Placement of an aortic stent graft induces anatomical deformations, particularly at the iliac level, because of the introduction of rigid devices. Anticipating these changes may improve the accuracy of fusion guidance during endovascular procedures. In this prospective single-center study, a deep learning-based sizing software was used to generate a deformation-adjusted three-dimensional fusion mask, which was compared with a conventional, nondeformed mask using intraoperative angiography as the reference standard. In a cohort of 31 patients, the deformation-adjusted mask significantly improved the fusion accuracy for renal artery ostia positioning, iliac artery alignment, and guidewire localization, without any additional acquisition time. These findings suggest that deformation-aware fusion masks may enhance intraoperative navigation during aortic endovascular repair.

BACKGROUND:The aim of this study was to evaluate a new measurement tool for the preoperative computed tomography (CT) scan to estimate the endograft apposition surface on the proximal aortic neck of the aneurysm that would predict the risk of late type IA endoleak (EL1A). METHODS:From databases of 4 high-volume centers between 2004 and 2020, all patients who underwent Endovascular Aneuruysm Repair (EVAR) with the possibility of analyzing pre and post-EVAR scans and who presented late (>2 years) EL1A in follow-up were included retrospectively (EL1A group). A control group of randomly selected EL1A-free patients were included (control group) and followed-up beyond 4 years. The measurement tool (Endosize, Therenva) was used to model a virtual cylinder constrained by the centreline whose diameter was determined from the diameter of the implanted endograft. The apposition surface was calculated by determining the areas of the virtual cylinder mesh outside the mesh of the neighboring portion of the segmented aorta by an intersection calculation of surface normals. This apposition surface and the usual anatomical data of the proximal neck were compared between the groups with univariate and multivariate analyses. Intra and interobserver variability in the measurement of apposition surface was analyzed using the intraclass correlation coefficient (ICC) and the method of Bland and Altman. RESULTS:Sixty-six comparable patients were included as follows: 33 in the EL1A group and 33 in the control group. The mean time to EL1A onset was 50.2 ± 24.6 months in the EL1A group, and the mean follow-up time was 79.8 ± 26.8 months in the control group. In univariate analysis, only the endograft apposition surface ratio on the proximal neck was significantly lower in the EL1A group compared with the control group (50.3% vs. 75.5%; P < 0.0001). There was no significant difference in classic anatomical neck characteristics. There was no difference in proximal endograft oversizing or in the length from the lowest renal artery to the beginning of the covered portion of the endograft. In multivariate analysis, only the apposition surface ratio was independently correlated with the presence of EL1A (P < 0.0001). The estimation of the intra and interobserver variability indicates excellent reliability (ICC = 0.992; ICC = 0.956). CONCLUSIONS:The method estimating the preoperative endograft apposition surface on the proximal aortic neck appears to be a reliable measuring technique in predicting the risk of occurrence of late EL1A after EVAR.

To improve the prediction of renal ostia displacement in biomechanical finite element simulations of the vascular structure deformation during endovascular aneurysm repair (EVAR). An existing finite element model to compute the deformation of the vascular structure due to tools insertion during EVAR, previously validated against clinical data in terms of guidewire position, is confronted in terms of renal ostia displacement to clinical intraoperative data from 16 patients undergoing EVAR and experiencing significant ostia displacement during the procedure (average vertical displacement of 10.38 mm from the preoperative to intraoperative configurations). This yields an update of the mechanical support parametrization. A score quantifies the predictive performance of the existing and updated parametrizations for both the renal displacement and the iliac arteries deformation. The updated model demonstrates a significant improvement in predictive accuracy for renal ostia deviation during EVAR. The axial mean displacement error is improved from 7.41 mm (previous parametrization) to 2.99 mm (updated parametrization). The score shows that this new parametrization improves the predictive performance of the simulation for the renal ostia displacement without compromising the iliac deformations prediction. The updated parametrization significantly enhances the predictive capability for arterial deformations during EVAR. A better prediction of the renal ostia displacement can significantly improve surgical planning and intraoperative guidance.

Selective Internal Radio Therapy, is a treatment for hepatocellular carcinoma, consisting in the administration of radioactive microsphere inside the arterial blood-flow feeding the tumors [1]. It is an effective treatment that requires a precise planning to maximize the radioactive activity delivered to the tumor and minimize the toxicity to healthy structures. It is in this clinical context that we are developing a deep learning assisted method to infer a patient specific activity distribution from pre-interventional, diagnostic images, to assist the planning and patient selection for the SIRT. Two networks are presented in this study, one using a diagnostic scanner, guided by post-operative information, and a second, using only data derived from pre-operative imagery.

Renal impairment after endovascular treatment of abdominal aortic aneurysms (AAA) is multifactorial. Decreased renal vascularity is a possible factor, but it is not well reflected by creatinine clearance, whereas renal volume decreases with infusion failure. The purpose of this study was to see if there is a change in renal volume 24 months after exclusion of AAA and to compare fenestrated (FEVAR) and infrarenal (EVAR) stents.

Endovascular navigation tools are being increasingly used in the endovascular treatment of abdominal aortic aneurysms, either in hybrid rooms or using mobile systems. At the same time, there is growing awareness of the problems associated with radiation protection. The purpose of our study was to analyze the impact of the use of the Endonaut navigation system on radiation protection in the endovascular treatment of abdominal aortic aneurysms.

Catheterization of target arteries during placement of a fenestrated stentgraft (FEVAR) may be difficult and contributes to a significant radiation exposure during these procedures and the duration of the procedure. The objective of this study was to correlate the anatomical parameters of the target arteries with the procedure time.

Endovascular revascularization is becoming the established first-line treatment of peripheral artery disease (PAD). Ultrasound (US) imaging is used pre-operatively to make the first diagnosis and is often followed by a CT angiography (CTA). US provides a non-invasive and non-ionizing method for the visualization of arteries and lesion(s). This paper proposes to generate a 3D stretched reconstruction of the femoral artery from a sequence of 2D US B-mode frames. The proposed method is solely image-based. A Mask-RCNN is used to segment the femoral artery on the 2D US frames. In-plane registration is achieved by aligning the artery segmentation masks. Subsequently, a convolutional neural network (CNN) predicts the out-of-plane translation. After processing all input frames and re-sampling the volume according to the vessel’s centerline, the whole femoral artery can be visualized on a single slice of the resulting stretched view. 111 tracked US sequences of the left or right femoral arteries have been acquired on 18 healthy volunteers. fivefold cross-validation was used to validate our method and achieve an absolute mean error of 0.28 ± 0.28 mm and a median drift error of 8.98%. This study demonstrates the feasibility of freehand US stretched reconstruction following a deep learning strategy for imaging the femoral artery. Stretched views are generated and can give rich diagnosis information in the pre-operative planning of PAD procedures. This visualization could replace traditional 3D imaging in the pre-operative planning process, and during the pre-operative diagnosis phase, to identify, locate, and size stenosis/thrombosis lesions.

Background: Endovascular treatment has become the first-line strategy for peripheral arterial disease (PAD). Given the number of procedures required, any technology associated with a reduction in radiation exposure and contrast volume is highly relevant. In the present study, we evaluated whether two-dimensional (2D) fusion imaging could reduce the radiation exposure and contrast volume during endovascular treatment of occlusive PAD. Methods: Our consecutive, retrospective, single-center, nonrandomized comparative trial included patients with PAD at the femoral, popliteal, and/or tibial level, at any clinical stage, if they were candidates for endovascular revascularization. Patients were treated with or without the EndoNaut 2D fusion imaging system (Therenva, Rennes, France) in a nonhybrid room with the same Cios Alpha mobile C-arm (Siemens, Munich, Germany). The indirect dose-area product and contrast medium volume were recorded. Results: Between March 2018 and April 2020, 255 patients underwent endovascular femoropopliteal revascularization with (n = 124) or without (n = 131) 2D fusion imaging. The volume of injected contrast medium (34.7 +/- 13.8 mL vs 51.3 +/- 26.7 mL; P <.001) and dose-area product (8.9 +/- 9.9 Gy/cm(2) vs 13.5 +/- 14.0 Gy/cm(2); P=.003) were significantly lower for the 2D fusion imaging group than for the control group. A subgroup analysis of complex (TransAtlantic Inter-Society Consensus for the Management of Peripheral Arterial Disease C/D) lesions showed similar results. Stratification of the fusion imaging group into three subgroups, according to the procedure dates, showed no effect of a potential learning curve on the operative parameters. Conclusions: The results from the present study showed a significant reduction in the contrast volume and radiation dose for endovascular treatment of PAD when applying 2D fusion imaging technology. Overall, a reduction of >30% was observed for both operative parameters, without excessive training requirements, highlighting the potential benefits of using 2D fusion imaging when performing endovascular revascularization for PAD.

The aim of this study is to develop an automated deep-learning-based whole heart segmentation of ECG-gated computed tomography data. After 21 exclusions, CT acquired before transcatheter aortic valve implantation in 71 patients were reviewed and randomly split in a training (n = 55 patients), validation (n = 8 patients), and a test set (n = 8 patients). A fully automatic deep-learning method combining two convolutional neural networks performed segmentation of 10 cardiovascular structures, which was compared with the manually segmented reference by the Dice index. Correlations and agreement between myocardial volumes and mass were assessed. The algorithm demonstrated high accuracy (Dice score = 0.920; interquartile range: 0.906–0.925) and a low computing time (13.4 s, range 11.9–14.9). Correlations and agreement of volumes and mass were satisfactory for most structures. Six of ten structures were well segmented. Deep-learning-based method allowed automated WHS from ECG-gated CT data with a high accuracy. Challenges remain to improve right-sided structures segmentation and achieve daily clinical application.

Background: Fusion imaging makes it possible to improve endovascular procedures and is mainly used in hybrid rooms for aortic procedures. The objective of this study was to evaluate the feasibility of fusion imaging for femoropopliteal endovascular procedures with a mobile flat plane sensor and dedicated software to assist endovascular navigation. Materials and methods: Between May and December 2017, 41 patients requiring femoropopliteal endovascular revascularization were included. Interventions were carried out in a conventional surgical room equipped with a mobile plane sensor (Cios Alpha, Siemens). The numerical video stream was transmitted to an angionavigation station (EndoNaut (EN), Therenva). The software created an osseous and arterial panorama of the treated limb from the angiographies carried out at the beginning of procedure. After each displacement of the table, the software relocated the current image on the osseous panorama, with 2D-2D resetting, and amalgamated the mask of the arterial panorama. The success rates of creation of osseous and arterial panorama and the success of relocation were evaluated. The data concerning irradiation, the volume of contrast (VC) injected, and operative times were recorded. Results: Osseous panoramas could be automatically generated for the 41 procedures, without manual adjustment in 33 cases (80.5%). About 35 relocations based on a 2D-2D resetting could be obtained in the 41 procedures, with a success rate of 85%. The causes of failure were a change in table height or arch angulation. The average duration of intervention was 74.5 min. The irradiation parameters were duration of fluoroscopy 17.8 +/- 13.1 min, air kerma 80.5 +/- 68.4 mGy, and dose area product 2140 +/- 1599 mu Gy m(2). The average VC was 24.5 +/- 14 mL. Conclusions: This preliminary study showed that fusion imaging is possible in a nonhybrid room for peripheral procedures. Imagery of mobile C-arms can be improved for femoropopliteal endovascular procedures without heavy equipment. These imagery tools bring an operative comfort and could probably reduce irradiation and the injected VC. The clinical benefit must be evaluated in more patients in a randomized comparative study with a rigorous methodology.

Abstract Introduction Intra-arterial therapy is an effective way of performing chemotherapy or radiation therapy in patients with primary liver cancer (i.e. hepatocellular carcinoma). Although this minimally invasive approach is now an established treatment option, support tools for pre-operative planning and intra-operative assistance might be helpful. Material and methods We developed an approach for semi-automatic segmentation of computed tomography angiography images of the main arterial branches (required for access path to the treatment site), automatic segmentation of the liver, arterial and venous tree, and interactive segmentation of the tumors (required for procedure-specific planning). This approach was then integrated into a liver-specific workflow within EndoSize® solution, a planning software for endovascular procedures. The main branches extraction approach was qualitatively evaluated inside the software, while the automatic segmentation methods were quantitatively assessed. Results Main branches extraction provides a success rate of 85% (i.e. all arteries correctly extracted) in a dataset of 172 patients. On public databases, a mean DICE of 0.91, 0.47 and 0.92 was obtained for liver, venous and arterial trees segmentation, respectively. Conclusions This pipeline is suitable for directly accessing the treatment site, giving anatomic measurements, and visualizing the hepatic trees, liver, and surrounding arteries during the pre-operative planning. Abbreviations HCC: hepatocellular carcinoma; TACE: transarterial chemoembolization; SIRT: selective internal radiation therapy; CT: computed tomography; CTA: computed tomography angiography; AMS: superior mesenteric artery; LGA: left gastric artery; RHA: right hepatic artery; LHA: left hepatic artery; rbHA: right branch of the hepatic artery; lbHA: left branch of the hepatic artery; GDA: gastroduodenal artery; VOI: volume of interest; SD: standard deviation; MICCAI: medical image computing and computer assisted interventions; MR: magnetic resonance

Purpose The fusion of pre/intraoperative images may improve catheter manipulation during radioembolization (RE) interventions by adding relevant information. The objective of this work is to propose and evaluate the performance of a RE guidance strategy relying on structure-driven intensity-based registration between preoperative CTA and intraoperative X-ray images. Methods The navigation strategy is decomposed into three image fusion steps, supporting the catheter navigation from the femoral artery till reaching the injection site (IS). During the pretreatment assessment intervention, the aorta and the origins of its side branches are projected on the intraoperative 2D fluoroscopy following a 3D/2D bone-based registration process, to assist the celiac trunk access. Subsequently, a similar approach consisting in projecting the hepatic vasculature on intraoperative DSA through 3D/2D vessel-based registration is performed to assist the IS location. Lastly, the selected IS is reproduced during the treatment intervention by employing 2D/2D image-based registration between pretreatment and treatment fluoroscopic images. Results The three fusion steps were independently evaluated on subsets of 20, 19 and 5 patient cases, respectively. Best results were obtained with gradient difference as similarity measure and with a delimited preoperative vascular structure for vessel-based registration. The approach resulted in qualitatively appropriate anatomical correspondences when projecting the preoperative structures on intraoperative images. With the best configuration, the registration steps showed accuracy and feasibility in aligning data, with global mean landmarks errors of 1.59 mm, 2.32 mm and 2.17 mm, respectively, a computation time that never exceeded 5 s, 25 s and 11 s, respectively, and a user interaction limited to manual initialization of the 3D/2D registration. Conclusion An image fusion-based approach has been specifically proposed for RE procedures guidance. The catheter manipulation strategy based on the fusion of pre- and intraoperative images has the potential to support different steps of the RE clinical workflow and to guide the overall procedure.

In a clinical decision support system, the purpose of case-based reasoning is to help clinicians make convenient decisions for diagnoses or interventional gestures. Past experience, which is represented by a case-base of previous patients, is exploited to solve similar current problems using four steps-retrieve, reuse, revise, and retain. The proposed case-based reasoning has been focused on transcatheter aortic valve implantation to respond to clinical issues pertaining vascular access and prosthesis choices. The computation of a relevant similarity measure is an essential processing step employed to obtain a set of retrieved cases from a case-base. A hierarchical similarity measure that is based on a clinical decision tree is proposed to better integrate the clinical knowledge, especially in terms of case representation, case selection and attributes weighting. A case-base of 138 patients is used to evaluate the case-based reasoning performance, and retrieve- and reuse-based criteria have been considered. The sensitivity for the vascular access and the prosthesis choice is found to 0.88 and 0.94, respectively, with the use of the hierarchical similarity measure as opposed to 0.53 and 0.79 for the standard similarity measure. Ninety percent of the suggested solutions are correctly classified for the proposed metric when four cases are retrieved. Using a dedicated similarity measure, with relevant and weighted attributes selected through a clinical decision tree, the set of retrieved cases, and consequently, the decision suggested by the case-based reasoning are substantially improved over state-of-the-art similarity measures.

Purpose: Minimally invasive trans-catheter aortic valve implantation (TAVI) has emerged as a treatment of choice for high-risk patients with severe aortic stenosis. However, the planning of TAVI procedures would greatly benefit from automation to speed up, secure and guide the deployment of the prosthetic valve. We propose a hybrid approach allowing the computation of relevant anatomical measurements along with an enhanced visualization. Material and methods: After an initial step of centerline detection and aorta segmentation, model-based and statistical-based methods are used in combination with 3 D active contour models to exploit the complementary aspects of these methods and automatically detect aortic leaflets and coronary ostia locations. Important anatomical measurements are then derived from these landmarks. Results: A validation on 50 patients showed good precision with respect to expert sizing for the ascending aorta diameter calculation (2.2 +/- 2.1 mm), the annulus diameter (1.31 +/- 0.75 mm), and both the right and left coronary ostia detection (1.96 +/- 0.87 mm and 1.80 +/- 0.74 mm, respectively). The visualization is enhanced thanks to the aorta and aortic root segmentation, the latter showing good agreement with manual expert delineation (Jaccard index: 0.96 +/- 0.03). Conclusion: This pipeline is promising and could greatly facilitate TAVI planning. [GRAPHICS] .

Background: During endovascular aneurysm repair (EVAR), complex iliac anatomy is a source of complications such as unintentional coverage of the hypogastric artery. The aim of our study was to evaluate ability to predict coverage of the hypogastric artery using a biomechanical model simulating arterial deformations caused by the delivery system. Methods: The biomechanical model of deformation has been validated by many publications. The simulations were performed on 38 patients included retrospectively, for a total of 75 iliac arteries used for the study. On the basis of objective measurements, two groups were formed: one with "complex'' iliac anatomy (n = 38 iliac arteries) and the other with "simple'' iliac anatomy (n = 37 iliac arteries). The simulation enabled measurement of the lengths of the aorta and the iliac arteries once deformed by the device. Coverage of the hypogastric artery was predicted if the deformed renal/iliac bifurcation length (L-pre) was less than the length of the implanted device (L-stent-measured on the postoperative computed tomography [CT]) and nondeformed L-pre was greater than L-stent. Results: Nine (12%) internal iliac arteries were covered unintentionally. Of the coverage attributed to perioperative deformations, 1 case (1.3%) occurred with simple anatomy and 6 (8.0%) with complex anatomy (P = 0.25). All cases of unintentional coverage were predicted by the simulation. The simulation predicted hypogastric coverage in 35 cases (46.7%). There were therefore 26 (34.6%) false positives. The simulation had a sensitivity of 100% and a specificity of 60.6%. On multivariate analysis, the factors significantly predictive of coverage were the iliac tortuosity index (P = 0.02) and the predicted margin between the termination of the graft limb and the origin of the hypogastric artery in nondeformed (P = 0.009) and deformed (P = 0.001) anatomy. Conclusions: Numerical simulation is a sensitive tool for predicting the risk of hypogastric coverage during EVAR and allows more precise preoperative sizing. Its specificity is liable to be improved by using a larger cohort.

PURPOSE:To quantify the displacement of the vascular structures after insertion of stiff devices during endovascular aneurysm repair (EVAR) of abdominal aortic aneurysm and to identify potential parameters influencing this displacement. MATERIALS AND METHODS:A total of 50 patients from a single center undergoing EVAR were prospectively enrolled between January 2016 and December 2017. Fusion imaging was employed using the EndoNaut (Therenva, Rennnes, France) station through a 3-dimensional (3D)/2-dimensional (2D) technology synchronizing the 3D computed tomography scan to the live intraoperative fluoroscopy. The accuracy of the fusion roadmap was evaluated before deployment by conventional digital subtraction angiogram on a single plane (with different C-arm incidences). RESULTS:The mean displacement error of the ostium of the lowest renal artery was 4.1 ± 2.4 mm (range, 0-11.7 mm), with a left/right displacement of 1.6 ± 1.7 mm (range, 0-6.9 mm) and a craniocaudal displacement of 3.5 ± 2.4 mm (range, 0-11.3 mm). The correction required for the ostium of the lower renal artery was mostly cranial and to the left. Multiple linear regression analysis revealed only the sharpest angle between the aneurysm neck and sac as the factor influencing the accuracy of fusion imaging. All other parameters did not show any correlation. CONCLUSIONS:This study identified the sources of fusion error after insertion of rigid material during EVAR. As the sharpest angulation between aneurysm neck and sac increases, the overall accuracy of the fusion might be affected.

Background: Fusion imaging is a technique that facilitates endovascular navigation but is only available in hybrid rooms. The goal of this study was to evaluate the feasibility of fusion imaging with a mobile C-arm in a conventional operating room through the use of an angionavigation station. Methods: From May 2016 to June 2017, the study included all patients who underwent an aortic stent graft procedure in a conventional operating room with a mobile flat-panel detector (Cios Alpha, Siemens) connected to an angionavigation station (EndoNaut, Therenva). The intention was to perform preoperative 3D computerized tomography/perioperative 2D fluoroscopy fusion imaging using an automatic registration process. Registration was considered successful when the software was able to correctly overlay preoperative 3D vascular structures onto the fluoroscopy image. For EVAR, contrast dose, operation time, and fluoroscopy time (FT) were compared with those of a control group drawn from the department's database who underwent a procedure with a C-arm image intensifier. Results: The study included 54 patients, and the procedures performed were 49 EVAR, 2 TEVAR, 2 IBD, and 1 FEVAR. Of the 178 registrations that were initialized, it was possible to use the fusion imaging in 170 cases, that is, a 95.5% success rate. In the EVAR comparison, there were no difference with the control group (n = 103) for FT (21.9 +/- 12 vs. 19.5 +/- 13 min; P = 0.27), but less contrast agent was used in the group undergoing a procedure with the angionavigation station (42.3 +/- 22 mL vs. 81.2 +/- 48 mL; P < 0.001), and operation time was shorter (114 +/- 44 vs. 140.8 +/- 38 min; P < 0.0001). Conclusions: Fusion imaging is feasible with a mobile C-arm in a conventional operating room and thus represents an alternative to hybrid rooms. Its clinical benefits should be evaluated in a randomized series, but our study already suggests that EVAR procedures might be facilitated with an angionavigation system.

Interventional endovascular treatment has become the first line of management in the treatment of peripheral artery disease (PAD). However, contrast and radiation exposure continue to limit the feasibility of these procedures. This paper presents a novel hybrid image fusion system for endovascular intervention of PAD. We present two different roadmapping methods from intra- and pre-interventional imaging that can be used either simultaneously or independently, constituting the navigation system.