A framework for geometric analysis of vascular structures: application to cerebral aneurysms

Vessel-targeted compensation of deformable motion in interventional cone-beam CT

The present standard of care for unresectable liver cancer is transarterial chemoembolization (TACE), which involves using chemotherapeutic particles to selectively embolize the arteries supplying hepatic tumors. Accurate volumetric identification of intricate fine vascularity is crucial for selective embolization. Three-dimensional imaging, particularly cone-beam CT (CBCT), aids in visualization and targeting of small vessels in such highly variable anatomy, but long image acquisition time results in intra-scan patient motion, which distorts vascular structures and tissue boundaries. To improve clarity of vascular anatomy and intra-procedural utility, this work proposes a targeted motion estimation and compensation framework that removes the need for any prior information or external tracking and for user interaction. Motion estimation is performed in two stages: (i) a target identification stage that segments arteries and catheters in the projection domain using a multi-view convolutional neural network to construct a coarse 3D vascular mask; and (ii) a targeted motion estimation stage that iteratively solves for the time-varying motion field via optimization of a vessel-enhancing objective function computed over the target vascular mask. The vessel-enhancing objective is derived through eigenvalues of the local image Hessian to emphasize bright tubular structures. Motion compensation is achieved via spatial transformer operators that apply time-dependent deformations to partial angle reconstructions, allowing efficient minimization via gradient backpropagation. The framework was trained and evaluated in anatomically realistic simulated motion-corrupted CBCTs mimicking TACE of hepatic tumors, at intermediate (3.0 mm) and large (6.0 mm) motion magnitudes. Motion compensation substantially improved median vascular DICE score (from 0.30 to 0.59 for large motion), image SSIM (from 0.77 to 0.93 for large motion), and vessel sharpness (0.189 mm-1 to 0.233 mm-1 for large motion) in simulated cases. Motion compensation also demonstrated increased vessel sharpness (0.188 mm-1 before to 0.205 mm-1 after) and reconstructed vessel length (median increased from 37.37 to 41.00 mm) on a clinical interventional CBCT. The proposed anatomy-aware motion compensation framework presented a promising approach for improving the utility of CBCT for intra-procedural vascular imaging, facilitating selective embolization procedures.

Automatic segmentation of intraluminal thrombosis of abdominal aortic aneurysms from CT angiography using a mixed-scale-driven multiview perception network (M2Net) model

Intraluminal thrombosis (ILT) plays a critical role in the progression of abdominal aortic aneurysms (AAA). Understanding the role of ILT can improve the evaluation and management of AAAs. However, compared with highly developed automatic vessel lumen segmentation methods, ILT segmentation is challenging. Angiographic contrast agents can enhance the vessel lumen but cannot improve boundary delineation of the ILT regions; the lack of intrinsic contrast in the ILT structure significantly limits the accurate segmentation of ILT. Additionally, ILT is not evenly distributed within AAAs; its sparsity and scattered distributions in the imaging data pose challenges to the learning process of neural networks. Thus, we propose a multiview fusion approach, allowing us to obtain high-quality ILT delineation from computed tomography angiography (CTA) data. Our multiview fusion network is named Mixed-scale-driven Multiview Perception Network (M2Net), and it consists of two major steps. Following image preprocessing, the 2D mixed-scale ZoomNet segments ILT from each orthogonal view (i.e., Axial, Sagittal, and Coronal views) to enhance the prior information. Then, the proposed context-aware volume integration network (CVIN) effectively fuses the multiview results. Using contrast-enhanced computed tomography angiography (CTA) data from human subjects with AAAs, we evaluated the proposed M2Net. A quantitative analysis shows that the proposed deep-learning M2Net model achieved superior performance (e.g., DICE scores of 0.88 with a sensitivity of 0.92, respectively) compared with other state-of-the-art deep-learning models. In closing, the proposed M2Net model can provide high-quality delineation of ILT in an automated fashion and has the potential to be translated into the clinical workflow.

Role of the vessel morphology on the lenticulostriate arteries hemodynamics during atrial fibrillation: A CFD-based multivariate regression analysis

BACKGROUND AND OBJECTIVE

Atrial fibrillation (AF) is the most common cardiac arrhythmia, inducing accelerated and irregular beating. Beside well-known disabling symptoms - such as palpitations, reduced exercise tolerance, and chest discomfort - there is growing evidence that an alteration of deep cerebral hemodynamics due to AF increases the risk of vascular dementia and cognitive impairment, even in the absence of clinical strokes. The alteration of deep cerebral circulation in AF represents one of the least investigated among the possible mechanisms. Lenticulostriate arteries (LSAs) are small perforating arteries mainly departing from the middle cerebral artery (MCA) and susceptible to small vessel disease, which is one of the mechanisms of subcortical vascular dementia development. The purpose of this study is to investigate the impact of different LSAs morphologies on the cerebral hemodynamics during AF.

METHODS

By combining a computational fluid dynamics (CFD) analysis of LSAs with 7T high-resolution magnetic resonance imaging (MRI), we performed different CFD-based multivariate regression analyses to detect which geometrical and morphological vessel features mostly affect AF hemodynamics in terms of wall shear stress. We exploited 17 cerebral 7T-MRI derived LSA vascular geometries extracted from 10 subjects and internal carotid artery data from validated 0D cardiovascular-cerebral modeling as inflow conditions.

RESULTS

Our results revealed that few geometrical variables - namely the size of the MCA and the bifurcation angles between MCA and LSA - are able to satisfactorily predict the AF impact. In particular, the present study indicates that LSA morphologies exhibiting markedly obtuse LSA-MCA inlet angles and small MCA size downstream of the LSA-MCA bifurcation may be more prone to vascular damage induced by AF.

CONCLUSIONS

The present MRI-based computational study has been able for the first time to: (i) investigate the net impact of LSAs vascular morphologies on cerebral hemodynamics during AF events; (ii) detect which combination of morphological features worsens the hemodynamic response in the presence of AF. Awaiting necessary clinical confirmation, our analysis suggests that the local hemodynamics of LSAs is affected by their geometrical features and some LSA morphologies undergo greater hemodynamic alterations in the presence of AF.

A Deep Learning Framework for Analysis of the Eustachian Tube and the Internal Carotid Artery

OBJECTIVE

Obtaining automated, objective 3-dimensional (3D) models of the Eustachian tube (ET) and the internal carotid artery (ICA) from computed tomography (CT) scans could provide useful navigational and diagnostic information for ET pathologies and interventions. We aim to develop a deep learning (DL) pipeline to automatically segment the ET and ICA and use these segmentations to compute distances between these structures.

STUDY DESIGN

Retrospective cohort.

SETTING

Tertiary referral center.

METHODS

From a database of 30 CT scans, 60 ET and ICA pairs were manually segmented and used to train an nnU-Net model, a DL segmentation framework. These segmentations were also used to develop a quantitative tool to capture the magnitude and location of the minimum distance point (MDP) between ET and ICA. Performance metrics for the nnU-Net automated segmentations were calculated via the average Hausdorff distance (AHD) and dice similarity coefficient (DSC).

RESULTS

The AHD for the ET and ICA were 0.922 and 0.246 mm, respectively. Similarly, the DSC values for the ET and ICA were 0.578 and 0.884. The mean MDP from ET to ICA in the cartilaginous region was 2.6 mm (0.7-5.3 mm) and was located on average 1.9 mm caudal from the bony cartilaginous junction.

CONCLUSION

This study describes the first end-to-end DL pipeline for automated ET and ICA segmentation and analyzes distances between these structures. In addition to helping to ensure the safe selection of patients for ET dilation, this method can facilitate large-scale studies exploring the relationship between ET pathologies and the 3D shape of the ET.

Efficient and Accurate 3D Thickness Measurement in Vessel Wall Imaging: Overcoming Limitations of 2D Approaches Using the Laplacian Method

The clinical significance of measuring vessel wall thickness is widely acknowledged. Recent advancements have enabled high-resolution 3D scans of arteries and precise segmentation of their lumens and outer walls; however, most existing methods for assessing vessel wall thickness are 2D. Despite being valuable, reproducibility and accuracy of 2D techniques depend on the extracted 2D slices. Additionally, these methods fail to fully account for variations in wall thickness in all dimensions. Furthermore, most existing approaches are difficult to be extended into 3D and their measurements lack spatial localization and are primarily confined to lumen boundaries. We advocate for a shift in perspective towards recognizing vessel wall thickness measurement as inherently a 3D challenge and propose adapting the Laplacian method as an outstanding alternative. The Laplacian method is implemented using convolutions, ensuring its efficient and rapid execution on deep learning platforms. Experiments using digital phantoms and vessel wall imaging data are conducted to showcase the accuracy, reproducibility, and localization capabilities of the proposed approach. The proposed method produce consistent outcomes that remain independent of centerlines and 2D slices. Notably, this approach is applicable in both 2D and 3D scenarios. It allows for voxel-wise quantification of wall thickness, enabling precise identification of wall volumes exhibiting abnormal wall thickness. Our research highlights the urgency of transitioning to 3D methodologies for vessel wall thickness measurement. Such a transition not only acknowledges the intricate spatial variations of vessel walls, but also opens doors to more accurate, localized, and insightful diagnostic insights.

Novel Domain Knowledge-Encoding Algorithm Enables Label-Efficient Deep Learning for Cardiac CT Segmentation to Guide Atrial Fibrillation Treatment in a Pilot Dataset

Background: Segmenting computed tomography (CT) is crucial in various clinical applications, such as tailoring personalized cardiac ablation for managing cardiac arrhythmias. Automating segmentation through machine learning (ML) is hindered by the necessity for large, labeled training data, which can be challenging to obtain. This article proposes a novel approach for automated, robust labeling using domain knowledge to achieve high-performance segmentation by ML from a small training set. The approach, the domain knowledge-encoding (DOKEN) algorithm, reduces the reliance on large training datasets by encoding cardiac geometry while automatically labeling the training set. The method was validated in a hold-out dataset of CT results from an atrial fibrillation (AF) ablation study. Methods: The DOKEN algorithm parses left atrial (LA) structures, extracts "anatomical knowledge" by leveraging digital LA models (available publicly), and then applies this knowledge to achieve high ML segmentation performance with a small number of training samples. The DOKEN-labeled training set was used to train a nnU-Net deep neural network (DNN) model for segmenting cardiac CT in N = 20 patients. Subsequently, the method was tested in a hold-out set with N = 100 patients (five times larger than training set) who underwent AF ablation. Results: The DOKEN algorithm integrated with the nn-Unet model achieved high segmentation performance with few training samples, with a training to test ratio of 1:5. The Dice score of the DOKEN-enhanced model was 96.7% (IQR: 95.3% to 97.7%), with a median error in surface distance of boundaries of 1.51 mm (IQR: 0.72 to 3.12) and a mean centroid-boundary distance of 1.16 mm (95% CI: -4.57 to 6.89), similar to expert results (r = 0.99; p < 0.001). In digital hearts, the novel DOKEN approach segmented the LA structures with a mean difference for the centroid-boundary distances of -0.27 mm (95% CI: -3.87 to 3.33; r = 0.99; p < 0.0001). Conclusions: The proposed novel domain knowledge-encoding algorithm was able to perform the segmentation of six substructures of the LA, reducing the need for large training data sets. The combination of domain knowledge encoding and a machine learning approach could reduce the dependence of ML on large training datasets and could potentially be applied to AF ablation procedures and extended in the future to other imaging, 3D printing, and data science applications.

3D airway geometry analysis of factors in airway navigation failure for lung nodules

BACKGROUND

This study aimed to quantitatively reveal contributing factors to airway navigation failure during radial probe endobronchial ultrasound (R-EBUS) by using geometric analysis in a three-dimensional (3D) space and to investigate the clinical feasibility of prediction models for airway navigation failure.

METHODS

We retrospectively reviewed patients who underwent R-EBUS between January 2017 and December 2018. Geometric quantification was analyzed using in-house software built with open-source python libraries including the Vascular Modeling Toolkit ( http://www.vmtk.org ), simple insight toolkit ( https://sitk.org ), and sci-kit image ( https://scikit-image.org ). We used a machine learning-based approach to explore the utility of these significant factors.

RESULTS

Of the 491 patients who were eligible for analysis (mean age, 65 years +/- 11 [standard deviation]; 274 men), the target lesion was reached in 434 and was not reached in 57. Twenty-seven patients in the failure group were matched with 27 patients in the success group based on propensity scores. Bifurcation angle at the target branch, the least diameter of the last section, and the curvature of the last section are the most significant and stable factors for airway navigation failure. The support vector machine can predict airway navigation failure with an average area under the curve of 0.803.

CONCLUSIONS

Geometric analysis in 3D space revealed that a large bifurcation angle and a narrow and tortuous structure of the closest bronchus from the lesion are associated with airway navigation failure during R-EBUS. The models developed using quantitative computer tomography scan imaging show the potential to predict airway navigation failure.

Validating a Curvature-Based Marker of Cervical Carotid Tortuosity for Risk Assessment in Heritable Aortopathies

BACKGROUND

Cervical arterial tortuosity is associated with adverse outcomes in Loeys-Dietz syndrome and other heritable aortopathies.

METHODS AND RESULTS

A method to assess tortuosity based on curvature of the vessel centerline in 3-dimensional space was developed. We measured cervical carotid tortuosity in 65 patients with Loeys-Dietz syndrome from baseline computed tomography angiogram/magnetic resonance angiogram and all serial images during follow-up. Relations between baseline carotid tortuosity, age, aortic root diameter, and its change over time were compared. Patients with unoperated aortic roots were assessed for clinical end point (type A aortic dissection or aortic root surgery during 4 years of follow-up). Logistic regression was performed to assess the likelihood of clinical end point according to baseline carotid tortuosity. Total absolute curvature at baseline was 11.13±5.76 and was relatively unchanged at 8 to 10 years (fold change: 0.026±0.298, P=1.00), whereas tortuosity index at baseline was 0.262±0.131, with greater variability at 8 to 10 years (fold change: 0.302±0.656, P=0.818). Baseline total absolute curvature correlated with aortic root diameter (r=0.456, P=0.004) and was independently associated with aortic events during the 4-year follow-up (adjusted odds ratio [OR], 2.64 [95% CI, 1.02-6.85]). Baseline tortuosity index correlated with age (r=0.532, P<0.001) and was not associated with events (adjusted OR, 1.88 [95% CI, 0.79-4.51]). Finally, baseline total absolute curvature had good discrimination of 4-year outcomes (area under the curve=0.724, P=0.014), which may be prognostic or predictive.

CONCLUSIONS

Here we introduce cervical carotid tortuosity as a promising quantitative biomarker with validated, standardized characteristics. Specifically, we recommend the adoption of a curvature-based measure, total absolute curvature, for early detection or monitoring of disease progression in Loeys-Dietz syndrome.

Predicting cerebrovascular age and its clinical relevance: Modeling using 3D morphological features of brain vessels

Aging manifests as many phenotypes, among which age-related changes in brain vessels are important, but underexplored. Thus, in the present study, we constructed a model to predict age using cerebrovascular morphological features, further assessing their clinical relevance using a novel pipeline. Age prediction models were first developed using data from a normal cohort (n = 1181), after which their relevance was tested in two stroke cohorts (n = 564 and n = 455). Our novel pipeline adapted an existing framework to compute generic vessel features for brain vessels, resulting in 126 morphological features. We further built various machine learning models to predict age using only clinical factors, only brain vessel features, and a combination of both. We further assessed deviation from healthy aging using the age gap and explored its clinical relevance by correlating the predicted age and age gap with various risk factors. The models constructed using only brain vessel features and those combining clinical factors with vessel features were better predictors of age than the clinical factor-only model (r = 0.37, 0.48, and 0.26, respectively). Predicted age was associated with many known clinical factors, and the associations were stronger for the age gap in the normal cohort. The age gap was also associated with important factors in the pooled cohort atherosclerotic cardiovascular disease risk score and white matter hyperintensity measurements. Cerebrovascular age, computed using the morphological features of brain vessels, could serve as a potential individualized marker for the early detection of various cerebrovascular diseases.

Design space exploration of flow diverter hydraulic resistance parameters in sidewall intracranial aneurysms

Intracranial aneurysms are nowadays treated with endovascular flow diverter devices to avoid sac rupture. This study explores how different linear and quadratic hydrodynamic resistance parameters reduce the flow in the sac for five patient-specific sidewall aneurysms.The 125 performed blood flow simulations included the stents using a Darcy-Forcheimer porous layer approach based on real-life stent characteristics. Time- and space-averaged velocity magnitudes were strongly affected by the linear coefficient with a power-law relationship. Quadratic coefficients alter the flow in a minor way due to the low-velocity levels in the aneurysm sac and neck region.

Hemodynamic characteristics and mechanism for intracranial aneurysms initiation with the circle of Willis anomaly

Clinically, circle of Willis (CoW) is prone to anomaly and is also the predominant incidence site of intracranial aneurysms (IAs). This study aims to investigate the hemodynamic characteristics of CoW anomaly, and ascertain the mechanism of IAs initiation from the perspective of hemodynamics. Thus, the flow of IAs and pre-IAs were analyzed for one type of cerebral artery anomaly, that is, anterior cerebral artery A1 segment (ACA-A1) unilateral absence. Three patient geometrical models with IAs were selected from Emory University Open Source Data Center. IAs were virtually removed from the geometrical models to simulate the pre-IAs geometry. For calculation methods, a one-dimensional (1-D) solver and a three-dimensional (3-D) solver were combined to obtain the hemodynamic characteristics. The numerical simulation revealed that the average flow of Anterior Communicating Artery (ACoA) is almost zero when CoW is complete. In contrast, ACoA flow increases significantly in the case of ACA-A1 unilateral absence. For per-IAs geometry, the jet flow is found at the bifurcation between contralateral ACA-A1 and ACoA, which exhibits characteristics of high Wall Shear Stress (WSS) and high wall pressure in the impact region. It triggers the initiation of IAs from the perspective of hemodynamics. The vascular anomaly that leads to jet flow should be considered as a risk factor for IAs initiation.

Modification of Woven Endo-Bridge After Intracranial Aneurysm Treatment: A Methodology for Three-Dimensional Analysis of Shape and Relative Position Changes

During follow-up of patients treated with WEB devices, shape changes have been observed. The quantitative three-dimensional measurement of the WEB shape modification (WSM) would offer useful information to be studied in association with the anatomical results and try to better understand mechanisms implicated in this modification phenomenon. We present a methodology to quantify the morphology and position of the WEB device in relation to the vascular anatomy. Three-dimensional rotational angiography (3DRA) images of seven aneurysms patients treated with WEBs were used, which also accompanied by a post-treatment 3DRA image and a follow-up 3DRA image. The device was manually segmented, obtaining the 3D models after treatment and at the follow-up. Volume, surface area, height, maximum diameter and WSM ratio of both surfaces were calculated. Position changes were evaluated measuring WEB axis and relative position between post-treatment and follow-up. Changes in WEB volume and surface area were observed with a mean modification of - 5.04 % ( ± 14.19 ) and - 1.68 % ( ± 8.29 ) , respectively. The positional variables also showed differences, mean change of device axis direction was 26.25 % ( ± 24.09 ) and mean change of distance l b was 5.87 % ( ± 10.59 ) . Inter-observer and intra-observer variability analyses did not show differences (ANOVA p > 0.05 ). This methodology allows quantifying the morphological and position changes suffered by the WEB device after treatment, offering new information to be studied in relation to the occurrence of WEB shape modification.

A Sewing Approach to the Fabrication of Eco/bioresorbable Electronics

Eco/bioresorbable electronics represent an emerging class of technology defined by an ability to dissolve or otherwise harmlessly disappear in environmental or biological surroundings after a period of stable operation. The resulting devices provide unique capabilities as temporary biomedical implants, environmental sensors, and related systems. Recent publications report schemes to overcome challenges in fabrication that follow from the low thermostability and/or high chemical reactivity of the eco/bioresorbable constituent materials. Here, this work reports the use of high-speed sewing machines, as the basis for a high-throughput manufacturing technique that addresses many requirements for these applications, without the need for high temperatures or reactive solvents. Results demonstrate that a range of eco/bioresorbable metal wires and polymer threads can be embroidered into complex, user-defined conductive patterns on eco/bioresorbable substrates. Functional electronic components, such as stretchable interconnects and antennas are possible, along with fully integrated systems. Examples of the latter include wirelessly powered light-emitting diodes, radiofrequency identification tags, and temporary cardiac pacemakers. These advances add to a growing range of options in high-throughput, automated fabrication of eco/bioresorbable electronics.

Comparative Study of Automated Algorithms for Brain Arteriovenous Malformation Nidus Extent Identification Using 3DRA

PURPOSE

When performing a brain arteriovenous malformation (bAVMs) intervention, computer-assisted analysis of bAVMs can aid clinicians in planning precise therapeutic alternatives. Therefore, we aim to assess currently available methods for bAVMs nidus extent identification over 3DRA. To this end, we establish a unified framework to contrast them over the same dataset, fully automatising the workflows.

MATERIALS AND METHODS

We retrospectively collected contrast-enhanced 3DRA scans of patients with bAVMs. A segmentation network was used to automatically acquire the brain vessels segmentation for each case. We applied the nidus extent identification algorithms over each of the segmentations, computing overlap measurements against manual nidus delineations.

RESULTS

We evaluated the methods over a private dataset with 22 3DRA scans of individuals with bAVMs. The best-performing alternatives resulted in [Formula: see text] and [Formula: see text] dice coefficient values.

CONCLUSIONS

The mathematical morphology-based approach showed higher robustness through inter-case variability. The skeleton-based approach leverages the skeleton topomorphology characteristics, while being highly sensitive to anatomical variations and the skeletonisation method employed. Overall, nidus extent identification algorithms are also limited by the quality of the raw volume, as the consequent imprecise vessel segmentation will hinder their results. Performance of the available alternatives remains subpar. This analysis allows for a better understanding of the current limitations.

Association of Arterial Tortuosity with Hemodynamic Parameters-A Computational Fluid Dynamics Study

BACKGROUND

Tortuosity of intracranial arteries has been proven to be associated with the risk of intracranial aneurysm development. We decided to analyze which aspects of tortuosity are correlated with hemodynamics parameters promoting intracranial aneurysm development.

METHODS

We constructed 73 idealized models of tortuous artery (length: 25 mm, diameter: 2.5 mm) with single bifurcation. For each model, on the course of segment before bifurcation, we placed 1-3 angles with measures 15, 30, 45, 60, or 75 degrees and arc lengths 2, 5, 7, 10, or 15 mm. We performed computational fluid dynamics analysis. Blood was modeled as Newtonian fluid. We have set velocity wave of 2 cardiac cycles. After performing simulation we calculated following hemodynamic parameters at the bifurcation: time average wall shear stress (TAWSS), time average wall shear stress gradient (TAWSSG), oscillatory shear index (OSI), and relative residence time (RRT).

RESULTS

We found a significant positive correlation with number of angles and TAWSS (R = 0.329; P < 0.01), TAWSSG (R = 0.317; P < 0.01), and negative with RRT (R = -0.335; P < 0.0.01). Similar results were obtained in terms of arcs lengths. On the other hand, mean angle measure was negatively correlated to TAWSS (R = -0.333; P < 0.01), TAWSSG (R = -0.473 P < 0.01), OSI (R = -0.463; P < 0.01), and positively to RRT (R = 0.332; P < 0.01). On the basis of the obtained results, we developed new tortuosity descriptor, which considered angle measures normalized to its arc length and distance from bifurcation. For such descriptor we found strong negative correlation with TAWSS (R = -0.701; P < 0.01), TAWSSG (R = 0.778; P < 0.01), OSI (R = -0.776; P < 0.01), and positive with RRT (R = 0.747; P < 0.01).

CONCLUSIONS

Hemodynamic parameters promoting aneurysm development are correlated with larger number of smaller angles located on larger arcs.

Significant unfavorable geometrical changes in ascending aorta despite stable diameter at follow-up

BACKGROUND

The clinical importance of following-up the ascending aortic diameter lies on the fundamental presumption that wall pathology eventually manifests as change in shape. However, the diameter describes the vessel locally, and the 55mm criterion fails to prevent most dissections. We hypothesized that geometric changes across the ascending aorta are not necessarily imprinted on its diameter; i.e. the maximum diameter correlates weakly and insignificantly with elongation, surface stretching, engorgement, and tortuosity.

METHODS

Two databases were interrogated for patients who had undergone at least 2 ECG-gated CT scans. The absence of motion artifacts permitted the generation of exact copies of the ascending aorta which then underwent three-dimensional analysis producing objective and accurate measurements of the centreline length, surface, volume, and tortuosity. The correlations of these global variables with the diameter were explored.

RESULTS

Twenty-two patients, 13 male and 9 females, were included. The mean age at the first and last scan was 63.7 and 67.1y, respectively. The mean diameter increase was approximately 1mm/y. There were no dissections, while 7 patients underwent preemptive surgery. The yearly change rate of the global variables, normalized to height if applicable, showed statistically insignificant, weak or negligible correlation with diameter increments at follow-up. Most characteristically, a patient's aorta maintained its diameter, while undergoing 1mm/y elongation, 151mm2/(y∙m) stretching, 2366mm3/(y∙m) engorgement, and 0.02/y tortuosity.

CONCLUSIONS

Maximum diameter provides a local description for the ascending aorta and cannot fully portray the pathological process across this vessel. Following-up the diameter is not suggestive of length, surface, volume and tortuosity changes.

Segmenting computed tomograms for cardiac ablation using machine learning leveraged by domain knowledge encoding

Background

Segmentation of computed tomography (CT) is important for many clinical procedures including personalized cardiac ablation for the management of cardiac arrhythmias. While segmentation can be automated by machine learning (ML), it is limited by the need for large, labeled training data that may be difficult to obtain. We set out to combine ML of cardiac CT with domain knowledge, which reduces the need for large training datasets by encoding cardiac geometry, which we then tested in independent datasets and in a prospective study of atrial fibrillation (AF) ablation.

Methods

We mathematically represented atrial anatomy with simple geometric shapes and derived a model to parse cardiac structures in a small set of N = 6 digital hearts. The model, termed "virtual dissection," was used to train ML to segment cardiac CT in N = 20 patients, then tested in independent datasets and in a prospective study.

Results

In independent test cohorts (N = 160) from 2 Institutions with different CT scanners, atrial structures were accurately segmented with Dice scores of 96.7% in internal (IQR: 95.3%-97.7%) and 93.5% in external (IQR: 91.9%-94.7%) test data, with good agreement with experts (r = 0.99; p < 0.0001). In a prospective study of 42 patients at ablation, this approach reduced segmentation time by 85% (2.3 ± 0.8 vs. 15.0 ± 6.9 min, p < 0.0001), yet provided similar Dice scores to experts (93.9% (IQR: 93.0%-94.6%) vs. 94.4% (IQR: 92.8%-95.7%), p = NS).

Conclusions

Encoding cardiac geometry using mathematical models greatly accelerated training of ML to segment CT, reducing the need for large training sets while retaining accuracy in independent test data. Combining ML with domain knowledge may have broad applications.

Morphological characterisation of pediatric Turner syndrome aortae: Insights from a small cohort study

Cardiovascular disease is widespread in girls and women living with Turner syndrome (TS). Despite this prevalence, cardiovascular risk evaluation using the current guidelines has seen life-threatening aortic events occurring at dimensions classified within the normal threshold. In this study, we characterized the three-dimensional aortic geometries of Turner syndrome children and their age-matched healthy counterparts to evaluate various morphological parameters. Turner syndrome girls had overall greater values in ten out of fifteen parameters examined (p > 0.05), when compared to healthy children: the aortic arch height and width; the ascending aorta, aortic arch (2 locations), and descending aorta diameters; the ratio of the ascending to descending aorta diameter; average curvature; average torsion; and average curvature-torsion score. Additionally, significant associations were found in the TS group: body surface area and both arch height (p = 0.03) and arch height to width ratio (p = 0.05), and aortic arch diameter and both body surface area (p = 0.04) and weight (p = 0.04). The new information resulting from this small cohort study contributes to an improved understanding of the morphological parameters affecting the hemodynamic environment in TS, and the clinical assessment of the increased cardiovascular risk in this population.

Volumetric surveillance of brain aneurysms: Pitfalls of MRA

INTRODUCTION

Untreated brain aneurysms are usually surveilled with serial MR imaging and evaluated with 2D multiplanar measurements. The assessment of aneurysm growth may be more accurate with volumetric analysis. We evaluated the accuracy of a magnetic resonance angiography (MRA) segmentation pipeline for aneurysm volume measurement and surveillance.

METHODS

A pipeline to determine aneurysm volume was developed and tested on two aneurysm phantoms imaged with time-of flight (TOF) MRA and 3D rotational angiography (3DRA). The accuracy of the pipeline was then evaluated by reconstructing 10 aneurysms imaged with contrast enhanced-MRA (CE-MRA) and 3DRA. This calibrated and refined post-processing pipeline was subsequently used to analyse aneurysms from our prospectively acquired database. Volume changes above the threshold of error were considered true volume changes. The accuracy of these measurements was analysed.

RESULTS

TOF-MRA reconstructions were not as accurate as CE-MRA reconstructions. When compared to 3DRA, CE-MRA underestimated aneurysm volume by 7.8% and did not accurately register the presence of blebs. Eighteen aneurysms (13 saccular and 5 fusiform) were analysed with the optimized 3D volume reconstruction pipeline, with a mean follow-up time of 11 months. Artifact accounted for 10.2% error in volume measurements using serial CE-MRA. When this margin of error was used to assess aneurysms volume in serial imaging with CE-MRA, only two fusiform aneurysms changed in volume. The variations in volume of these two fusiform aneurysms were caused by intra-mural and intrasaccular thrombosis.

CONCLUSIONS

CE-MRA and TOF-MRA 3D volume reconstructions may not register minor morphological changes such as the appearance of blebs. CE-MRA underestimates volume by 7.8% compared to 3DRA. Serial CE-MRA volume measurements had a larger margin of error of approximately 10.2%. MRA-based volumetric measurements may not be appropriate for aneurysm surveillance.

Reconstruction and Validation of Arterial Geometries for Computational Fluid Dynamics Using Multiple Temporal Frames of 4D Flow-MRI Magnitude Images

PURPOSE

Segmentation and reconstruction of arterial blood vessels is a fundamental step in the translation of computational fluid dynamics (CFD) to the clinical practice. Four-dimensional flow magnetic resonance imaging (4D Flow-MRI) can provide detailed information of blood flow but processing this information to elucidate the underlying anatomical structures is challenging. In this study, we present a novel approach to create high-contrast anatomical images from retrospective 4D Flow-MRI data.

METHODS

For healthy and clinical cases, the 3D instantaneous velocities at multiple cardiac time steps were superimposed directly onto the 4D Flow-MRI magnitude images and combined into a single composite frame. This new Composite Phase-Contrast Magnetic Resonance Angiogram (CPC-MRA) resulted in enhanced and uniform contrast within the lumen. These images were subsequently segmented and reconstructed to generate 3D arterial models for CFD. Using the time-dependent, 3D incompressible Reynolds-averaged Navier-Stokes equations, the transient aortic haemodynamics was computed within a rigid wall model of patient geometries.

RESULTS

Validation of these models against the gold standard CT-based approach showed no statistically significant inter-modality difference regarding vessel radius or curvature (p > 0.05), and a similar Dice Similarity Coefficient and Hausdorff Distance. CFD-derived near-wall hemodynamics indicated a significant inter-modality difference (p > 0.05), though these absolute errors were small. When compared to the in vivo data, CFD-derived velocities were qualitatively similar.

CONCLUSION

This proof-of-concept study demonstrated that functional 4D Flow-MRI information can be utilized to retrospectively generate anatomical information for CFD models in the absence of standard imaging datasets and intravenous contrast.

On the major role played by the lumen curvature of intracranial aneurysms walls in determining their mechanical response, local hemodynamics, and rupture likelihood

The properties of intracranial aneurysms (IAs) walls are known to be driven by the underlying hemodynamics adjacent to the IA sac. Different pathways exist explaining the connections between hemodynamics and local tissue properties. The emergence of such theories is essential if one wishes to compute the mechanical response of a patient-specific IA wall and predict its rupture. Apart from the hemodynamics and tissue properties, one could assume that the mechanical response also depends on the local morphology, more specifically, the curvature of the luminal surface, with larger values at highly-curved wall portions. Nonetheless, this contradicts observations of IA rupture sites more often found at the dome, where the curvature is lower. This seeming contradiction indicates a complex interaction between the hemodynamics adjacent to the aneurysm wall, its morphology, and mechanical response, which warrants further investigation. This was the main goal of this work. We accomplished this by analyzing the stress and stretch fields in different regions of the wall for a sample of IAs, which have been classified based on particular hemodynamics conditions and lumen curvature. Pulsatile numerical simulations were performed using the one-way fluid-solid interaction strategy implemented in OpenFOAM (solids4foam toolbox). We found that the variable best correlated with regions of high stress and stretch was the lumen curvature. Additionally, our data suggest a connection between the local curvature and particular hemodynamics conditions adjacent to the wall, indicating that the lumen curvature is a property that could be used to assess both mechanical response and hemodynamic conditions, and, moreover, suggest new rupture indicators based on the curvature.

Observational study of endoluminal mural thrombotic apposition in popliteal artery aneurysm stenting and its relationship with stent-graft geometrical features

Introduction

The development of intrastent thrombosis is one of the mechanisms related to medium- to long-term failure of endovascular treatment of popliteal artery aneurysm. The present study aims to investigate possible links between the development of endoluminal mural thrombotic apposition in the stented zone (EMTS) with both geometrical features of stent-graft(s) and time of follow-up.

Methods

Patients with popliteal artery aneurysm who underwent endovascular treatment were recruited during the follow-up period. Segmentation of computed tomography angiography scan was performed to detect femoropopliteal artery lumen, leg bones, EMTS, and stent-graft(s). The following parameters were assessed: number, diameter, and length of stent-graft(s); and shape, volume, and length of thrombotic apposition within the stent(s). The spiral shape of the thrombotic apposition was evaluated as well.

Results

Eighteen male patients were recruited in the study. EMTS was observed in 13 of them (72%) during the follow-up analysis. An average of 1.8 ± 0.79 stents-grafts were implanted per patient with a median diameter and length of 6.2 (1.9) mm and 125 (50) mm, respectively. The percentage of the stent length where EMTS was present was 42.1 on average (interquartile range: 42.4%) with a mean volume of 206.8 mm3. A positive correlation was found between the length and volume of EMTS (R-squared = 0.71, p < 0.01). Moreover, EMTS had a helical shape in 8/13 patients, with 4/5 with counterclockwise rotation with stent-grafts in the left leg and 3/3 with clockwise direction treated in the right leg. A higher frequency of EMTS was observed in patients with longer follow-up and higher risk factors, as well.

Conclusions

EMTS is observed in most of the patients under analysis, especially in those with medium- to long-term follow-up. The pattern of such EMTS follows a helical shape having a direction that depends on which leg, right or left, is treated. Our results suggest a close surveillance of popliteal aneurysm stenting by follow-up examinations to control the onset and progression of EMTS.

Computerized Differentiation of Growth Status for Abdominal Aortic Aneurysms: A Feasibility Study

Fast-growing abdominal aortic aneurysms (AAA) have a high rupture risk and poor outcomes if not promptly identified and treated. Our primary objective is to improve the differentiation of small AAAs' growth status (fast versus slow-growing) through a combination of patient health information, computational hemodynamics, geometric analysis, and artificial intelligence. 3D computed tomography angiography (CTA) data available for 70 patients diagnosed with AAAs with known growth status were used to conduct geometric and hemodynamic analyses. Differences among ten metrics (out of ninety metrics) were statistically significant discriminators between fast and slow-growing groups. Using a support vector machine (SVM) classifier, the area under receiving operating curve (AUROC) and total accuracy of our best predictive model for differentiation of AAAs' growth status were 0.86 and 77.50%, respectively. In summary, the proposed analytics has the potential to differentiate fast from slow-growing AAAs, helping guide resource allocation for the management of patients with AAAs.

Prediction of guidewire-induced aortic deformations during EVAR: a finite element and in vitro study

Introduction and aims: During an Endovascular Aneurysm Repair (EVAR) procedure a stiff guidewire is inserted from the iliac arteries. This induces significant deformations on the vasculature, thus, affecting the pre-operative planning, and the accuracy of image fusion. The aim of the present work is to predict the guidewire induced deformations using a finite element approach validated through experiments with patient-specific additive manufactured models. The numerical approach herein developed could improve the pre-operative planning and the intra-operative navigation. Material and methods: The physical models used for the experiments in the hybrid operating room, were manufactured from the segmentations of pre-operative Computed Tomography (CT) angiographies. The finite element analyses (FEA) were performed with LS-DYNA Explicit. The material properties used in finite element analyses were obtained by uniaxial tensile tests. The experimental deformed configurations of the aorta were compared to those obtained from FEA. Three models, obtained from Computed Tomography acquisitions, were investigated in the present work: A) without intraluminal thrombus (ILT), B) with ILT, C) with ILT and calcifications. Results and discussion: A good agreement was found between the experimental and the computational studies. The average error between the final in vitro vs. in silico aortic configurations, i.e., when the guidewire is fully inserted, are equal to 1.17, 1.22 and 1.40 mm, respectively, for Models A, B and C. The increasing trend in values of deformations from Model A to Model C was noticed both experimentally and numerically. The presented validated computational approach in combination with a tracking technology of the endovascular devices may be used to obtain the intra-operative configuration of the vessels and devices prior to the procedure, thus limiting the radiation exposure and the contrast agent dose.

Automated Aortic Anatomy Analysis: from Image to Clinical Indicators

Most cerebrovascular diseases (including strokes and aneurysms) are treated endovascularly with catheters that are navigated from the groin through the vessels to the brain. Many patients have complex anatomy of the aortic arch and supra-aortic vessels, which can make it difficult to select the best catheters for navigation, resulting in longer procedures and more complications or failures. To this end, we propose a framework dedicated to the analysis of the aortic arch and supra-aortic trunks. This framework can automatically compute anatomical and geometrical features from meshes segmented beforehand via CNN-based pipeline. These features such as arch type, tortuosity and angulations describe the navigational difficulties encountered during catheterization. Quantitative and qualitative validation was performed by experienced neuroradiologists, leading to reliable vessel characterization.Clinical relevance- This method allows clinicians to determine the type and the anatomy of the aortic arch and its supra-aortic trunks before endovascular procedures. This is essential in interventional neuroradiology, such as navigation with catheters in this complex area.

Learning the Hidden Signature of Fetal Arch Anatomy: a Three-Dimensional Shape Analysis in Suspected Coarctation of the Aorta

Neonatal coarctation of the aorta (CoA) is a common congenital heart defect. Its antenatal diagnosis remains challenging, and its pathophysiology is poorly understood. We present a novel statistical shape modeling (SSM) pipeline to study the role and predictive value of arch shape in CoA in utero. Cardiac magnetic resonance imaging (CMR) data of 112 fetuses with suspected CoA was acquired and motion-corrected to three-dimensional volumes. Centerlines from fetal arches were extracted and used to build a statistical shape model capturing relevant anatomical variations. A linear discriminant analysis was used to find the optimal axis between CoA and false positive cases. The CoA shape risk score classified cases with an area under the curve of 0.907. We demonstrate the feasibility of applying a SSM pipeline to three-dimensional fetal CMR data while providing novel insights into the anatomical determinants of CoA and the relevance of in utero arch anatomy for antenatal diagnosis of CoA.

Atrial fibrillation ablation outcome prediction with a machine learning fusion framework incorporating cardiac computed tomography

BACKGROUND

Structural changes in the left atrium (LA) modestly predict outcomes in patients undergoing catheter ablation for atrial fibrillation (AF). Machine learning (ML) is a promising approach to personalize AF management strategies and improve predictive risk models after catheter ablation by integrating atrial geometry from cardiac computed tomography (CT) scans and patient-specific clinical data. We hypothesized that ML approaches based on a patient's specific data can identify responders to AF ablation.

METHODS

Consecutive patients undergoing AF ablation, who had preprocedural CT scans, demographics, and 1-year follow-up data, were included in the study for a retrospective analysis. The inputs of models were CT-derived morphological features from left atrial segmentation (including the shape, volume of the LA, LA appendage, and pulmonary vein ostia) along with deep features learned directly from raw CT images, and clinical data. These were merged intelligently in a framework to learn their individual importance and produce the optimal classification.

RESULTS

Three hundred twenty-one patients (64.2 ± 10.6 years, 69% male, 40% paroxysmal AF) were analyzed. Post 10-fold nested cross-validation, the model trained to intelligently merge and learn appropriate weights for clinical, morphological, and imaging data (AUC 0.821) outperformed those trained solely on clinical data (AUC 0.626), morphological (AUC 0.659), or imaging data (AUC 0.764).

CONCLUSION

Our ML approach provides an end-to-end automated technique to predict AF ablation outcomes using deep learning from CT images, derived structural properties of LA, augmented by incorporation of clinical data in a merged ML framework. This can help develop personalized strategies for patient selection in invasive management of AF.

USING CONVOLUTIONAL NEURAL NETWORK-BASED SEGMENTATION FOR IMAGE-BASED COMPUTATIONAL FLUID DYNAMICS SIMULATIONS OF BRAIN ANEURYSMS: INITIAL EXPERIENCE IN AUTOMATED MODEL CREATION

"Image-based" computational fluid dynamics (CFD) simulations provide insights into each patient's hemodynamic environment. However, current standard procedures for creating CFD models start with manual segmentation and are time-consuming, hindering the clinical translation of image-based CFD simulations. This feasibility study adopts deep-learning-based image segmentation (hereafter referred to as Artificial Intelligence (AI) segmentation) to replace manual segmentation to accelerate CFD model creation. Two published convolutional neural network-based AI methods (MIScnn and DeepMedic) were selected to perform CFD model extraction from three-dimensional (3D) rotational angiography data containing intracranial aneurysms. In this study, aneurysm morphological and hemodynamic results using models generated by AI segmentation methods were compared with those obtained by two human users for the same data. Interclass coefficients (ICC), Bland-Altman plots, and Pearson's correlation coefficients (PCC) were combined to assess how well AI-generated CFD models were performed. We found that almost perfect agreement was obtained between the human and AI results for all eleven morphological and five out of eight hemodynamic parameters, while a moderate agreement was obtained from the remaining three hemodynamic parameters. Given this level of agreement, using AI segmentation to create CFD models is feasible, given more developments.

Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields

Computational fluid dynamics (CFD) can be used to simulate vascular haemodynamics and analyse potential treatment options. CFD has shown to be beneficial in improving patient outcomes. However, the implementation of CFD for routine clinical use is yet to be realised. Barriers for CFD include high computational resources, specialist experience needed for designing simulation set-ups, and long processing times. The aim of this study was to explore the use of machine learning (ML) to replicate conventional aortic CFD with automatic and fast regression models. Data used to train/test the model consisted of 3,000 CFD simulations performed on synthetically generated 3D aortic shapes. These subjects were generated from a statistical shape model (SSM) built on real patient-specific aortas (N = 67). Inference performed on 200 test shapes resulted in average errors of 6.01% ±3.12 SD and 3.99% ±0.93 SD for pressure and velocity, respectively. Our ML-based models performed CFD in ∼0.075 seconds (4,000x faster than the solver). This proof-of-concept study shows that results from conventional vascular CFD can be reproduced using ML at a much faster rate, in an automatic process, and with reasonable accuracy.

Automated in-depth cerebral arterial labelling using cerebrovascular vasculature reframing and deep neural networks

Identifying the cerebral arterial branches is essential for undertaking a computational approach to cerebrovascular imaging. However, the complexity and inter-individual differences involved in this process have not been thoroughly studied. We used machine learning to examine the anatomical profile of the cerebral arterial tree. The method is less sensitive to inter-subject and cohort-wise anatomical variations and exhibits robust performance with an unprecedented in-depth vessel range. We applied machine learning algorithms to disease-free healthy control subjects (n = 42), patients with stroke with intracranial atherosclerosis (ICAS) (n = 46), and patients with stroke mixed with the existing controls (n = 69). We trained and tested 70% and 30% of each study cohort, respectively, incorporating spatial coordinates and geometric vessel feature vectors. Cerebral arterial images were analyzed based on the 'segmentation-stacking' method using magnetic resonance angiography. We precisely classified the cerebral arteries across the exhaustive scope of vessel components using advanced geometric characterization, redefinition of vessel unit conception, and post-processing algorithms. We verified that the neural network ensemble, with multiple joint models as the combined predictor, classified all vessel component types independent of inter-subject variations in cerebral arterial anatomy. The validity of the categorization performance of the model was tested, considering the control, ICAS, and control-blended stroke cohorts, using the area under the receiver operating characteristic (ROC) curve and precision-recall curve. The classification accuracy rarely fell outside each image's 90-99% scope, independent of cohort-dependent cerebrovascular structural variations. The classification ensemble was calibrated with high overall area rates under the ROC curve of 0.99-1.00 [0.97-1.00] in the test set across various study cohorts. Identifying an all-inclusive range of vessel components across controls, ICAS, and stroke patients, the accuracy rates of the prediction were: internal carotid arteries, 91-100%; middle cerebral arteries, 82-98%; anterior cerebral arteries, 88-100%; posterior cerebral arteries, 87-100%; and collections of superior, anterior inferior, and posterior inferior cerebellar arteries, 90-99% in the chunk-level classification. Using a voting algorithm on the queued classified vessel factors and anatomically post-processing the automatically classified results intensified quantitative prediction performance. We employed stochastic clustering and deep neural network ensembles. Ma-chine intelligence-assisted prediction of vessel structure allowed us to personalize quantitative predictions of various types of cerebral arterial structures, contributing to precise and efficient decisions regarding the cerebrovascular disease.

Comprehensive morphomechanical analysis of brain aneurysms

OBJECTIVE

Brain aneurysms comprise different compartments that undergo unique biological processes. A detailed multimodal analysis incorporating 3D aneurysm wall enhancement (AWE), computational fluid dynamics (CFD), and finite element analysis (FEA) data can provide insights into the aneurysm wall biology.

METHODS

Unruptured aneurysms were prospectively imaged with 7 T high-resolution MRI (HR-MRI). 3D AWE color maps of the entire aneurysm wall were generated and co-registered with contour plots of morphomechanical parameters derived from CFD and FEA. A multimodal analysis of the entire aneurysm was performed using 3D circumferential AWE (3D-CAWE), wall tension (WT), time-averaged wall shear stress (TAWSS), wall shear stress gradient (WSSG), and oscillatory shear index (OSI). A detailed compartmental analysis of each aneurysm's dome, bleb, and neck was also performed.

RESULTS

Twenty-six aneurysms were analyzed. 3D-CAWE + aneurysms had higher WT (p = 0.03) and higher TAWSS (p = 0.045) than 3D-CAWE- aneurysms. WT, TAWSS, and WSSG were lower in areas of focal AWE in the aneurysm dome compared to the neck (p = 0.009, p = 0.049, and p = 0.040, respectively), whereas OSI was higher in areas of focal AWE compared to the neck (p = 0.020). When compared to areas of no AWE of the aneurysm sac (AWE = 0.92 vs. 0.49, p = 0.001), blebs exhibited lower WT (1.6 vs. 2.45, p = 0.010), lower TAWSS (2.6 vs. 6.34), lower OSI (0.0007 vs. 0.0010), and lower WSSG (2900 vs. 5306). Fusiform aneurysms had a higher 3D-CAWE and WT than saccular aneurysms (p = 0.046 and p = 0.003, respectively).

CONCLUSIONS

Areas of focal high AWE in the sac and blebs are associated with low wall tension, low wall shear stress, and low flow conditions (TAWSS and WSSG). Conversely, the neck had average AWE, high wall tension, high wall shear stress, and high flow conditions. The aneurysm dome and the aneurysm neck have different morphomechanical environments, with increased mechanical load at the neck.

Predicting reachability to peripheral lesions in transbronchial biopsies using CT-derived geometrical attributes of the bronchial route

PURPOSE

The bronchoscopist's ability to locate the lesion with the bronchoscope is critical for a transbronchial biopsy. However, much less study has been done on the transbronchial biopsy route. This study aims to determine whether the geometrical attributes of the bronchial route can predict the difficulty of reaching tumors in bronchoscopic intervention.

METHODS

This study included patients who underwent bronchoscopic diagnosis of lung tumors using electromagnetic navigation. The biopsy instrument was considered "reached" and recorded as such if the tip of the tracked bronchoscope or extended working channel was in the tumors. Four geometrical indices were defined: Local curvature (LC), plane rotation (PR), radius, and global relative angle. A Mann-Whitney U test and logistic regression analysis were performed to analyze the difference in geometrical indices between the reachable and unreachable groups. Receiver operating characteristic analysis (ROC) was performed to evaluate the geometrical indices to predict reachability.

RESULTS

Of the 41 patients enrolled in the study, 16 patients were assigned to the unreachable group and 25 patients to the reachable group. LC, PR, and radius have significantly higher values in unreachable cases than in reachable cases ([Formula: see text], [Formula: see text], [Formula: see text]). The logistic regression analysis showed that LC and PR were significantly associated with reachability ([Formula: see text], [Formula: see text]). The areas under the curve with ROC analysis of the LC and PR index were 0.903 and 0.618. The LC's cut-off value was 578.25.

CONCLUSION

We investigated whether the geometrical attributes of the bronchial route to the lesion can predict the difficulty of reaching the lesions in the bronchoscopic biopsy. LC, PR, and radius have significantly higher values in unreachable cases than in reachable cases. LC and PR index can be potentially used to predict the navigational success of the bronchoscope.

An attention residual u-net with differential preprocessing and geometric postprocessing: Learning how to segment vasculature including intracranial aneurysms

OBJECTIVE

Intracranial aneurysms (IA) are lethal, with high morbidity and mortality rates. Reliable, rapid, and accurate segmentation of IAs and their adjacent vasculature from medical imaging data is important to improve the clinical management of patients with IAs. However, due to the blurred boundaries and complex structure of IAs and overlapping with brain tissue or other cerebral arteries, image segmentation of IAs remains challenging. This study aimed to develop an attention residual U-Net (ARU-Net) architecture with differential preprocessing and geometric postprocessing for automatic segmentation of IAs and their adjacent arteries in conjunction with 3D rotational angiography (3DRA) images.

METHODS

The proposed ARU-Net followed the classic U-Net framework with the following key enhancements. First, we preprocessed the 3DRA images based on boundary enhancement to capture more contour information and enhance the presence of small vessels. Second, we introduced the long skip connections of the attention gate at each layer of the fully convolutional decoder-encoder structure to emphasize the field of view (FOV) for IAs. Third, residual-based short skip connections were also embedded in each layer to implement in-depth supervision to help the network converge. Fourth, we devised a multiscale supervision strategy for independent prediction at different levels of the decoding path, integrating multiscale semantic information to facilitate the segmentation of small vessels. Fifth, the 3D conditional random field (3DCRF) and 3D connected component optimization (3DCCO) were exploited as postprocessing to optimize the segmentation results.

RESULTS

Comprehensive experimental assessments validated the effectiveness of our ARU-Net. The proposed ARU-Net model achieved comparable or superior performance to the state-of-the-art methods through quantitative and qualitative evaluations. Notably, we found that ARU-Net improved the identification of arteries connecting to an IA, including small arteries that were hard to recognize by other methods. Consequently, IA geometries segmented by the proposed ARU-Net model yielded superior performance during subsequent computational hemodynamic studies (also known as "patient-specific" computational fluid dynamics [CFD] simulations). Furthermore, in an ablation study, the five key enhancements mentioned above were confirmed.

CONCLUSIONS

The proposed ARU-Net model can automatically segment the IAs in 3DRA images with relatively high accuracy and potentially has significant value for clinical computational hemodynamic analysis.

Customizable tubular model for n-furcating blood vessels and its application to 3D reconstruction of the cerebrovascular system

Understanding the 3D cerebral vascular network is one of the pressing issues impacting the diagnostics of various systemic disorders and is helpful in clinical therapeutic strategies. Unfortunately, the existing software in the radiological workstation does not meet the expectations of radiologists who require a computerized system for detailed, quantitative analysis of the human cerebrovascular system in 3D and a standardized geometric description of its components. In this study, we show a method that uses 3D image data from magnetic resonance imaging with contrast to create a geometrical reconstruction of the vessels and a parametric description of the reconstructed segments of the vessels. First, the method isolates the vascular system using controlled morphological growing and performs skeleton extraction and optimization. Then, around the optimized skeleton branches, it creates tubular objects optimized for quality and accuracy of matching with the originally isolated vascular data. Finally, it optimizes the joints on n-furcating vessel segments. As a result, the algorithm gives a complete description of shape, position in space, position relative to other segments, and other anatomical structures of each cerebrovascular system segment. Our method is highly customizable and in principle allows reconstructing vascular structures from any 2D or 3D data. The algorithm solves shortcomings of currently available methods including failures to reconstruct the vessel mesh in the proximity of junctions and is free of mesh collisions in high curvature vessels. It also introduces a number of optimizations in the vessel skeletonization leading to a more smooth and more accurate model of the vessel network. We have tested the method on 20 datasets from the public magnetic resonance angiography image database and show that the method allows for repeatable and robust segmentation of the vessel network and allows to compute vascular lateralization indices.

Assessment of shape-based features ability to predict the ascending aortic aneurysm growth

The current guidelines for the ascending aortic aneurysm (AsAA) treatment recommend surgery mainly according to the maximum diameter assessment. This criterion has already proven to be often inefficient in identifying patients at high risk of aneurysm growth and rupture. In this study, we propose a method to compute a set of local shape features that, in addition to the maximum diameter D, are intended to improve the classification performances for the ascending aortic aneurysm growth risk assessment. Apart from D, these are the ratio DCR between D and the length of the ascending aorta centerline, the ratio EILR between the length of the external and the internal lines and the tortuosity T. 50 patients with two 3D acquisitions at least 6 months apart were segmented and the growth rate (GR) with the shape features related to the first exam computed. The correlation between them has been investigated. After, the dataset was divided into two classes according to the growth rate value. We used six different classifiers with input data exclusively from the first exam to predict the class to which each patient belonged. A first classification was performed using only D and a second with all the shape features together. The performances have been evaluated by computing accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUROC) and positive (negative) likelihood ratio LHR+ (LHR-). A positive correlation was observed between growth rate and DCR (r = 0.511, p = 1.3e-4) and between GR and EILR (r = 0.472, p = 2.7e-4). Overall, the classifiers based on the four metrics outperformed the same ones based only on D. Among the diameter-based classifiers, k-nearest neighbours (KNN) reported the best accuracy (86%), sensitivity (55.6%), AUROC (0.74), LHR+ (7.62) and LHR- (0.48). Concerning the classifiers based on the four shape features, we obtained the best accuracy (94%), sensitivity (66.7%), specificity (100%), AUROC (0.94), LHR+ (+∞) and LHR- (0.33) with support vector machine (SVM). This demonstrates how automatic shape features detection combined with risk classification criteria could be crucial in planning the follow-up of patients with ascending aortic aneurysm and in predicting the possible dangerous progression of the disease.

A numerical investigation of the mechanics of intracranial aneurysms walls: Assessing the influence of tissue hyperelastic laws and heterogeneous properties on the stress and stretch fields

Numerical simulations have been extensively used in the past two decades for the study of intracranial aneurysms (IAs), a dangerous disease that occurs in the arteries that reach the brain and affect overall 3.2% of a population without comorbidity with up to 60% mortality rate, in case of rupture. The majority of those studies, though, assumed a rigid-wall model to simulate the blood flow. However, to also study the mechanics of IAs walls, it is important to assume a fluid-solid interaction (FSI) modeling. Progress towards more reliable FSI simulations is limited because FSI techniques pose severe numerical difficulties, but also due to scarce data on the mechanical behavior and material constants of IA tissue. Additionally, works that have investigated the impact of different wall modeling choices for patient-specific IAs geometries are a few and often with limited conclusions. Thus our present study investigated the effect of different modeling approaches to simulate the motion of an IA. We used three hyperelastic laws - the Yeoh law, the three-parameter Mooney-Rivlin law, and a Fung-like law with a single parameter - and two different ways of modeling the wall thickness and tissue mechanical properties - one assumed that both were uniform while the other accounted for the heterogeneity of the wall by using a "hemodynamics-driven" approach in which both thickness and material constants varied spatially with the cardiac-cycle-averaged hemodynamics. Pulsatile numerical simulations, with patient-specific vascular geometries harboring IAs, were carried out using the one-way fluid-solid interaction solution strategy implemented in solids4foam, an extension of OpenFOAM®, in which the blood flow is solved and applied as the driving force of the wall motion. We found that different wall morphology models yield smaller absolute differences in the mechanical response than different hyperelastic laws. Furthermore, the stretch levels of IAs walls were more sensitive to the hyperelastic and material constants than the stress. These findings could be used to guide modeling decisions on IA simulations, since the computational behavior of each law was different, for example, with the Yeoh law being the fastest to converge.

A PDE-regularized smoothing method for space-time data over manifolds with application to medical data

We propose an innovative statistical-numerical method to model spatio-temporal data, observed over a generic two-dimensional Riemanian manifold. The proposed approach consists of a regression model completed with a regularizing term based on the heat equation. The model is discretized through a finite element scheme set on the manifold, and solved by resorting to a fixed point-based iterative algorithm. This choice leads to a procedure which is highly efficient when compared with a monolithic approach, and which allows us to deal with massive datasets. After a preliminary assessment on simulation study cases, we investigate the performance of the new estimation tool in practical contexts, by dealing with neuroimaging and hemodynamic data.

Predicting tracheal work of breathing in neonates based on radiological and pulmonary measurements

Tracheomalacia is an airway condition in which the trachea excessively collapses during breathing. Neonates diagnosed with tracheomalacia require more energy to breathe, and the effect of tracheomalacia can be quantified by assessing flow-resistive work of breathing (WOB) in the trachea using computational fluid dynamics (CFD) modeling of the airway. However, CFD simulations are computationally expensive; the ability to instead predict WOB based on more straightforward measures would provide a clinically useful estimate of tracheal disease severity. The objective of this study is to quantify the WOB in the trachea using CFD and identify simple airway and/or clinical parameters that directly relate to WOB. This study included 30 neonatal intensive care unit subjects (15 with tracheomalacia and 15 without tracheomalacia). All subjects were imaged using ultrashort echo time (UTE) MRI. CFD simulations were performed using patient-specific data obtained from MRI (airway anatomy, dynamic motion, and airflow rates) to calculate the WOB in the trachea. Several airway and clinical measurements were obtained and compared with the tracheal resistive WOB. The maximum percent change in the tracheal cross-sectional area (ρ = 0.560, P = 0.001), average glottis cross-sectional area (ρ = -0.488, P = 0.006), minute ventilation (ρ = 0.613, P < 0.001), and lung tidal volume (ρ = 0.599, P < 0.001) had significant correlations with WOB. A multivariable regression model with three independent variables (minute ventilation, average glottis cross-sectional area, and minimum of the eccentricity index of the trachea) can be used to estimate WOB more accurately (R2 = 0.726). This statistical model may allow clinicians to estimate tracheal resistive WOB based on airway images and clinical data.NEW & NOTEWORTHY The work of breathing due to resistance in the trachea is an important metric for quantifying the effect of tracheal abnormalities such as tracheomalacia, but currently requires complex dynamic imaging and computational fluid dynamics simulation to calculate it. This study produces a method to predict the tracheal work of breathing based on readily available imaging and clinical metrics.

Automated generation of 0D and 1D reduced-order models of patient-specific blood flow

Three-dimensional (3D) cardiovascular fluid dynamics simulations typically require hours to days of computing time on a high-performance computing cluster. One-dimensional (1D) and lumped-parameter zero-dimensional (0D) models show great promise for accurately predicting blood bulk flow and pressure waveforms with only a fraction of the cost. They can also accelerate uncertainty quantification, optimization, and design parameterization studies. Despite several prior studies generating 1D and 0D models and comparing them to 3D solutions, these were typically limited to either 1D or 0D and a singular category of vascular anatomies. This work proposes a fully automated and openly available framework to generate and simulate 1D and 0D models from 3D patient-specific geometries, automatically detecting vessel junctions and stenosis segments. Our only input is the 3D geometry; we do not use any prior knowledge from 3D simulations. All computational tools presented in this work are implemented in the open-source software platform SimVascular. We demonstrate the reduced-order approximation quality against rigid-wall 3D solutions in a comprehensive comparison with N = 72 publicly available models from various anatomies, vessel types, and disease conditions. Relative average approximation errors of flows and pressures typically ranged from 1% to 10% for both 1D and 0D models, measured at the outlets of terminal vessel branches. In general, 0D model errors were only slightly higher than 1D model errors despite requiring only a third of the 1D runtime. Automatically generated ROMs can significantly speed up model development and shift the computational load from high-performance machines to personal computers.

Early Experience with Ovation Alto Stent-Graft

INTRODUCTION

Alto is the latest generation of the Ovation stent-graft platform for endovascular aneurysm repair (EVAR). Its ultra-low profile and its proximal sealing zone close to the lowest renal artery (≥7 mm), increase standard EVAR eligibility. We report early clinical and technical outcomes with the Alto stent-graft in our University Hospital Center, after CE Mark approval in August 2020.

METHODS

Seven patients (all male, mean age 76,1±6.2 years) underwent EVAR with Ovation Alto stent-graft between June 2021 and February 2022. All the EVAR procedures were performed by a team of vascular surgeons experienced on EVAR with previous generation of Ovation platform. Follow-up consisted of duplex ultrasounds examination (DUS) at 1, 3 and 6 months and of a 1-month control computed tomography angiography (CTA). Patients treated gave consent to participate in this case series and publication. A descriptive analysis of variables was performed. SPSS (Version 25; SPSS Inc, Chicago, IL, USA) and Excel (Microsoft Corporation, Redmond, WA, USA) were used for statistical analysis.

RESULTS

Most of patients had a fusiform AAA (n = 5; 71,4%). The median maximal transversal aortic diameter (DT) was 5,06 cm (range, 3,98 - 6,99). Due to hostile aortic neck anatomy, on-label EVAR was considered feasible only with Ovation Alto stent-graft. Narrow iliac arteries (<6 mm) were also present in 2 cases. All procedures were performed according to the instruction for use (IFU) of the device. Technical success was achieved in all cases. No type IA/IB/III endoleak occurred at completion angiography. No distal migration (>10 mm), but two distal displacements (≥ 2 mm) were observed at control CTA. During follow-up, DUS and CTA showed no type I/III endoleak, no stent-graft migration (>10 mm), and no proximal aortic neck variations (p=ns). 3 patients (42.8%) are under strict surveillance because of low-flow type II endoleak not associated with sac variations.

CONCLUSION

Our early experience shows promising technical and clinical success with Alto stent-graft. The proximal relocation of the proximal sealing rings and the ultra-low profile delivery system allow on-label EVAR in a wider range of aortic anatomies. Notwithstanding, further studies, metanalysis and prospective registries are mandatory to evaluate mid and long-term efficacy and safety of this latest Ovation platform.

Predicting the risk of postsplenectomy thrombosis in patients with portal hypertension using computational hemodynamics models: A proof-of-concept study

BACKGROUND

The high incidence of thrombosis in the portal venous system following splenectomy (a frequently adopted surgery for treating portal hypertension in patients with splenomegaly and hypersplenism) is a critical clinical issue. The aim of this study was to address whether quantification of postsplenectomy hemodynamics has potential value for assessing the risk of postsplenectomy thrombosis.

METHODS

Computational models were constructed for three portal hypertensive patients treated with splenectomy based on their preoperative clinical data to quantify hemodynamics in the portal venous system before and after splenectomy, respectively. Each patient was followed up for three or five months after surgery and examined with CT to screen potential thrombosis.

FINDINGS

The area ratio of wall regions exposed to low wall shear stress was small before splenectomy in all patients, which increased markedly after splenectomy and exhibited enlarged inter-patient differences. The largest area ratio of low wall shear stress and most severe flow stagnation after splenectomy were predicted for the patient suffering from postsplenectomy thrombosis, with the wall regions exposed to low wall shear stress corresponding well with the CT-detected distribution of thrombus. Further analyses revealed that postoperative hemodynamic characteristics were considerably influenced by the anatomorphological features of the portal venous system.

INTERPRETATION

Postoperative hemodynamic conditions in the portal venous system are highly patient-specific and have a potential link to postsplenectomy thrombosis, which indicates that patient-specific hemodynamic studies may serve as a complement to routine clinical assessments for refining risk stratification and postoperative patient management.

Chanalyzer: A Computational Geometry Approach for the Analysis of Protein Channel Shape and Dynamics

Morphological analysis of protein channels is a key step for a thorough understanding of their biological function and mechanism. In this respect, molecular dynamics (MD) is a very powerful tool, enabling the description of relevant biological events at the atomic level, which might elude experimental observations, and pointing to the molecular determinants thereof. In this work, we present a computational geometry-based approach for the characterization of the shape and dynamics of biological ion channels or pores to be used in combination with MD trajectories. This technique relies on the earliest works of Edelsbrunner and on the NanoShaper software, which makes use of the alpha shape theory to build the solvent-excluded surface of a molecular system in an aqueous solution. In this framework, a channel can be simply defined as a cavity with two entrances on the opposite sides of a molecule. Morphological characterization, which includes identification of the main axis, the corresponding local radius, and the detailed description of the global shape of the cavity, is integrated with a physico-chemical description of the surface facing the pore lumen. Remarkably, the possible existence or temporary appearance of fenestrations from the channel interior towards the outer lipid matrix is also accounted for. As a test case, we applied the present approach to the analysis of an engineered protein channel, the mechanosensitive channel of large conductance.

Biomechanics of Transcatheter Aortic Valve Implant

Transcatheter aortic valve implantation (TAVI) has grown exponentially within the cardiology and cardiac surgical spheres. It has now become a routine approach for treating aortic stenosis. Several concerns have been raised about TAVI in comparison to conventional surgical aortic valve replacement (SAVR). The primary concerns regard the longevity of the valves. Several factors have been identified which may predict poor outcomes following TAVI. To this end, the lesser-used finite element analysis (FEA) was used to quantify the properties of calcifications which affect TAVI valves. This method can also be used in conjunction with other integrated software to ascertain the functionality of these valves. Other imaging modalities such as multi-detector row computed tomography (MDCT) are now widely available, which can accurately size aortic valve annuli. This may help reduce the incidence of paravalvular leaks and regurgitation which may necessitate further intervention. Structural valve degeneration (SVD) remains a key factor, with varying results from current studies. The true incidence of SVD in TAVI compared to SAVR remains unclear due to the lack of long-term data. It is now widely accepted that both are part of the armamentarium and are not mutually exclusive. Decision making in terms of appropriate interventions should be undertaken via shared decision making involving heart teams.

Simulation of intra-saccular devices for pre-operative device size selection: Method and validation for sizing and porosity simulation

Intra-saccular devices (ID) are novel braided devices used for complex intracranial aneurysms treatment. Treatment success is associated with correct device size selection. A technique that predicts the ID size within the aneurysm before intervention will provide a powerful computational tool to aid the interventionist during device selection. We present a method to calculate the device's final height, radial expansion and porosity within the patient's anatomy, which allows assessing different device sizes before treatment takes place. The proposed sizing technique was tested in-vitro and in real patient's geometries obtained from 3DRA angiographic images of 8 unruptured aneurysms previously treated with IDs. The obtained simulated height was compared to the real height, with a mean error of less than 0.28 mm (±0.44). The porosity calculation method was tested in-vitro with an error of 0.02 (±0.022). The results of both sizing and porosity experiments resemble well measures from real patients. This methodology could be used before treatment to provide the interventionist with additional information that allows selecting the device that best fits the patient's aneurysm to be treated.

Bifurcation geometry remodelling of vessels in de novo and growing intracranial aneurysms: a multicenter study

BACKGROUND

Geometrical parameters, including arterial bifurcation angle, tortuosity, and arterial diameters, have been associated with the pathophysiology of intracranial aneurysm (IA) formation. The aim of this study was to investigate whether these parameters were present before or if they resulted from IA formation and growth.

METHODS

Patients from nine academic centers were retrospectively identified if they presented with a de novo IA or a significant IA growth on subsequent imaging. For each patient, geometrical parameters were extracted using a semi-automated algorithm and compared between bifurcations with IA formation or growth (aneurysmal group), and their contralateral side without IA (control group). These parameters were compared at two different times using univariable models, multivariable models, and a sensitivity analysis with paired comparison.

RESULTS

46 patients were included with 21 de novo IAs (46%) and 25 significant IA growths (54%). The initial angle was not different between the aneurysmal and control groups (129.7±42.1 vs 119.8±34.3; p=0.264) but was significantly wider at the final stage (140.4±40.9 vs 121.5±34.1; p=0.032), with a more important widening of the aneurysmal angle (10.8±15.8 vs 1.78±7.38; p=0.001). Variations in other parameters were not significant. These results were confirmed by paired comparisons.

CONCLUSION

Our study suggests that wider bifurcation angles that have long been deemed causal factors for IA formation or growth may be secondary to IA formation at pathologic bifurcation sites. This finding has implications for our understanding of IA formation pathophysiology.

Shape Trumps Size: Image-Based Morphological Analysis Reveals That the 3D Shape Discriminates Intracranial Aneurysm Disease Status Better Than Aneurysm Size

Background

To date, it remains difficult for clinicians to reliably assess the disease status of intracranial aneurysms. As an aneurysm's 3D shape is strongly dependent on the underlying formation processes, it is believed that the presence of certain shape features mirrors the disease status of the aneurysm wall. Currently, clinicians associate irregular shape with wall instability. However, no consensus exists about which shape features reliably predict instability. In this study, we present a benchmark to identify shape features providing the highest predictive power for aneurysm rupture status.

Methods

3D models of aneurysms were extracted from medical imaging data (3D rotational angiographies) using a standardized protocol. For these aneurysm models, we calculated a set of metrics characterizing the 3D shape: Geometry indices (such as undulation, ellipticity and non-sphericity); writhe- and curvature-based metrics; as well as indices based on Zernike moments. Using statistical learning methods, we investigated the association between shape features and aneurysm disease status. This processing was applied to a clinical dataset of 750 aneurysms (261 ruptured, 474 unruptured) registered in the AneuX morphology database. We report here statistical performance metrics [including the area under curve (AUC)] for morphometric models to discriminate between ruptured and unruptured aneurysms.

Results

The non-sphericity index NSI (AUC = 0.80), normalized Zernike energies ZNsurf (AUC = 0.80) and the modified writhe-index W¯meanL1 (AUC = 0.78) exhibited the strongest association with rupture status. The combination of predictors further improved the predictive performance (without location: AUC = 0.82, with location AUC = 0.87). The anatomical location was a good predictor for rupture status on its own (AUC = 0.78). Different protocols to isolate the aneurysm dome did not affect the prediction performance. We identified problems regarding generalizability if trained models are applied to datasets with different selection biases.

Conclusions

Morphology provided a clear indication of the aneurysm disease status, with parameters measuring shape (especially irregularity) being better predictors than size. Quantitative measurement of shape, alone or in conjunction with information about aneurysm location, has the potential to improve the clinical assessment of intracranial aneurysms.

Geometric features of middle cerebral artery are associated with spontaneous basal ganglia intracerebral haemorrhage

BACKGROUND AND PURPOSE

Haemodynamics around the middle cerebral artery (MCA) and lenticulostriate arteries is believed to play important roles in the vascular rupture and local haemodynamics is subject to vascular geometry. Nonetheless, the relationship between the geometric features of MCA and spontaneous basal ganglia intracerebral haemorrhage (ICH) has not been investigated. To examine the relationship between the MCA geometric features and spontaneous basal ganglia ICH.

METHODS

This study was of retrospective and observational nature. The study recruited 158 consecutive hospitalised patients with consecutive CT-confirmed unilateral spontaneous basal ganglia ICH. Clinical data were extracted from electronic medical records, and imaging data were evaluated by two trained radiologists. The MCA-related geometric features were examined and their relationship with spontaneous basal ganglia ICH was analysed. Haemodynamic analyses under different MCA structural features were conducted.

RESULTS

Compared with the contralateral MCA, the ipsilateral MCA had greater M1 diameter ratio (proximal/distal) and a smaller M1/M2 angle and MCA bifurcation angle (p＜0.01). Imaging study showed differences in the MCA shape in both sides on coronal plane (p＜0.05). These MCA features were significantly correlated with the spontaneous ICH in basal ganglia. The greater M1 diameter ratio (proximal/distal), the inferior-oriented M1, the smaller M1/M2 angle and the superior-oriented M1 conditions increased the pressure, from high to low. The greater M1 diameter ratio (proximal/distal) and the inferior-oriented M1 increased the shear stress at the distal end of M1 segment.

CONCLUSIONS

The geometric features of MCA were significantly related to the spontaneous ICH in basal ganglia. The risk of haemorrhage, from high to low, included the greater M1 diameter ratio (proximal/distal), the inferior-oriented M1 (distal end), the smaller M1/M2 angle and the superior-oriented M1. Mechanistically, these vascular structural features contribute to increased vascular wall pressure and shear stress, which eventually lead to haemorrhage.

Analysis of Cerebral Aneurysm Wall Tension and Enhancement Using Finite Element Analysis and High-Resolution Vessel Wall Imaging

Biomechanical computational simulation of intracranial aneurysms has become a promising method for predicting features of instability leading to aneurysm growth and rupture. Hemodynamic analysis of aneurysm behavior has helped investigate the complex relationship between features of aneurysm shape, morphology, flow patterns, and the proliferation or degradation of the aneurysm wall. Finite element analysis paired with high-resolution vessel wall imaging can provide more insight into how exactly aneurysm morphology relates to wall behavior, and whether wall enhancement can describe this phenomenon. In a retrospective analysis of 23 unruptured aneurysms, finite element analysis was conducted using an isotropic, homogenous third order polynomial material model. Aneurysm wall enhancement was quantified on 2D multiplanar views, with 14 aneurysms classified as enhancing (CR_stalk≥0.6) and nine classified as non-enhancing. Enhancing aneurysms had a significantly higher 95th percentile wall tension (μ = 0.77 N/cm) compared to non-enhancing aneurysms (μ = 0.42 N/cm, p < 0.001). Wall enhancement remained a significant predictor of wall tension while accounting for the effects of aneurysm size (p = 0.046). In a qualitative comparison, low wall tension areas concentrated around aneurysm blebs. Aneurysms with irregular morphologies may show increased areas of low wall tension. The biological implications of finite element analysis in intracranial aneurysms are still unclear but may provide further insights into the complex process of bleb formation and aneurysm rupture.