better accommodate smaller and greater curvature curves, respectively [ 39 ]. Our results revealed that, compared to CL, the VSI created based on the GL led to a more accurate prediction of graft pathway in TEAVR. However, the prediction of the abdominal aortic stent pathway could not solely rely on the GL. Due to the presence of an AAA, the straight and short neck portion was further stretched with the implantation of the stent body. The stent body was deployed according to the CL. Consistent with the thoracic aortic stent, the iliac artery stent was susceptible to bending due to the GL. Hence, we adopted the same strategy as deploying the iliac artery stent for the thoracic aortic stenting graft. Not surprisingly, our results confirmed for the first time that this CL-GL model was more accurate in predicting the pathway of abdominal aortic stents. In summary, our data indicated that the accuracy and consistency of diameter and length determined using the fully automated measurement software DetecteMicro were more reliable than manual and semi-automatic measurement software (PACS and 3mensio). Basing on DetectMicro's measurement function, the virtual stent implantation is expected to become a reliable tool for preoperative planning of EVAR treatment. Declarations Finding This work was supported by the National Natural Science Foundation of China [No. 82270520], Hubei Province Key Research and Development Plan [No. 2022BCA024] and Free Innovation Pre Research Fund of Tongji Medical College Affiliated Union Hospital of Huazhong University of Science and Technology [No. 2021xhyn09]. Conflict of Interest The authors declare no competing interests. Acknowledgements Thanks for all the support and contributions of participators. Thank you to Boea Wisdom (Hangzhou) Network Technology Co. for providing technical support for this research. Author contributions Chao Yang, Yiqing Li and Qi Li conceived the ideas. Linlin Guo, Xiaoyu Qi and Ming Yang designed and performed the experiments. Linlin Guo, Fei Cai and Peng Zhou analyzed the data. Gezheng Chen, Wanying Wu, and Bingjie Zhu provided critical materials. Linlin Guo and Xiaoyu Qi wrote the manuscript. Chao Yang, Yiqing Li and Qi Li supervised the study. All the authors have read and approved the final version of the manuscript for publication. Data availability The datasets generated and/or analyzed in the current study are available from the corresponding author upon reasonable request. References Hensley, S. E., Upchurch G. R. Jr. Repair of Abdominal Aortic Aneurysms: JACC Focus Seminar, Part 1. J Am Coll Cardiol. 80(8):821–831 (2022). Sakalihasan, N., Limet, R., Defawe, O. D. Abdominal aortic aneurysm. Lancet 2005. 365(9470):1577-89 (2022). Baman, J. R., Eskandari, M. K. What Is an Abdominal Aortic Aneurysm? JAMA. 328(22):2280 (2022). Tang, G. H. L., Tadros, R. O. Endovascular Aortic Repair in Nonagenarians: Select Well and Time Appropriately. J Am Coll Cardiol. 77(15):1900–1902 (2021). Lederle, F. A., et al. OVER Veterans Affairs Cooperative Study Group. Open versus Endovascular Repair of Abdominal Aortic Aneurysm. N Engl J Med. 380(22):2126–2135 (2019). Patel, R., Sweeting, M. J., Powell, J. T., Greenhalgh, R. M. EVAR trial investigators. Endovascular versus open repair of abdominal aortic aneurysm in 15-years' follow-up of the UK endovascular aneurysm repair trial 1 (EVAR trial 1): a randomised controlled trial. Lancet. 388(10058):2366–2374 (2016). Hostalrich, A., et al. Prospective Multicentre Cohort Study of Fenestrated and Branched Endografts After Failed Endovascular Infrarenal Aortic Aneurysm Repair with Type Ia Endoleak. Eur J Vasc Endovasc Surg. 62(4):540–548 (2021). Seike, Y., et al. Nationwide Analysis of Persistent Type II Endoleak and Late Outcomes of Endovascular Abdominal Aortic Aneurysm Repair in Japan: A Propensity-Matched Analysis. Circulation. 145(14):1056–1066 (2022). van Keulen, J. W., van Prehn, J., Prokop, M., Moll, F. L., van Herwaarden, J. A. Dynamics of the aorta before and after endovascular aneurysm repair: a systematic review. Eur J Vasc Endovasc Surg. 38(5):586–96 (2009). Kobus, K., et al. Cancer, cancer treatment and aneurysmatic ascending aorta growth within a retrospective single center study. Vasa. 52(1):38–45 (2023). von Tengg-Kobligk, H., Weber, T. F., Rengier, F., Böckler, D., Schumacher, H., Kauczor, H.U. Aktuelle Bildnachverarbeitung der aortalen CTA und MRA [Image postprocessing of aortic CTA and MRA]. Radiologe. 47(11):1003–11 (2007). Jeong, D. K., et al. On-the-fly generation of multiplanar reformation images independent of CT scanner type. J Digit Imaging. 21(3):306–11(2008). Khajouei, R., Jahromi, M. E., Ameri, A. Challenges of Implementing Picture Archiving and Communication System in Multiple Hospitals: Perspectives of Involved Staff and Users. J Med Syst. 43(7):182 (2009). Rengier, F., et al. Reliability of semiautomatic centerline analysis versus manual aortic measurement techniques for TEVAR among non-experts. Eur J Vasc Endovasc Surg. 42(3):324–31 (2011). Kaladji, A., Lucas, A., Kervio, G., Haigron, P., Cardon, A. Sizing for endovascular aneurysm repair: clinical evaluation of a new automated three-dimensional software. Ann Vasc Surg. 24(7):912–20 (2010). Sprouse, L. R., et al. Is three-dimensional computed tomography reconstruction justified before endovascular aortic aneurysm repair? J Vasc Surg. 40(3):443–7 (2004). Parker, M. V., et al. What imaging studies are necessary for abdominal aortic endograft sizing? A prospective blinded study using conventional computed tomography, aortography, and three-dimensional computed tomography. J Vasc Surg. 41(2):199–205 (2005). Picel, A. C., Kansal, N. Essentials of endovascular abdominal aortic aneurysm repair imaging: preprocedural assessment. AJR Am J Roentgenol. 203(4): W347-57 (2014). Whittaker, D. R., Dwyer, J., Fillinger, M.F. Prediction of altered endograft path during endovascular abdominal aortic aneurysm repair with the Gore Excluder. J Vasc Surg. 41(4):575–83 (2005). White, G. H., et al. Shortening of endografts during deployment in endovascular AAA repair. J Endovasc Surg. 6(1):4–10 (1999). Lin, K. K., Kratzberg, J. A., Raghavan, M. L. Role of aortic stent graft oversizing and barb characteristics on folding. J Vasc Surg. 55(5):1401–9 (2012). Mora, C., Marcus, C., Barbe, C., Ecarnot, F., Long, A. Measurement of maximum diameter of native abdominal aortic aneurysm by angio-CT: reproducibility is better with the semi-automated method. Eur J Vasc Endovasc Surg. 47(2):139–50 (2014). Abada, H. T., Sapoval, M. R., Paul, J. F., de Maertelaer, V., Mousseaux, E., Gaux JC. Aneurysmal sizing after endovascular repair in patients with abdominal aortic aneurysm: interobserver variability of various measurement protocols and its clinical relevance. Eur Radiol. 13(12):2699–704 (2003). Diehm, N., et al. Automated software supported versus manual aorto-iliac diameter measurements in CT angiography of patients with abdominal aortic aneurysms: assessment of inter- and intraobserver variation. Vasa. 34(4):255–61 (2005). England, A., García-Fiñana, M., How, T. V., Vallabhaneni, S. R., McWilliams, R. G. The accuracy of computed tomography central luminal line measurements in quantifying stent graft migration. J Vasc Surg. 55(4):895–905 (2012). Xian, R., Lugu, R., Peng, H., Yang, Q., Luo, X., Wang, J. Edge Detection Method Based on Nonlinear Spiking Neural Systems. Int J Neural Syst. 33(1):2250060 (2023). Al-Dmour, H., Al-Ani, A. Quality optimized medical image information hiding algorithm that employs edge detection and data coding. Comput Methods Programs Biomed. 127:24–43 (2016). Huan, L., Xue, N., Zheng, X., He, W., Gong, J., Xia, G. S. Unmixing Convolutional Features for Crisp Edge Detection. IEEE Trans Pattern Anal Mach Intell. 44(10):6602–6609 (2022). Sabra, W., Khouzam, M., Chanu, A., Martel, S. Use of 3D Potential Field and an Enhanced Breadth-first Search Algorithms for the Path Planning of Microdevices Propelled in the Cardiovascular System. Conf Proc IEEE Eng Med Biol Soc. 2005:3916–20 (2005). Sternbergh, W. C., Money, S. R., Greenberg, R. K., Chuter, T. A. Zenith Investigators. Influence of endograft oversizing on device migration, endoleak, aneurysm shrinkage, and aortic neck dilation: results from the Zenith Multicenter Trial. J Vasc Surg. 39(1):20–6 (2004). Conners, M. S., Sternbergh, W. C., Carter, G., Tonnessen, B. H., Yoselevitz, M., Money, S. R. Endograft migration one to four years after endovascular abdominal aortic aneurysm repair with the AneuRx device: a cautionary note. J Vasc Surg. 36(3):476–84 (2002). Oliveira, N. F. G., et al. Risk Factors, Dynamics, and Clinical Consequences of Aortic Neck Dilatation after Standard Endovascular Aneurysm Repair. Eur J Vasc Endovasc Surg. 62(1):26–35 (2021). Mohan, I. V., Laheij, R. J., Harris, P. L. EUROSTAR COLLABORATORS. Risk factors for endoleak and the evidence for stent-graft oversizing in patients undergoing endovascular aneurysm repair. Eur J Vasc Endovasc Surg. 21(4):344–9 (2001). Sultan, S., Evoy, D., Nicholls, S., Colgan, M.P., Moore, D., Shanik, G. Endoluminal stent grafts in the management of infrarenal abdominal aortic aneurysms: a realistic assessment. Eur J Vasc Endovasc Surg. 21(1):70–4 (2001). Lorbeer, R., et al. Reference values of vessel diameters, stenosis prevalence, and arterial variations of the lower limb arteries in a male population sample using contrast-enhanced MR angiography. PLoS One. 13(6):e0197559 (2018). Dalainas, I. A word of caution before killing hypogastric arteries. Cardiovasc Intervent Radiol. 31(4):835; author reply 836 (2008). Lin, P. H., Bush, R. L., Lumsden, A. B. Sloughing of the scrotal skin and impotence subsequent to bilateral hypogastric artery embolization for endovascular aortoiliac aneurysm repair. J Vasc Surg. 34(4):748–50 (2001). Iwakoshi, S., et al. Measuring the greater curvature length of virtual stent graft can provide accurate prediction of stent graft position for thoracic endovascular aortic repair. J Vasc Surg. 69(4):1021–1027 (2019). Tricarico, R., et al. Hemodynamic and Anatomic Predictors of Renovisceral Stent-Graft Occlusion Following Chimney Endovascular Repair of Juxtarenal Aortic Aneurysms. J Endovasc Ther. 24(6):880–888 (2017). Additional Declarations No competing interests reported. Supplementary Files Supplementaryfile.docx Cite Share Download PDF Status: Posted Version 1 posted 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-4857239","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":351099944,"identity":"085ba6f8-427d-46e7-a22e-0ad60080b2c0","order_by":0,"name":"Linlin Guo","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Linlin","middleName":"","lastName":"Guo","suffix":""},{"id":351099946,"identity":"e5cb18c1-66c2-42ff-8d86-60c92a243670","order_by":1,"name":"Xiaoyu Qi","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyu","middleName":"","lastName":"Qi","suffix":""},{"id":351099947,"identity":"49f0fab9-ab72-49d6-b8be-3208733ff43c","order_by":2,"name":"Ming Yang","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Ming","middleName":"","lastName":"Yang","suffix":""},{"id":351099948,"identity":"8c3fbecb-b9a9-4e71-95a1-5dbb22817258","order_by":3,"name":"Fei Cai","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Fei","middleName":"","lastName":"Cai","suffix":""},{"id":351099949,"identity":"a76e9eb8-3e2a-41d7-9ae1-d6c23c07d308","order_by":4,"name":"Peng Zhou","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Peng","middleName":"","lastName":"Zhou","suffix":""},{"id":351099950,"identity":"703b75e4-8333-4d5d-b298-c0d7cd0ac7c1","order_by":5,"name":"Gezheng Chen","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Gezheng","middleName":"","lastName":"Chen","suffix":""},{"id":351099951,"identity":"65e8eca3-ddea-4784-afab-bae9069ca8cb","order_by":6,"name":"Wanying Wu","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Wanying","middleName":"","lastName":"Wu","suffix":""},{"id":351099952,"identity":"ca64bf8e-1e4b-4a1c-ae63-75746ea5f6a8","order_by":7,"name":"Bingjie Zhu","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Bingjie","middleName":"","lastName":"Zhu","suffix":""},{"id":351099953,"identity":"2c82fd85-e377-4e89-841c-1af0685d6497","order_by":8,"name":"Qin Li","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Qin","middleName":"","lastName":"Li","suffix":""},{"id":351099954,"identity":"ddf56780-d332-4693-958e-bd5e5d281318","order_by":9,"name":"Yiqing Li","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Yiqing","middleName":"","lastName":"Li","suffix":""},{"id":351099955,"identity":"0f4d3c9e-4d2d-4021-a45d-9386ebf86c46","order_by":10,"name":"Chao Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxklEQVRIiWNgGAWjYFACHiCuOIBgE6nlDFALG0laGNtI0WJw/OzBDx/n3UmcP7+B8cHbNgZ5c4JazuQlS87c9ixxwzEGZsO5bQyGOxsIabnBYyDNu+1w4gY2BjZp3jaGBIMDhLUY/+adczhxfhsD+29itZhJ8zYcTmw4xsDGTJQWyTM5ZpYzjh023nAssVlyzjkJww2EtPAdP2N840PNYdn5zYcPfnhTZiNP0BYFhALGBiAhQUA9EMg3EFYzCkbBKBgFIx0AAHe7QxoPKR+hAAAAAElFTkSuQmCC","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Chao","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2024-08-04 14:32:39","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4857239/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4857239/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":66374693,"identity":"3458ec30-8611-4a42-88d0-3df74de48106","added_by":"auto","created_at":"2024-10-11 05:28:06","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2451427,"visible":true,"origin":"","legend":"\u003cp\u003eFlow chart of the computed tomography angiography (CTA) database constitution. AAA = abdominal aortic aneurysm.\u003c/p\u003e","description":"","filename":"Fig.1.png","url":"https://assets-eu.researchsquare.com/files/rs-4857239/v1/dd39c2029589afd9afcbc479.png"},{"id":66374692,"identity":"606c734c-e4e4-434a-9f6f-976708f7ee05","added_by":"auto","created_at":"2024-10-11 05:28:06","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":12295394,"visible":true,"origin":"","legend":"\u003cp\u003eSizing step of DetecMicro. A. CTA Data Loading and Visualization. Scale bars, 50 mm. B. Vessel segmentation. Scale bars, 50 mm. C. Centerline extraction. Scale bars, 50 mm. D. Template measurement. E. Sizing report. Scale bars, 50 mm.\u003c/p\u003e","description":"","filename":"Fig.2.png","url":"https://assets-eu.researchsquare.com/files/rs-4857239/v1/cc19b12d4071dd94f19ddfce.png"},{"id":66374695,"identity":"5005ad94-962f-46a6-b4ae-fa49911c02f7","added_by":"auto","created_at":"2024-10-11 05:28:06","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":3124192,"visible":true,"origin":"","legend":"\u003cp\u003eAccuracy and reproducibility of three measurement software. A. Measurement error of 12 diameter parameters in each group (n = 90 samples: 3 observers ×30 patients for each parameter, data are shown as mean and SD). B. Measurement error of 3 length parameters in each group (n = 90 samples: 3 observers ×30 patients for each parameter, data are shown as mean and SD). C. Statistical analysis of all diameter measurement errors (n = 1080 samples: 30 patients × 3 observers × 12 diameter parameters in each group). D. Statistical analysis of all length measurement errors (n = 270 samples: 30 patients × 3 observers × 3 length parameters in each group). E. Intra-observer reproducibility of diameter measurement (n = 36 samples: 3 observers × 12 diameter parameters in each group). F. Intra-observer reproducibility of length measurement (n = 9 samples: 3 observers × 3 length parameters in each group). G. Inter-observer reproducibility of diameter measurement in each group(n = 36 samples: 12 diameter parameters × 3 times). H. Intra-observer reproducibility of length measurement in each group. (n = 9 samples: 3 length parameters ×3 times), ns. P \u0026gt; 0.05, * P \u0026lt; 0.05, ** P \u0026lt; 0.01, *** P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"Fig.3.png","url":"https://assets-eu.researchsquare.com/files/rs-4857239/v1/1e6277cb406c0fc935d872c5.png"},{"id":66374689,"identity":"3fb4a392-8da4-4010-97dd-58108f2dc57f","added_by":"auto","created_at":"2024-10-11 05:28:05","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2108027,"visible":true,"origin":"","legend":"\u003cp\u003eInfluence of measurement error on stent graft selection. A. Absolute diameter differences ≤ 2 mm in each group (n = 1080 samples: 3 observers ×30 patients × 12 diameter parameters in each group). B. Absolute length differences ≤ 5 mm in each group (n = 270 samples: 3 observers ×30 patients × 3 length parameters in each group). C. Accuracy of stenting selection for each group (n = 9 sample: 3 observers × 3 times stenting selection in each group). ns. P \u0026gt; 0.05, *** P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"Fig.4.png","url":"https://assets-eu.researchsquare.com/files/rs-4857239/v1/64422aad61e335c15725d0e7.png"},{"id":66374690,"identity":"02401e24-72ce-45f8-9df9-5af3320fa8f3","added_by":"auto","created_at":"2024-10-11 05:28:05","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":938787,"visible":true,"origin":"","legend":"\u003cp\u003eLearning curve for each group (n = 3 observers).\u003c/p\u003e","description":"","filename":"Fig.5.png","url":"https://assets-eu.researchsquare.com/files/rs-4857239/v1/3a5d472afa0e7cab95721775.png"},{"id":66374955,"identity":"c3382158-6338-4165-8408-864354aac420","added_by":"auto","created_at":"2024-10-11 05:36:05","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":2112778,"visible":true,"origin":"","legend":"\u003cp\u003eAccuracy and consistency of DetecMicro‘s fully automated measurement function. A. Diameter difference of fully automatical DetecMicro (n = 30 samples: 3 observers ×10 patients for each parameter). B. Length difference of fully automatical DetecMicro (n = 30 samples: 3 observers ×10 patients). C. Statistical analysis of all diameter difference in both semi-automatic and fully automatic measurements by DetecMicro (n = 1080 samples: 12 diameter parameters × 30 patients × 3 times). D. Statistical analysis of all length difference in both semi-automatic and fully automatic measurements by DetecMicro. (n = 270 samples: 3 length parameters × 30 patients × 3 times). ns. P \u0026gt; 0.05.\u003c/p\u003e","description":"","filename":"Fig.6.png","url":"https://assets-eu.researchsquare.com/files/rs-4857239/v1/943539bf7f4f5cce313c7774.png"},{"id":66376318,"identity":"0082674b-54ed-46a8-808a-8bb2eefd3b19","added_by":"auto","created_at":"2024-10-11 06:00:06","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":13610882,"visible":true,"origin":"","legend":"\u003cp\u003eVirtual stenting implantation. A. Simulated stening based on centerline (CL model) and greater curvature length (GL model), and actual post-TEVAR CTA-reconstructed models (CTA model). Scale bars, 50 mm. B. Statistical results for the distance difference between the end of the virtual stent and the mark point in two different models of thoracic aortic stent placement compared to the CTA model (n = 3 patients). C. Process of preoperative centerline adjustment in patients with abdominal aortic aneurysm. Scale bars, 50 mm. D. Simulated stenting based on CL model and GL model, and the actual post-EVAR in patients with abdominal aortic aneurysm CTA model. Scale bars, 50 mm. E. Statistical results for the distance difference between the end of the virtual stent and the mark point in two different models of abdominal aortic virtual stent placement compared to the CTA model (n = 6 paired samples from 3 patients). ** P \u0026lt; 0.01.\u003c/p\u003e","description":"","filename":"Fig.7.png","url":"https://assets-eu.researchsquare.com/files/rs-4857239/v1/294e74e0b9378b7b1753c026.png"},{"id":68783988,"identity":"ae18f607-2305-44ce-a84c-167a2770401b","added_by":"auto","created_at":"2024-11-12 03:32:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":57101206,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4857239/v1/73f07ad4-1450-40e0-85d3-c71c2ea72481.pdf"},{"id":66374696,"identity":"89572bde-e35c-46e9-b515-1c10c1abebb2","added_by":"auto","created_at":"2024-10-11 05:28:06","extension":"docx","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":621216,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryfile.docx","url":"https://assets-eu.researchsquare.com/files/rs-4857239/v1/64f523ca4ace87e730a8b7e9.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Accurate preoperative planning for abdominal aortic aneurysm using fully automated measurement software: A randomized controlled trial","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAbdominal aortic aneurysm (AAA) is characterized by the dilation of the three vascular layers of the abdominal aorta, with a diameter of 30 mm or greater [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Endovascular aneurysm repair (EVAR) has emerged as the main treatment method for AAA [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Compared with traditional open repair, EVAR significantly reduces intraoperative bleeding, shortens hospital stays, and enhances postoperative quality of life for patients due to its minimally invasive nature [\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe proper preoperative selection of intravascular stent graft is of paramount importance for the success of EVAR, as inaccuracies can lead to graft thrombosis, graft dislocation, and endoleaks [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The preoperative assessment of abdominal aortic anatomy and the selection of stent graft sizes rely on precise computed tomography combined with angiography (CTA) [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. CTA postprocessing software facilitates the determination of relevant abdominal aortic parameters. In the early stages of EVAR's proposal, the preprocedural planning of EVAR has predominantly utilized two-dimensional (2D) imaging modalities such as picture archiving and communication system (PACS) workstations [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. The measurement of PACS workstations relies on manually adjusting the sagittal, coronal, and transverse planes on dual tilt multiplanar reconstruction (MPR) to obtain a direction perpendicular to the aorta [\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. However, the accuracy and consistency of PACS measurements are notably suboptimal [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Over the past decade, the semi-automatic measurement software based on the central luminal line, has been employed in preoperative planning of EVAR [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Three-dimensional (3D) reconstructed images contribute to improving consistency between observers and reducing variability in preoperative measurements [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. The fully automated measurement software we have developed, DetecMicro, utilizing mature and reliable graphic image algorithms to analyze digital imaging and communications in medicine (DICOM) images, creates a 3D model of the lesion site. This software employs 3D geometry and morphological algorithms to measure plane size and 3D size, providing morphological parameters, dimensions, angles, and other critical information. However, the accuracy of the sizing method employed by DetecMicro warrants further investigation.\u003c/p\u003e \u003cp\u003eSuccessful EVAR relies on the appropriate selection of stent-grafts and precise positioning to strengthen the aorta [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Many centers still rely solely on the less sophisticated vessel centerline during preoperative planning to measure AAA-related lengths and predict the path for \"virtual stenting\" [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. However, after implanting the vascular stent, the interaction between the stent and the blood vessel causes local folding and bending of the graft, resulting in a shorter stent coverage length than the predicted path [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Additionally, stent-induced remodeling of blood vessels leads to a shortened graft pathway, particularly in curved anatomical structures [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. If the stent model is selected based on the centerline length measurement, it may cause the graft endpoint to be further away, posing a risk of obstructing the internal iliac artery [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Therefore, relying solely on the \"centerline only\" method for prediction may result in significant deviations. Accurately predicting the end position of an arterial stent after determining the proximal rivet point remains a persistent challenge.\u003c/p\u003e \u003cp\u003eThe purpose of this study was to assess the accuracy and reproducibility of DetecMicro in preoperative planning for AAA and evaluate the correctness of the path for virtual stent implantation (VSI).\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCTA database constitution\u003c/h2\u003e \u003cp\u003eA retrospective identification was conducted for 30 male patients with AAA, each exhibiting a maximum diameter of \u0026ge;\u0026thinsp;30 mm in any direction. These patients underwent CTA between January 1, 2023, and April 15, 2023. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents a flowchart illustrating the patient selection process. The CTA scans of the 30 patients were organized numerically from 1 to 30. The Ethics Committee of Tongji Medical College Affiliated Union Hospital, Huazhong University of Science and Technology approved this study (approval reference number: UHCT-IEC-SOP-016-03-01), and all patients provided written informed consent. The study design complied with the Declaration of Helsinki.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eCTA protocol\u003c/h2\u003e \u003cp\u003eAll imaging examinations were performed on a dual-source CT system (Somatom Force; Siemens Healthineers, Forchheim, Germany). The scanning parameters were as follows: slice thickness of 0.75 mm, collimation of 92 \u0026times; 0.6 mm, tube current time product of 110 mAs, tube voltage of 100 kV, and pitch of 0.8. Imaging was initiated after administering 60 mL of low-osmolar iodinated contrast agent (iopromide 370 mgI/mL; administered at a rate of 3.5 mL/s; Ultravist, Bayer, Wayne, NJ). Soft tissue window parameters of 40 HU center and 400 HU width were applied.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eStudy design and participants\u003c/h2\u003e \u003cp\u003eThe study was a three armed, stratified randomized controlled trial. To evaluate the accuracy and reproducibility of DetecMicro (version V01.03.10.01), PACS, a 2D measurement software, and 3mensio, a 3D semi-automatic software (version 10.3) were used to compare measurement results of the same sample. At the Vascular Surgery Department of Union Hospital affiliated with Huazhong University of Science and Technology, a total of 18 surgeons were recruited, including 9 junior residents (J) and 9 specialists (S) in vascular surgery, with 1\u0026ndash;2 years and more than 10 years of experience, respectively. To ensure randomization, a computer-generated list of random numbers was used to assign the surgeons into three groups: PACS group, 3mensio group, and DetecMicro group. Each group included 3 junior residents and 3 specialists, with the allocation being stratified based on clinical experience. Before conducting the measurements, all surgeons received standardized training. They were then required to complete three repeated measurements of CTA for 30 patients, with a 4-week interval between each measurement.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eImage alalysis\u003c/h2\u003e \u003cp\u003eAs shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, DetecMicro was using four steps as follows:\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStep 1: CT data loading and visualization (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA)\u003c/h2\u003e \u003cp\u003eThe 2D slice views and the smooth volume rendering view were obtained.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eStep 2: Vessel segmentation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB)\u003c/h2\u003e \u003cp\u003eObservers were instructed to select a portion of the vessel without calcification, and the blood vessel segment was automatically extracted. All vessels designated for assessment, such as the external iliac artery, renal artery, and celiac trunk angiography, were included in the segmentation, while unwanted tissues like bones and kidneys were eliminated. Observers could adjust the range of vessel segmentation using cut and add buttons.\u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003eStep 3: Centerline extraction (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC)\u003c/h2\u003e \u003cp\u003eTo create an automatic bifurcated centerline, four landmarks of the centerline needed to be indicated on the volume rendering: the start point, the bifurcation point, and two endpoints (one on the left and another on the right). It is essential to note that the starting point should be positioned above the celiac artery, while the endpoint should be located below the common iliac artery to evaluate the entire vascular pathway.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\n\u003cdiv class=\"Heading\"\u003e\u003cb\u003eStep 4: Template measurement (\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD\u003cb\u003e)\u003c/b\u003e\u003c/div\u003e \u003cp\u003eThe position along the centerline path was represented by a series of points in 3D coordinates. After selecting any point, the system automatically generated an orthogonal image perpendicular to the center-lumen-line and provided the minimum/maximum diameter of the plane. During the diameter measurement process, observers manually drew to mitigate the impact of thrombosis and calcification on the measurement results. When multiple points were selected, the system automatically calculated the distance between them. The AAA template measurement program required clicking on the following 10 points in sequence: the inferior margin of the lower renal artery, aortic bifurcation, left proximal iliac artery, right proximal iliac artery, left distal iliac artery, and right distal iliac artery. Subsequently, all diameters and lengths were automatically generated and displayed in the report within a few seconds (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE).\u003c/p\u003e \u003cp\u003eFor the fully automatic measurement function, the observer was only required to click the \"Automatic calculation\" button, and DetecMicro automatically provided all parameters in the template.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCustomization of Gold Standards\u003c/h2\u003e \u003cp\u003eThree experienced radiology experts conducted a one-time measurement on 30 samples using three different software. In cases where the analyzed variables exceeded 1 mm, disagreements among the three experts were resolved through reevaluation by a fourth expert. Finally, the gold standard for each quantitative variable was determined as the mean of the measurement results from all four experts.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eSelection of stenting\u003c/h2\u003e \u003cp\u003eBased on the aneurysm neck diameter (D1) and anatomical morphology, each observers selected the appropriate Medtronic stent following the principle of a stenting diameter 15\u0026ndash;20% larger than D1.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eVirtual stenting implantation\u003c/h2\u003e \u003cp\u003eAfter considering the length of the stent implanted in patients who underwent EVAR, the \"segment selection measurement\" tool of the DetecMicro software was used to simulate the path of stent implantation. The intersection point between the celiac trunk and the abdominal aorta, as well as the bifurcation point of the common iliac artery, were served as markers. Subsequently, the distance between the end of the stenting and the marked point was compared with the postoperative CTA to assess the accuracy of VSI.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eThe measured variable errors of three groups of observers were used to evaluate the accuracy of each software. For quantitative variables, the correlation was evaluated using the intraclass coefficient correlation (ICC) with the Statistical Product and Service Software Automatically (SPSSAU). Quantitative results were analyzed using GraphPad Prism (v.7.0.0). One-way analysis of variance (ANOVA) and Student's t-test were used for mean comparisons. A p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e "},{"header":"Results","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e1. Accuracy of three measurement software\u003c/h2\u003e \u003cp\u003eThe accuracy of various software tools was evaluated by examining the measurement deviations of observers from the \"Gold Standards\". In the APCS and 3mensio groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, B), junior residents showed larger standard deviations and mean measurement errors for most parameters compared to specialists, unlike the DetecMicro group. Statistical analysis was conducted to compare the precision differences in diameter and length errors (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC, D). The DetecMicro groups demonstrated superior accuracy in diameter and length measurements (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), especially the DetecMicro-S group (diameter error: 0.49 mm, length error: 1.13 mm), but no significant difference was found between the DetecMicro-S and DetecMicro-J groups (diameter error: 0.54 mm, length error: 1.50 mm) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC, D, diameter: P\u0026thinsp;=\u0026thinsp;0.592, length: P\u0026thinsp;=\u0026thinsp;0.864).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e2. Intra-observer and inter-observer reproducibility of measurements\u003c/h2\u003e \u003cp\u003eThe study involved measuring 15 parameters three times by each observer, including 12 diameters and 3 lengths, to assess consistency. Novices in the PACS and 3mensio groups showed significantly poorer intra-observer reproducibility of diameter compared to specialists (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE, APCS: P\u0026thinsp;=\u0026thinsp;0.049, 3mensio: P\u0026thinsp;=\u0026thinsp;0.039). DetecMicro significantly improved consistency in diameter measurements for beginners compared to PACS and 3mensio (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Compared with the APCS-J group, DetecMicro-J group exhibited higher consistency in length measurements (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001). No significant differences in ICC were observed among other intra-observer groups in terms of length measurements (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE, P\u0026thinsp;=\u0026thinsp;0.052). Subsequently, interobserver agreement was computed, revealing lower ICC values for diameter and length measurements in the PACS-J and 3mensio-J groups compared to their corresponding experienced groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG, H, P\u0026thinsp;=\u0026thinsp;0.035). However, this phenomenon did not occur in the DetecMicro-J group (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG, H, P\u0026thinsp;\u0026gt;\u0026thinsp;0.999). In addition, DetecMicro-J group also exhibited higher consistency in diameter and length measurements among observers compared to PACS-J and 3mensio-J groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG, H, P\u0026thinsp;=\u0026thinsp;0.011). Among experienced observers, those using DetecMicro showed heightened inter-observer consistency, although not statistically significant compared to the 3mensio-S group (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG, H, diameter: P\u0026thinsp;=\u0026thinsp;0.244, length: P\u0026thinsp;\u0026gt;\u0026thinsp;0.999).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3. Influence of measurement error on the selection of stent graft models\u003c/h2\u003e \u003cp\u003eDiameters and lengths with absolute differences less than 2 mm and 5 mm, respectively, were individually tallied for each group. In DetecMicro-J and DetecMicro-S groups, 90.34% and 90.43% of diameter differences were below 2 mm, with none exceeding 5 mm (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). The PACS and 3mensio groups had a reduced probability of diameter differences less than 2 mm. Similarly, DetecMicro-J and DetecMicro-S also showed advantages in length measurement, with 96.42% and 98.77% of differences less than 5 mm. To determine whether the measurement error affects stent selection, the stent models implanted in 30 patients were considered as a standard to evaluate the accuracy of stenting selection by other observers. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC shows stent selection accuracy, seemingly unaffected by clinical experience (P\u0026thinsp;\u0026gt;\u0026thinsp;0.999). However, DetecMicro-J and DetecMicro-S significantly improved accuracy compared to PACS and 3mensio, reaching 96.67% and 96.30%, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e4. Learning curve\u003c/h2\u003e \u003cp\u003eThe learning curve assessed the efficiency of each software. PACS-J consistently took more time for the initial sample measurement, gradually decreasing until reaching a plateau. PACS-S reached a plateau early, with an average time of 11.70 mins, which was only 0.77 times that of PACS-J's average time of 15.17 mins (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). In contrast, both the 3mensio and DetecMicro groups showed flatter learning curves, indicating a significant reduction in the initial sample measurement time. Unlike 3mensio-S, which required an average time of 5.45 mins for a single sample measurement, 3mensio-J required more time, with an average of 8.38 mins (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). However, the DetecMicro J and S groups demonstrated similar average measurement times of 4.74 and 4.58 mins, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e5. Accuracy and consistency of DetecMicro's fully automated measurement function\u003c/h2\u003e \u003cp\u003eTo assess DetecMicro's fully automatic measurement function, we compared it with semi-automatic measurements conducted by experienced clinical doctors. The results demonstrated that 97.41% of fully automatic diameter measurements had an error of under 2 mm, indicating a 6.98% improvement over the semi-automatic measurements (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). In terms of length, the fully automatic measurements achieved the same level of accuracy as the semi-automatic measurements, with no errors exceeding 5 mm (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB, \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). In addition, compared to semi-automatic measurement, fully automatic measurement has significant advantages in the consistency of diameter and length measurement. (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC, D, P\u0026thinsp;=\u0026thinsp;0.362). The ICC for three consecutive fully automatic measurements exceeded 0.971, indicating strong repeatability (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCorrelation of three times fully automatic measurements and inter-observer measurement consistency in semi-automatic measurements using DetecMicro.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParameter\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eICC of semi-automatic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC of fully automatic\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD1min\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.995\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.985\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD1max\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.998\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD2min\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.968\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.994\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD2max\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.962\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.997\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD3min\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.959\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD3max\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.973\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.971\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD4min\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.974\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.986\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD4max\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.944\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.985\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD5min\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.994\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD5max\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.986\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.994\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD6min\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.952\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.984\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD6max\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.929\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.988\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eL1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.983\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.999\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eL2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.993\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.998\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eL3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.989\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.997\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e6. Accuracy of the virtual stenting implantation\u003c/h2\u003e \u003cp\u003eThe accuracy of VSI was assessed by comparing the distance between the stent's endpoint and the marker point with the actual post-EVAR CTA reconstruction model (CTA model). In the VSI simulation for thoracic endovascular aortic repair (TEVAR), two models were used to represent the stenting length: the centerline-based model (CL model) and greater curvature length-based model (GL model). As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA, the GL model showed a smaller difference in distance (1.42 mm) compared to the CL model (32.01 mm) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA, B, P\u0026thinsp;=\u0026thinsp;0.004). Relying solely on CL or GL numerical values for simulating abdominal aortic stent implantation, especially in tortuous anatomical structures of AAAs, seemed unreasonable. The position of the stent bifurcation was determined by the proximal rivet point (inferior margin of the lowermost renal artery). Subsequently, the centerline path was readjusted to minimize the distance between the stent's bifurcation and the proximal ends of the left and right common iliac arteries (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC). Since the main stent body had less curvature, the CL represented the stent body length. Simultaneously, considering the significant curvature of the common iliac artery, which was prone to stent deformation, the GL model was used to represent the length of the iliac branch stenting in addition to the CL model (CL-GL model, Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD). Compared to the CL model, the CL-GL model yielded satisfactory results with an error controlled within 1 mm (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eE, P\u0026thinsp;=\u0026thinsp;0.009).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis investigation represented the first use of a three-armed, stratified randomized controlled trial to assess the accuracy and reproducibility of three different AAA preoperative planning software. Importantly, our findings suggested that DetecMicro demonstrated reliability and reproducibility in sizing. When using the PACS workstation for diameter measurement, the orthogonal plane determined by the observer is based on subjective selection to adjust the sagittal and coronal planes perpendicular to the vessel centerline [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Abada et al. [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] proposed that when the observer chose the orthogonal planes for measurement, the variability was greater compared to pre-selecting the slice. In the case of length measurement using PACS, it is hard to adjust the sagittal, coronal, and cross-sectional views of vessels with complex anatomical structures to achieve a consistent measurement range at one planes, and this process is significantly influenced by subjective factors [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. The 3mensio and DetecMciro generated the centerline of vessels and provided a cross-section perpendicular to the centerline of the lumen. Previous research suggested that software providing a plane perpendicular to the centerline of the lumen improved the accuracy and repeatability of measurements [\u003cspan additionalcitationids=\"CR25\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. However, surprisingly, 3mensio did not show a significant advantage in diameter measurement compared to PACS, possibly due to its inaccurate automatic identification of the vascular range, which requires manual adjustments (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). In contrast, DetecMicro has achieved fast and accurate segmentation of target blood vessels by utilizing edge detection algorithms, morphological operations, and other methods. Multiple studies have also confirmed the high accuracy of these algorithms in model extraction [\u003cspan additionalcitationids=\"CR28\" citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Additively, compared with specialists, the accuracy, consistency, and time consumption of PACS and 3mensio measurements performed worse among beginners, but DetecMicro did not show this trend. Our study demonstrated that DetecMicro reduced the impact of clinical experience on measurement outcomes, allowing even young physicians to perform precise AAA preoperative planning.\u003c/p\u003e \u003cp\u003eGiven the 2\u0026ndash;4 mm diameter differences in various Medtronic abdominal aortic stent models used in EVAR, our study evaluated the percentage of diameter errors below 2 mm (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.medtronic.com/\u003c/span\u003e\u003cspan address=\"https://www.medtronic.com/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Our results revealed that among the three types of software, a higher proportion of measurement errors less than 2 mm appeared in the DetecMicro group. However, it deserves further investigation whether the measurement errors of the three types of software affect the selection of stents. Precise sizing of aortic stent grafts has increasingly been recognized as a major element in achieving optimal early and late outcomes after EVAR [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Multiple studies have shown that device sizes larger than 20% are associated with late aortic neck dilation and subsequent graft migration [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. However, the risk of type I endoleaks was remarkably increased with oversizing of \u0026lt;\u0026thinsp;10% [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Therefore, currently clinical doctors are more inclined to choose vascular grafts that are 15\u0026ndash;20% larger in size to avoid these postoperative complications [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Our results in this study revealed DetecMicro could sensibly improve the accuracy of stent selection by reducing measurement errors. In addition, the normal diameter of the internal iliac artery is 5\u0026ndash;6 mm, providing perfusion blood flow to the pelvic cavity [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Previous studies have reported adverse clinical consequences of internal iliac artery occlusion, including colon ischemia, hip limp, erectile dysfunction and spinal cord ischemia [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Overall, the limits of length error were within the clinically accepted range [\u0026minus;\u0026thinsp;5 mm, +\u0026thinsp;5 mm][\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Unintentionally, our encouraging results indicated only DetecMciro could control over 99% of length errors within 5 mm, suggesting that this tool has a smaller possibility of excessive stent length covering the internal iliac artery due to the smaller measurement errors in comparison to PACS and 3mensio.\u003c/p\u003e \u003cp\u003ePrevious studies have highlighted the inaccuracy of using centerline-based measurements for preoperative planning of aortic aneurysms. Therefore, it is crucial to accurately predict the path of stent grafts in the blood vessel [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. When stent grafts are deployed in angled arteries, the stent gaps fold or elongate to better accommodate smaller and greater curvature curves, respectively [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Our results revealed that, compared to CL, the VSI created based on the GL led to a more accurate prediction of graft pathway in TEAVR. However, the prediction of the abdominal aortic stent pathway could not solely rely on the GL. Due to the presence of an AAA, the straight and short neck portion was further stretched with the implantation of the stent body. The stent body was deployed according to the CL. Consistent with the thoracic aortic stent, the iliac artery stent was susceptible to bending due to the GL. Hence, we adopted the same strategy as deploying the iliac artery stent for the thoracic aortic stenting graft. Not surprisingly, our results confirmed for the first time that this CL-GL model was more accurate in predicting the pathway of abdominal aortic stents.\u003c/p\u003e \u003cp\u003eIn summary, our data indicated that the accuracy and consistency of diameter and length determined using the fully automated measurement software DetecteMicro were more reliable than manual and semi-automatic measurement software (PACS and 3mensio). Basing on DetectMicro's measurement function, the virtual stent implantation is expected to become a reliable tool for preoperative planning of EVAR treatment.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFinding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Natural Science Foundation of China [No. 82270520], Hubei Province Key Research and Development Plan [No. 2022BCA024] and Free Innovation Pre Research Fund of Tongji Medical College Affiliated Union Hospital of Huazhong University of Science and Technology [No. 2021xhyn09].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThanks for all the support and contributions of participators. Thank you to Boea Wisdom (Hangzhou) Network Technology Co. for providing technical support for this research.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eChao Yang, Yiqing Li and Qi Li conceived the ideas. Linlin Guo, Xiaoyu Qi and Ming Yang designed and performed the experiments. Linlin Guo, Fei Cai and Peng Zhou analyzed the data. Gezheng Chen, Wanying Wu, and Bingjie Zhu provided critical materials. Linlin Guo and Xiaoyu Qi wrote the manuscript. Chao Yang, Yiqing Li and Qi Li supervised the study. All the authors have read and approved the final version of the manuscript for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated and/or analyzed in the current study are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eHensley, S. E., Upchurch G. R. Jr. Repair of Abdominal Aortic Aneurysms: JACC Focus Seminar, Part 1. J Am Coll Cardiol. 80(8):821\u0026ndash;831 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSakalihasan, N., Limet, R., Defawe, O. D. Abdominal aortic aneurysm. Lancet 2005. 365(9470):1577-89 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaman, J. R., Eskandari, M. K. What Is an Abdominal Aortic Aneurysm? JAMA. 328(22):2280 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang, G. H. L., Tadros, R. O. Endovascular Aortic Repair in Nonagenarians: Select Well and Time Appropriately. J Am Coll Cardiol. 77(15):1900\u0026ndash;1902 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLederle, F. A., et al. OVER Veterans Affairs Cooperative Study Group. Open versus Endovascular Repair of Abdominal Aortic Aneurysm. N Engl J Med. 380(22):2126\u0026ndash;2135 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatel, R., Sweeting, M. J., Powell, J. T., Greenhalgh, R. M. EVAR trial investigators. Endovascular versus open repair of abdominal aortic aneurysm in 15-years' follow-up of the UK endovascular aneurysm repair trial 1 (EVAR trial 1): a randomised controlled trial. Lancet. 388(10058):2366\u0026ndash;2374 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHostalrich, A., et al. Prospective Multicentre Cohort Study of Fenestrated and Branched Endografts After Failed Endovascular Infrarenal Aortic Aneurysm Repair with Type Ia Endoleak. Eur J Vasc Endovasc Surg. 62(4):540\u0026ndash;548 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeike, Y., et al. Nationwide Analysis of Persistent Type II Endoleak and Late Outcomes of Endovascular Abdominal Aortic Aneurysm Repair in Japan: A Propensity-Matched Analysis. Circulation. 145(14):1056\u0026ndash;1066 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Keulen, J. W., van Prehn, J., Prokop, M., Moll, F. L., van Herwaarden, J. A. Dynamics of the aorta before and after endovascular aneurysm repair: a systematic review. Eur J Vasc Endovasc Surg. 38(5):586\u0026ndash;96 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKobus, K., et al. Cancer, cancer treatment and aneurysmatic ascending aorta growth within a retrospective single center study. Vasa. 52(1):38\u0026ndash;45 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evon Tengg-Kobligk, H., Weber, T. F., Rengier, F., B\u0026ouml;ckler, D., Schumacher, H., Kauczor, H.U. Aktuelle Bildnachverarbeitung der aortalen CTA und MRA [Image postprocessing of aortic CTA and MRA]. Radiologe. 47(11):1003\u0026ndash;11 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJeong, D. K., et al. On-the-fly generation of multiplanar reformation images independent of CT scanner type. J Digit Imaging. 21(3):306\u0026ndash;11(2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhajouei, R., Jahromi, M. E., Ameri, A. Challenges of Implementing Picture Archiving and Communication System in Multiple Hospitals: Perspectives of Involved Staff and Users. J Med Syst. 43(7):182 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRengier, F., et al. Reliability of semiautomatic centerline analysis versus manual aortic measurement techniques for TEVAR among non-experts. Eur J Vasc Endovasc Surg. 42(3):324\u0026ndash;31 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaladji, A., Lucas, A., Kervio, G., Haigron, P., Cardon, A. Sizing for endovascular aneurysm repair: clinical evaluation of a new automated three-dimensional software. Ann Vasc Surg. 24(7):912\u0026ndash;20 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSprouse, L. R., et al. Is three-dimensional computed tomography reconstruction justified before endovascular aortic aneurysm repair? J Vasc Surg. 40(3):443\u0026ndash;7 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParker, M. V., et al. What imaging studies are necessary for abdominal aortic endograft sizing? A prospective blinded study using conventional computed tomography, aortography, and three-dimensional computed tomography. J Vasc Surg. 41(2):199\u0026ndash;205 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePicel, A. C., Kansal, N. Essentials of endovascular abdominal aortic aneurysm repair imaging: preprocedural assessment. AJR Am J Roentgenol. 203(4): W347-57 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhittaker, D. R., Dwyer, J., Fillinger, M.F. Prediction of altered endograft path during endovascular abdominal aortic aneurysm repair with the Gore Excluder. J Vasc Surg. 41(4):575\u0026ndash;83 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhite, G. H., et al. Shortening of endografts during deployment in endovascular AAA repair. J Endovasc Surg. 6(1):4\u0026ndash;10 (1999).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin, K. K., Kratzberg, J. A., Raghavan, M. L. Role of aortic stent graft oversizing and barb characteristics on folding. J Vasc Surg. 55(5):1401\u0026ndash;9 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMora, C., Marcus, C., Barbe, C., Ecarnot, F., Long, A. Measurement of maximum diameter of native abdominal aortic aneurysm by angio-CT: reproducibility is better with the semi-automated method. Eur J Vasc Endovasc Surg. 47(2):139\u0026ndash;50 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbada, H. T., Sapoval, M. R., Paul, J. F., de Maertelaer, V., Mousseaux, E., Gaux JC. Aneurysmal sizing after endovascular repair in patients with abdominal aortic aneurysm: interobserver variability of various measurement protocols and its clinical relevance. Eur Radiol. 13(12):2699\u0026ndash;704 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiehm, N., et al. Automated software supported versus manual aorto-iliac diameter measurements in CT angiography of patients with abdominal aortic aneurysms: assessment of inter- and intraobserver variation. Vasa. 34(4):255\u0026ndash;61 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEngland, A., Garc\u0026iacute;a-Fi\u0026ntilde;ana, M., How, T. V., Vallabhaneni, S. R., McWilliams, R. G. The accuracy of computed tomography central luminal line measurements in quantifying stent graft migration. J Vasc Surg. 55(4):895\u0026ndash;905 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXian, R., Lugu, R., Peng, H., Yang, Q., Luo, X., Wang, J. Edge Detection Method Based on Nonlinear Spiking Neural Systems. Int J Neural Syst. 33(1):2250060 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl-Dmour, H., Al-Ani, A. Quality optimized medical image information hiding algorithm that employs edge detection and data coding. Comput Methods Programs Biomed. 127:24\u0026ndash;43 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuan, L., Xue, N., Zheng, X., He, W., Gong, J., Xia, G. S. Unmixing Convolutional Features for Crisp Edge Detection. IEEE Trans Pattern Anal Mach Intell. 44(10):6602\u0026ndash;6609 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSabra, W., Khouzam, M., Chanu, A., Martel, S. Use of 3D Potential Field and an Enhanced Breadth-first Search Algorithms for the Path Planning of Microdevices Propelled in the Cardiovascular System. Conf Proc IEEE Eng Med Biol Soc. 2005:3916\u0026ndash;20 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSternbergh, W. C., Money, S. R., Greenberg, R. K., Chuter, T. A. Zenith Investigators. Influence of endograft oversizing on device migration, endoleak, aneurysm shrinkage, and aortic neck dilation: results from the Zenith Multicenter Trial. J Vasc Surg. 39(1):20\u0026ndash;6 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eConners, M. S., Sternbergh, W. C., Carter, G., Tonnessen, B. H., Yoselevitz, M., Money, S. R. Endograft migration one to four years after endovascular abdominal aortic aneurysm repair with the AneuRx device: a cautionary note. J Vasc Surg. 36(3):476\u0026ndash;84 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOliveira, N. F. G., et al. Risk Factors, Dynamics, and Clinical Consequences of Aortic Neck Dilatation after Standard Endovascular Aneurysm Repair. Eur J Vasc Endovasc Surg. 62(1):26\u0026ndash;35 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMohan, I. V., Laheij, R. J., Harris, P. L. EUROSTAR COLLABORATORS. Risk factors for endoleak and the evidence for stent-graft oversizing in patients undergoing endovascular aneurysm repair. Eur J Vasc Endovasc Surg. 21(4):344\u0026ndash;9 (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSultan, S., Evoy, D., Nicholls, S., Colgan, M.P., Moore, D., Shanik, G. Endoluminal stent grafts in the management of infrarenal abdominal aortic aneurysms: a realistic assessment. Eur J Vasc Endovasc Surg. 21(1):70\u0026ndash;4 (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLorbeer, R., et al. Reference values of vessel diameters, stenosis prevalence, and arterial variations of the lower limb arteries in a male population sample using contrast-enhanced MR angiography. PLoS One. 13(6):e0197559 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDalainas, I. A word of caution before killing hypogastric arteries. Cardiovasc Intervent Radiol. 31(4):835; author reply 836 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin, P. H., Bush, R. L., Lumsden, A. B. Sloughing of the scrotal skin and impotence subsequent to bilateral hypogastric artery embolization for endovascular aortoiliac aneurysm repair. J Vasc Surg. 34(4):748\u0026ndash;50 (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIwakoshi, S., et al. Measuring the greater curvature length of virtual stent graft can provide accurate prediction of stent graft position for thoracic endovascular aortic repair. J Vasc Surg. 69(4):1021\u0026ndash;1027 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTricarico, R., et al. Hemodynamic and Anatomic Predictors of Renovisceral Stent-Graft Occlusion Following Chimney Endovascular Repair of Juxtarenal Aortic Aneurysms. J Endovasc Ther. 24(6):880\u0026ndash;888 (2017).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Abdominal aortic aneurysm, Fully automatic measurement software, Hemodynamics, Preoperative planning, Virtual stent","lastPublishedDoi":"10.21203/rs.3.rs-4857239/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4857239/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSizing, the first step of endovascular aneurysm repair (EVAR), is essential for a successful procedure. This study evaluated the precision and reproducibility of EVAR sizing facilitated by a novel fully automated software, DetecMicro, in comparison to conventional manual and semi-automatic software. A total of 18 surgeons, consisting of 9 junior residents and 9 vascular surgery specialists, participated in a prospective single-center randomized controlled trial with three parallel arms, stratified based on clinical experience. Each surgeon conducted three repeated measurements for 450 parameters (360 diameter and 90 length parameters). Intra- and inter-observer variability were analyzed using the intraclass correlation coefficient (ICC). Subsequently, the stent size based on the measured results was assessed to determine the impact of measurement errors on stent selection. The reliability of virtual stent implantation (VSI) using DetecMicro was evaluated by comparing it with postoperative models. Compared to PACS and 3mensio, the DetecMicro group exhibited superior accuracy, with 90.39% of diameter measurements and 97.60% of length measurements falling within clinically acceptable ranges, [-2 mm, +\u0026thinsp;2 mm] and [-5 mm, +\u0026thinsp;5 mm], respectively. Intra-observer and inter-observer repeatability with DetecMicro demonstrated efficacy, with a mean ICC exceeding 0.9. In the DetecMicro group, clinical experience had a negligible impact on the aforementioned results. VSI, when compared with actual postoperative models, limited errors to within 2 mm. The integration of DetecMicro's measurement and VSI functions holds promise as a reliable tool for preoperative planning in EVAR treatment.\u003c/p\u003e","manuscriptTitle":"Accurate preoperative planning for abdominal aortic aneurysm using fully automated measurement software: A randomized controlled trial","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-10-11 05:28:00","doi":"10.21203/rs.3.rs-4857239/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"24886c76-9dca-4c2e-9aec-2c88ee022b8b","owner":[],"postedDate":"October 11th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":37281338,"name":"Health sciences/Diseases"},{"id":37281339,"name":"Health sciences/Medical research"},{"id":37281340,"name":"Physical sciences/Engineering"}],"tags":[],"updatedAt":"2024-11-12T03:24:10+00:00","versionOfRecord":[],"versionCreatedAt":"2024-10-11 05:28:00","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4857239","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4857239","identity":"rs-4857239","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}