Morphology of Abdominal Aortic Aneurysms and Correlation with Biomechanical Tests of Aneurysmal Wall Fragments.

BACKGROUND: Evaluate how specific morphologic aspects of abdominal aortic aneurysms (AAAs), including asymmetries, curvatures, tortuosities, and angulations, among others can influence the intrinsic biomechanical properties of the AAA's wall. This study analyzed the correlation of geometric measurements (1-dimensional, 2-dimensional, and 3-dimensional) of preoperative tomographic images of AAA with uniaxial biomechanical tests of the arterial wall fragments of these AAA obtained in open surgical repair of aneurysms. METHODS: It was a multicenter, experimental, and observational study, and initially 54 individuals were selected who underwent open surgical of AAA, with valid biomechanical tests of the anterior wall of the AAA. Seven individuals were excluded because they had poor preoperative quality computed tomography scans and/or artifacts that impeded image segmentation and extraction of AAA geometric indices. The aortic fragments were subjected to uniaxial biomechanical destructive tests to obtain the following data: maximum load, failure stress, failure tension, failure strain energy, strain, and fragment thickness. In the same patients, preoperative computed tomography scans were performed with the extraction of 26 geometric indices, subdivided into 9 1-dimensional indices, 6 2-dimensional indices, and 11 3-dimensional indices. Data were subjected to statistical analysis using SPSS version 28. RESULTS: Comparing ruptured and unruptured AAA, no statistical difference was observed between the biomechanical and geometric parameters. The fragment thickness of the ruptured AAA was lower than that of the unruptured AAA (P < 0.05). By comparing tomographic geometric indices and biomechanical parameters of the aortic fragments using Pearson's coefficient, positive and linear correlations (P < 0.05) were observed between the geometric variable maximum diameter (Dmax) of the AAA with maximum load (r = 0.408), failure tension (r = 0.372), and failure stress (r = 0.360). Positive and linear correlations were also observed between the variable diameter/height ratio (DHr) and the maximum load (r = 0.360), failure tension (r = 0.354), and failure stress (r = 0.289). The geometric variable DHr was dependent and correlated with Dmax. Simple regression analysis showed that R2 varied between 8.3% and 16.7%, and all models were significant (P < 0.05). CONCLUSIONS: Dmax and DHr were linearly and positively correlated with the resistance parameters (maximum load, failure tension, and failure stress) of the AAA fragments. The DHr variable is dependent and correlated with Dmax. There was no correlation between the other geometric indices and the biomechanical parameters of the AAA wall. The asymmetries did not globally influence the biomechanics of AAA wall; however, they may influence regionally. Larger AAAs were stronger than smaller ones. Therefore, such findings may point toward Dmax is still the main geometric parameter, which influences the anterior wall, and possibly globally in the AAA.

CT analysis of aortic calcifications to predict abdominal aortic aneurysm rupture.

BACKGROUND: Abdominal aortic aneurysm (AAA) rupture prediction based on sex and diameter could be improved. The goal was to assess whether aortic calcification distribution could better predict AAA rupture through machine learning and LASSO regression. METHODOLOGY: In this retrospective study, 80 patients treated for a ruptured AAA between January 2001 and August 2018 were matched with 80 non-ruptured patients based on maximal AAA diameter, age, and sex. Calcification volume and dispersion, morphologic, and clinical variables were compared between both groups using a univariable analysis with p = 0.05 and multivariable analysis through machine learning and LASSO regression. We used AUC for machine learning and odds ratios for regression to measure performance. RESULTS: Mean age of patients was 74.0 ± 8.4 years and 89% were men. AAA diameters were equivalent in both groups (80.9 ± 17.5 vs 79.0 ± 17.3 mm, p = 0.505). Ruptured aneurysms contained a smaller number of calcification aggregates (18.0 ± 17.9 vs 25.6 ± 18.9, p = 0.010) and were less likely to have a proximal neck (45.0% vs 76.3%, p < 0.001). In the machine learning analysis, 5 variables were associated to AAA rupture: proximal neck, antiplatelet use, calcification number, Euclidian distance between calcifications, and standard deviation of the Euclidian distance. A follow-up LASSO regression was concomitant with the findings of the machine learning analysis regarding calcification dispersion but discordant on calcification number. CONCLUSION: There might be more to AAA calcifications that what is known in the present literature. We need larger prospective studies to investigate if indeed, calcification dispersion affects rupture risk. CLINICAL RELEVANCE STATEMENT: Ruptured aneurysms are possibly more likely to have their calcification volume concentrated in a smaller geographical area. KEY POINTS: • Abdominal aortic aneurysm (AAA) rupture prediction based on sex and diameter could be improved. • For a given calcification volume, AAAs with well-distributed calcification clusters could be less likely to rupture. • A machine learning model including AAA calcifications better predicts rupture compared to a model based solely on maximal diameter and sex alone, although it might be prone to overfitting.

Source of errors and accuracy of a two-dimensional/three-dimensional fusion road map for endovascular aneurysm repair of abdominal aortic aneurysm.

PURPOSE: To evaluate the accuracy and source of errors using a two-dimensional (2D)/three-dimensional (3D) fusion road map for endovascular aneurysm repair (EVAR) of abdominal aortic aneurysm. MATERIALS AND METHODS: A rigid 2D/3D road map was tested in 16 patients undergoing EVAR. After 3D/3D manual registration of preoperative multidetector computed tomography (CT) and cone beam CT, abdominal aortic aneurysm outlines were overlaid on live fluoroscopy/digital subtraction angiography (DSA). Patient motion was evaluated using bone landmarks. The misregistration of renal and internal iliac arteries were estimated by 3 readers along head-feet and right-left coordinates (z-axis and x-axis, respectively) before and after bone and DSA corrections centered on the lowest renal artery. Iliac deformation was evaluated by comparing centerlines before and during intervention. A score of clinical added value was estimated as high (z-axis < 3 mm), good (3 mm ≤ z-axis ≤ 5 mm), and low (z-axis > 5 mm). Interobserver reproducibility was calculated by the intraclass correlation coefficient. RESULTS: The lowest renal artery misregistration was estimated at x-axis = 10.6 mm ± 11.1 and z-axis = 7.4 mm ± 5.3 before correction and at x-axis = 3.5 mm ± 2.5 and z-axis = 4.6 mm ± 3.7 after bone correction (P = .08), and at 0 after DSA correction (P < .001). After DSA correction, residual misregistration on the contralateral renal artery was estimated at x-axis = 2.4 mm ± 2.0 and z-axis = 2.2 mm ± 2.0. Score of clinical added value was low (n = 11), good (n= 0), and high (n= 5) before correction and low (n = 5), good (n = 4), and high (n = 7) after bone correction. Interobserver intraclass correlation coefficient for misregistration measurements was estimated at 0.99. Patient motion before stent graft delivery was estimated at x-axis = 8 mm ± 5.8 and z-axis = 3.0 mm ± 2.7. The internal iliac artery misregistration measurements were estimated at x-axis = 6.1 mm ± 3.5 and z-axis = 5.6 mm ± 4.0, and iliac centerline deformation was estimated at 38.3 mm ± 15.6. CONCLUSIONS: Rigid registration is feasible and fairly accurate. Only a partial reduction of vascular misregistration was observed after bone correction; minimal DSA acquisition is still required.

Virtual and analytical self-expandable braided stent treatment models.

The Flow Diverter is a self-expandable braided stent that has helped improve the effectiveness of cerebral aneurysm treatment during the last decade. The Flow Diverter's efficiency heavily relies on proper decision-making during the pre-operative phase, which is currently based on static measurements that fail to account for vessel or tissue deformation. In the context of providing realistic measurements, a biomechanical computational method is designed to aid physicians in predicting patient-specific treatment outcomes. The method integrates virtual and analytical treatment models, validated against experimental mechanical tests, and two patient treatment outcomes. In the case of both patients, deployed stent length was one of the validated result parameters, which displayed an error inferior to 1.5% for the virtual and analytical models. These results indicated both models' accuracy. However, the analytical model provided more accurate results with a 0.3% error while requiring a lower computational cost for length prediction. This computational method can offer designing and testing platforms for predicting possible intervention-related complications, patient-specific medical device designs, and pre-operative planning to automate interventional procedures.

Human-scale navigation of magnetic microrobots in hepatic arteries.

Using external actuation sources to navigate untethered drug-eluting microrobots in the bloodstream offers great promise in improving the selectivity of drug delivery, especially in oncology, but the current field forces are difficult to maintain with enough strength inside the human body (>70-centimeter-diameter range) to achieve this operation. Here, we present an algorithm to predict the optimal patient position with respect to gravity during endovascular microrobot navigation. Magnetic resonance navigation, using magnetic field gradients in clinical magnetic resonance imaging (MRI), is combined with the algorithm to improve the targeting efficiency of magnetic microrobots (MMRs). Using a dedicated microparticle injector, a high-precision MRI-compatible balloon inflation system, and a clinical MRI, MMRs were successfully steered into targeted lobes via the hepatic arteries of living pigs. The distribution ratio of the microrobots (roughly 2000 MMRs per pig) in the right liver lobe increased from 47.7 to 86.4% and increased in the left lobe from 52.2 to 84.1%. After passing through multiple vascular bifurcations, the number of MMRs reaching four different target liver lobes had a 1.7- to 2.6-fold increase in the navigation groups compared with the control group. Performing simulations on 19 patients with hepatocellular carcinoma (HCC) demonstrated that the proposed technique can meet the need for hepatic embolization in patients with HCC. Our technology offers selectable direction for actuator-based navigation of microrobots at the human scale.

Design of a Low-Cost, Self-Adaptive and MRI-Compatible Cardiac Gating System.

OBJECTIVE: Cardiac gating, synchronizing medical scans with cardiac activity, is widely used to make quantitative measurements of physiological events and to obtain high-quality scans free of pulsatile artefacts. This can provide important information for disease diagnosis, targeted control of medical microrobots, etc. The current work proposes a low-cost, self-adaptive, MRI-compatible cardiac gating system. METHOD: The system and its processing algorithm, based on the monitoring and analysis of blood pressure waveforms, are proposed. The system is tested in an in vitro experiment and two living pigs using four-dimensional (4D) flow magnetic resonance imaging (MRI) and two-dimensional phase-contrast (2D-PC) sequences. RESULTS: in vitro and in vivo experiments reveal that the proposed system can provide stable cardiac synchronicity, has good MRI compatibility, and can cope with the fringe magnetic field of the MRI scanner, radiofrequency signals during image acquisition, and heart rate changes. High-resolution 4D flow imaging is successfully acquired both in vivo and in vitro. The difference between the 2D and 4D measurements is ≤ 21%. The incidence of false triggers is 0% in all tests, which is unattainable for other known cardiac gating methods. CONCLUSION: The system has good MRI compatibility and can provide a stable and accurate trigger signal based on pressure waveform. It opens the door to applications where the previous gating methods were difficult to implement or not applicable.

Design of a Patient-Specific Respiratory-Motion-Simulating Platform for In Vitro 4D Flow MRI.

Four-dimensional (4D) flow magnetic resonance imaging (MRI) is a leading-edge imaging technique and has numerous medicinal applications. In vitro 4D flow MRI can offer some advantages over in vivo ones, especially in accurately controlling flow rate (gold standard), removing patient and user-specific variations, and minimizing animal testing. Here, a complete testing method and a respiratory-motion-simulating platform are proposed for in vitro validation of 4D flow MRI. A silicon phantom based on the hepatic arteries of a living pig is made. Under the free-breathing, a human volunteer's liver motion (inferior-superior direction) is tracked using a pencil-beam MRI navigator and is extracted and converted into velocity-distance pairs to program the respiratory-motion-simulating platform. With the magnitude displacement of about 1.3 cm, the difference between the motions obtained from the volunteer and our platform is ≤ 1 mm which is within the positioning error of the MRI navigator. The influence of the platform on the MRI signal-to-noise ratio can be eliminated even if the actuator is placed in the MRI room. The 4D flow measurement errors are respectively 0.4% (stationary phantom), 9.4% (gating window = 3 mm), 27.3% (gating window = 4 mm) and 33.1% (gating window = 7 mm). The vessel resolutions decreased with the increase of the gating window. The low-cost simulation system, assembled from commercially available components, is easy to be duplicated.

Multiband sensor using thick holographic gratings for sulfur detection by laser-induced breakdown spectroscopy.

Detection of sulfur by optical emission spectroscopy generally presents some difficulties because the strongest lines are in the vacuum UV below 185 nm and therefore are readily absorbed by oxygen molecules in air. A novel concept for a low-cost and efficient system to detect sulfur using near-IR bands by laser-induced breakdown spectroscopy is here proposed. This concept is based on customized thick holographic gratings as spectral filtering elements. The signal integration and the temporal synchronization are performed using built-in custom electronics that amplify and integrate or trigger photodiode output signals. In this work, we use the near-IR lines at 921.287 nm and a background reference at 900 nm. Preliminary results show a limit of detection comparable to that of a conventional high-end system.

Impact of calcification modeling to improve image fusion accuracy for endovascular aortic aneurysm repair.

Since the 1990s, endovascular aortic aneurysm repair (EVAR) has become a common alternative to open surgery for the treatment of abdominal aortic aneurysms (AAAs). To aid the deployment of stent-grafts, fluoroscopic image guidance can be enhanced using preoperative simulation and intraoperative image fusion techniques. However, the impact of calcification (Ca) presence on the guidance accuracy of such techniques is yet to be considered. In the present work, we introduce a guidance tool that accounts for patient-specific Ca presence. Numerical simulations of EVAR were developed for 12 elective AAA patients, both with (With-Ca) and without (No-Ca) Ca consideration. To assess the accuracy of the simulations, the image results were overlaid on corresponding intraoperative images and the overlay error was measured at selected anatomical landmarks. With this approach we gained insight into the impact of Ca presence on image fusion accuracy. Inclusion of Ca improved mean image fusion accuracy by 8.68 ± 4.59%. In addition, a positive correlation between the relative Ca presence and the image fusion accuracy was found (R = .753, p < .005). Our results suggest that considering Ca presence in patient-specific EVAR simulations increases the reliability of EVAR image guidance techniques that utilize numerical simulation, especially for patients with severe aortic Ca presence.

Quantification and 3D Localization of Magnetically Navigated Superparamagnetic Particles Using MRI in Phantom and Swine Chemoembolization Models.

OBJECTIVE: Superparamagnetic nanoparticles (SPIONs) can be combined with tumor chemoembolization agents to form magnetic drug-eluting beads (MDEBs), which are navigated magnetically in the MRI scanner through the vascular system. We aim to develop a method to accurately quantify and localize these particles and to validate the method in phantoms and swine models. METHODS: MDEBs were made of Fe3O4 SPIONs. After injected known numbers of MDEBs, susceptibility artifacts in three-dimensional (3D) volumetric interpolated breath-hold examination (VIBE) sequences were acquired in glass and Polyvinyl alcohol (PVA) phantoms, and two living swine. Image processing of VIBE images provided the volume relationship between MDEBs and their artifact at different VIBE acquisitions and post-processing parameters. Simulated hepatic-artery embolization was performed in vivo with an MRI-conditional magnetic-injection system, using the volume relationship to locate and quantify MDEB distribution. RESULTS: Individual MDEBs were spatially identified, and their artifacts quantified, showing no correlation with magnetic-field orientation or sequence bandwidth, but exhibiting a relationship with echo time and providing a linear volume relationship. Two MDEB aggregates were magnetically steered into desired liver regions while the other 19 had no steering, and 25 aggregates were injected into another swine without steering. The MDEBs were spatially identified and the volume relationship showed accuracy in assessing the number of the MDEBs, with small errors (≤ 8.8%). CONCLUSION AND SIGNIFICANCE: MDEBs were able to be steered into desired body regions and then localized using 3D VIBE sequences. The resulting volume relationship was linear, robust, and allowed for quantitative analysis of the MDEB distribution.

Navigation of Microrobots by MRI: Impact of Gravitational, Friction and Thrust Forces on Steering Success.

INTRODUCTION: Magnetic resonance navigation (MRN) uses MRI gradients to steer magnetic drug-eluting beads (MDEBs) across vascular bifurcations. We aim to experimentally verify our theoretical forces balance model (gravitational, thrust, friction, buoyant and gradient steering forces) to improve the MRN targeted success rate. METHOD: A single-bifurcation phantom (3 mm inner diameter) made of poly-vinyl alcohol was connected to a cardiac pump at 0.8 mL/s, 60 beats/minutes with a glycerol solution to reproduce the viscosity of blood. MDEB aggregates (25 ± 6 particles, 200 [Formula: see text]) were released into the main branch through a 5F catheter. The phantom was tilted horizontally from - 10° to +25° to evaluate the MRN performance. RESULTS: The gravitational force was equivalent to 71.85 mT/m in a 3T MRI. The gradient duration and amplitude had a power relationship (amplitude=78.717 [Formula: see text]). It was possible, in 15° elevated vascular branches, to steer 87% of injected aggregates if two MRI gradients are simultaneously activated ([Formula: see text] = +26.5 mT/m, [Formula: see text]= +18 mT/m for 57% duty cycle), the flow velocity was minimized to 8 cm/s and a residual pulsatile flow to minimize the force of friction. CONCLUSION: Our experimental model can determine the maximum elevation angle MRN can perform in a single-bifurcation phantom simulating in vivo conditions.

Anthropomorphic and biomechanical mockup for abdominal aortic aneurysm.

Abdominal aortic aneurysm (AAA) is an asymptomatic condition due to the dilation of abdominal aorta along with progressive wall degeneration, where rupture of AAA is life-threatening. Failures of AAA endovascular repair (EVAR) reflect our inadequate knowledge about the complex interaction between the aortic wall and medical devices. In this regard, we are presenting a hydrogel-based anthropomorphic mockup (AMM) to better understand the biomechanical constraints during EVAR. By adjusting the cryogenic treatments, we tailored the hydrogel to mimic the mechanical behavior of human AAA wall, thrombus and abdominal fat. A specific molding sequence and a pressurizing system were designed to reproduce the geometrical and diseased characteristics of AAA. A mechanically, anatomically and pathologically realistic AMM for AAA was developed for the first time, EVAR experiments were then performed with and without the surrounding fat. Substantial displacements of the aortic centerlines and vessel expansion were observed in the case without surrounding fat, revealing an essential framework created by the surrounding fat to account for the interactions with medical devices. In conclusion, the importance to consider surrounding tissue for the global deformation of AAA during EVAR was highlighted. Furthermore, potential use of this AMM for medical training was also suggested.

A Cohort Longitudinal Study Identifies Morphology and Hemodynamics Predictors of Abdominal Aortic Aneurysm Growth.

Abdominal aortic aneurysms (AAA) are localized, commonly occurring aortic dilations. Following rupture only immediate treatment can prevent morbidity and mortality. AAA maximal diameter and growth are the current metrics to evaluate the associated risk and plan intervention. Although these criteria alone lack patient specificity, predicting their evolution would improve clinical decision. If the disease is known to be associated with altered morphology and blood flow, intraluminal thrombus deposit and clinical symptoms, the growth mechanisms are yet to be fully understood. In this retrospective longitudinal study of 138 scans, morphological analysis and blood flow simulations for 32 patients with clinically diagnosed AAAs and several follow-up CT-scans, are performed and compared to 9 control subjects. Several metrics stratify patients between healthy, low and high risk groups. Local correlations between hemodynamic metrics and AAA growth are also explored but due to their high inter-patient variability, do not explain AAA heterogeneous growth. Finally, high-risk predictors trained with successively clinical, morphological, hemodynamic and all data, and their link to the AAA evolution are built from supervise learning. Predictive performance is high for morphological, hemodynamic and all data, in contrast to clinical data. The morphology-based predictor exhibits an interesting effort-predictability tradeoff to be validated for clinical translation.

Interstitial imaging with multiple diffusive reflectance spectroscopy projections for in vivo blood vessels detection during brain needle biopsy procedures.

Blood vessel injury during image-guided brain biopsy poses a risk of hemorrhage. Approaches that reduce this risk may minimize related patient morbidity. We present here an intraoperative imaging device that has the potential to detect the brain vasculature in situ. The device uses multiple diffuse reflectance spectra acquired in an outward-viewing geometry to detect intravascular hemoglobin, enabling the construction of an optical image in the vicinity of the biopsy needle revealing the proximity to blood vessels. This optical detection system seamlessly integrates into a commercial biopsy system without disrupting the neurosurgical clinical workflow. Using diffusive brain tissue phantoms, we show that this device can detect 0.5-mm diameter absorptive carbon rods up to ∼ 2 mm from the biopsy window. We also demonstrate feasibility and practicality of the technique in a clinical environment to detect brain vasculature in an in vivo model system. In situ brain vascular detection may add a layer of safety to image-guided biopsies and minimize patient morbidity.

A Numerical Preoperative Planning Model to Predict Arterial Deformations in Endovascular Aortic Aneurysm Repair.

Endovascular aneurysm repair is rapidly emerging as the primary preferred method for treating abdominal aortic aneurysm. In this image-guided interventional procedure, to obtain the roadmap and decrease contrast injections, preoperative CT images are overlaid onto live fluoroscopy images using various 2D/3D image fusion techniques. However, the structural changes due to the insertion of stiff tools degrade the fusion accuracy. To correct the mismatch and quantify the intraoperative deformations, we present a patient-specific biomechanical model of the aorto-iliac structure and its surrounding tissues. The predictive capability of the model was evaluated against intraoperative data for a group of four patients. Incorporating the perivascular tissues into the model significantly improved the results and the mean distance between the real and simulated endovascular tools was 2.99 ± 1.78 mm on the ipsilateral side and 4.59 ± 3.25 mm on the contralateral side. Moreover, the distance between the deformed iliac ostia and their corresponding landmarks on intraoperative images was 2.99 ± 2.48 mm.

Flow stagnation volume and abdominal aortic aneurysm growth: Insights from patient-specific computational flow dynamics of Lagrangian-coherent structures.

Abdominal aortic aneurysms (AAA) are localized, commonly-occurring dilations of the aorta. When equilibrium between blood pressure (loading) and wall mechanical resistance is lost, rupture ensues, and patient death follows, if not treated immediately. Experimental and numerical analyses of flow patterns in arteries show direct correlations between wall shear stress and wall mechano-adaptation with the development of zones prone to thrombus formation. For further insights into AAA flow topology/growth interaction, a workout of patient-specific computational flow dynamics (CFD) is proposed to compute finite-time Lyapunov exponents and extract Lagrangian-coherent structures (LCS). This computational model was first compared with 4-D phase-contrast magnetic resonance imaging (MRI) in 5 patients. To better understand the impact of flow topology and transport on AAA growth, hyperbolic, repelling LCS were computed in 1 patient during 8-year follow-up, including 9 volumetric morphologic AAA measures by computed tomography-angiography (CTA). LCS defined barriers to Lagrangian jet cores entering AAA. Domains enclosed between LCS and the aortic wall were considered to be stagnation zones. Their evolution was studied during AAA growth. Good correlation - 2-D cross-correlation coefficients of 0.65, 0.86 and 0.082 (min, max, SD) - was obtained between numerical simulations and 4-D MRI acquisitions in 6 specific cross-sections from 4 patients. In follow-up study, LCS divided AAA lumens into 3 dynamically-isolated zones: 2 stagnation volumes lying in dilated portions of the AAA, and circulating volume connecting the inlet to the outlet. The volume of each zone was tracked over time. Although circulating volume remained unchanged during 8-year follow-up, the AAA lumen and main stagnation zones grew significantly (8 cm3/year and 6 cm3/year, respectively). This study reveals that transient transport topology can be quantified in patient-specific AAA during disease progression by CTA, in parallel with lumen morphology. It is anticipated that analysis of the main AAA stagnation zones by patient-specific CFD on a yearly basis could help to predict AAA growth and rupture.

Automatic detection of selective arterial devices for advanced visualization during abdominal aortic aneurysm endovascular repair.

Here we address the automatic segmentation of endovascular devices used in the endovascular repair (EVAR) of abdominal aortic aneurysms (AAA) that deform vascular tissues. Using this approach, the vascular structure is automatically reshaped solving the issue of misregistration observed on 2D/3D image fusion for EVAR guidance. The endovascular devices we considered are the graduated pigtail catheter (PC) used for contrast injection and the stent-graft delivery device (DD). The segmentation of the DD was enhanced using an asymmetric Frangi filter. The segmented geometries were then analysed using their specific features to remove artefacts. The radiopaque markers of the PC were enhanced using a fusion of Hessian and newly introduced gradient norm shift filters. Extensive experiments were performed using a database of images taken during 28 AAA-EVAR interventions. This dataset was divided into two parts: the first half was used to optimize parameters and the second to compile performances using optimal values obtained. The radiopaque markers of the PC were detected with a sensitivity of 88.3% and a positive predictive value (PPV) of 96%. The PC can therefore be positioned with a majority of its markers localized while the artefacts were all located inside the vessel lumen. The major parts of the DD, the dilatator tip and the pusher surfaces, were detected accurately with a sensitivity of 85.9% and a PPV of 88.7%. The less visible part of the DD, the stent enclosed within the sheath, was segmented with a sensitivity of 63.4% because the radiopacity of this region is low and uneven. The centreline of the DD in this stent region was alternatively traced within a 0.74 mm mean error. The automatic segmentation of endovascular devices during EVAR is feasible and accurate; it could be useful to perform elastic registration of the vascular lumen during endovascular repair.

Guidewire tracking during endovascular neurosurgery.

This paper presents a new method for guidewire tracking on fluoroscopic images from endovascular brain intervention. The combination of algorithms chosen can be implemented in real time, so that it can be used in an augmented reality 3D representation to assist physicians performing these interventions. A ribbon-like morphing process combined with a minimal path optimization algorithm is used to track lateral motion between successive frames. Forward motions are then tracked with an endpoint tracking algorithm, based on a circular window processed with the Radon transform. The proposed method was tested on 6 fluoroscopic sequences presenting high-speed motions, which were saved during endovascular brain interventions. The experiments showed above-average precision and robust guidewire tracking, without any permanent error requiring manual correction.