Quantitative analysis of flow vortices: differentiation of unruptured and ruptured medium-sized middle cerebral artery aneurysms

Hemodynamic analysis of thrombosed intracranial aneurysms: a comparative correlation study

A small fraction of intracranial aneurysms (IA) contains intrasaccular thrombosis (IST). This study explores the hemodynamic causes of IST formation in IAs. We performed computational hemodynamic analysis in 26 IAs: 13 thrombosed and 13 non-thrombosed. The IAs in the two cohorts were matching in size and anatomical location. The computational hemodynamic analysis used "patient-specific" IA geometries derived from 3D magnetic resonance imaging. A comprehensive hemodynamic analysis was conducted using commonly used hemodynamic metrics (e.g., wall shear stress [WSS] and its derivative), flow vortex analysis, and velocity informatics. We observed that more flow eddies and endothelial cell activation potential (ECAP) are present in thrombosed IAs. The flow eddies in thrombosed IAs change size and location over a cardiac cycle more significantly than those in non-thrombosed IAs. These two factors, coupled with more flow stagnation and positive WSS divergence in the thrombosed IAs, promoted thrombus formation in the thrombosed IA cohort.

Investigating the role of blood models in predicting rupture status of intracranial aneurysms

Purpose. Selecting patients with high-risk intracranial aneurysms (IAs) is of clinical importance. Recent work in machine learning-based (ML) predictive modeling has demonstrated that lesion-specific hemodynamics within IAs can be combined with other information to provide critical insights for assessing rupture risk. However, how the adoption of blood rheology models (i.e., Newtonian and Non-Newtonian blood models) may influence ML-based predictive modeling of IA rupture risk has not been investigated.Methods and Materials.In this study, we conducted transient CFD simulations using Newtonian and non-Newtonian rheology (Carreau-Yasuda [CY]) models on a large cohort of 'patient-specific' IA geometries (>100) under pulsatile flow conditions to investigate how each blood model may affect the characterization of the IAs' rupture status. Key hemodynamic parameters were analyzed and compared, including wall shear stress (WSS) and vortex-based parameters. In addition, velocity-informatics features extracted from the flow velocity were utilized to train a support vector machine (SVM) model for rupture status prediction.Results.Our findings demonstrate significant differences between the two models (i.e., Newtonian versus CY) regarding the WSS-related metrics. In contrast, the parameters derived from the flow vortices and velocity informatics agree. Similar to other studies, using a non-Newtonian CY model results in lower peak WSS and higher oscillatory shear index (OSI) values. Furthermore, integrating velocity informatics and machine learning achieved robust performance for both blood models (area under the curve [AUC] ˃0.85).Conclusions.Our preliminary study found that ML-based rupture status prediction derived from velocity informatics and geometrical parameters yielded comparable results despite differences observed in aneurysmal hemodynamics using two blood rheology models (i.e., Newtonian versus CY).

Improving Prediction of Intracranial Aneurysm Rupture Status Using Temporal Velocity-Informatics

This study uses a spatial pattern analysis of time-resolved aneurysmal velocity fields to enhance the characterization of intracranial aneurysms' (IA) rupture status. We name this technique temporal velocity-informatics (TVI). In this study, using imaging data obtained from 112 subjects harboring IAs with known rupture status, we reconstructed 3D models to get aneurysmal velocity data by performing computational fluid dynamics (CFD) simulations and morphological information. TVI analyses were conducted for time-resolved velocity fields to quantitatively obtain spatial and temporal flow disturbance. Lastly, we employed four machine learning (ML) methods (e.g., support vector machine [SVM]) to evaluate the prediction performance of the proposed TVI. Overall, the SVM's prediction with TVI performed the best: an area under the curve (AUC) value of 0.92 and a total accuracy of 86%. With TVI, the SVM classifier correctly identified 77 and 92% of ruptured and unruptured IAs, respectively.

Systematic Review of Radiomics and Artificial Intelligence in Intracranial Aneurysm Management

Intracranial aneurysms, with an annual incidence of 2%-3%, reflect a rare disease associated with significant mortality and morbidity risks when ruptured. Early detection, risk stratification of high-risk subgroups, and prediction of patient outcomes are important to treatment. Radiomics is an emerging field using the quantification of medical imaging to identify parameters beyond traditional radiology interpretation that may offer diagnostic or prognostic significance. The general radiomic workflow involves image normalization and segmentation, feature extraction, feature selection or dimensional reduction, training of a predictive model, and validation of the said model. Artificial intelligence (AI) techniques have shown increasing interest in applications toward vascular pathologies, with some commercially successful software including AiDoc, RapidAI, and Viz.AI, as well as the more recent Viz Aneurysm. We performed a systematic review of 684 articles and identified 84 articles exploring the applications of radiomics and AI in aneurysm treatment. Most studies were published between 2018 and 2024, with over half of articles in 2022 and 2023. Studies included categories such as aneurysm diagnosis (25.0%), rupture risk prediction (50.0%), growth rate prediction (4.8%), hemodynamic assessment (2.4%), clinical outcome prediction (11.9%), and occlusion or stenosis assessment (6.0%). Studies utilized molecular data (2.4%), radiologic data alone (51.2%), clinical data alone (28.6%), and combined radiologic and clinical data (17.9%). These results demonstrate the current status of this emerging and exciting field. An increased pace of innovation in this space is likely with the expansion of clinical applications of radiomics and AI in multiple vascular pathologies.

Developing a nearly automated open-source pipeline for conducting computational fluid dynamics simulations in anterior brain vasculature: a feasibility study

Intracranial aneurysms (IA) pose significant health risks and are often challenging to manage. Computational fluid dynamics (CFD) simulation has emerged as a powerful tool for understanding lesion-specific hemodynamics in and around IAs, aiding in the clinical management of patients with an IA. However, the current workflow of CFD simulations is time-consuming, complex, and labor-intensive and, thus, does not fit the clinical environment. To address these challenges, we have developed a semi-automated pipeline integrating multiple open-source software packages to streamline the CFD simulation process. Specifically, the study utilized medical angiography data from 18 patients. An in-house open-source DL image segmentation model (ARU-Net) was employed to generate 3D computer models of the anterior circulation. The segmented intracranial vasculature models, including IAs, were further refined using the Vascular Modeling Toolkit (VMTK), an open-source Python package. This step involved smoothing the surface of the models and extending the inlet and outlet regions to ensure a realistic representation of the vascular geometry. The refined vascular models were then converted into computational meshes using an open-source mesh generator known as TetGen. This process was nearly automated and required minimal user interaction(s). Blood flow simulations of the cerebral vascular models were performed using established SimVascular solvers (an open-source finite element platform for vascular applications) through an application programming interface (API). The CFD simulation process was also conducted using the manual workflow for comparative purposes. The initial assessment compared the geometries derived from manual and DL-based segmentation. The DL-based segmentation demonstrated reliable performance, closely aligning with manually segmented results, evidenced by excellent Pearson correlation coefficient (PCC) values and low relative difference (RD) values ranging from 3% to 10% between the computed geometrical variables derived from both methods. The statistical analysis of the computed hemodynamic variables, including velocity informatics and WSS-related variables, indicated good to excellent reliability for most parameters (e.g., ICC of 0.85-0.95). Given the data investigated, the proposed automated workflow streamlines the process of conducting CFD simulations. It generates results consistent with the current standard manual CFD protocol while minimizing dependence on user input.

Computational Hemodynamics-Based Growth Prediction for Small Abdominal Aortic Aneurysms: Laminar Simulations Versus Large Eddy Simulations

Prior studies have shown that computational fluid dynamics (CFD) simulations help assess patient-specific hemodynamics in abdominal aortic aneurysms (AAAs); patient-specific hemodynamic stressors are frequently used to predict an AAA's growth. Previous studies have utilized both laminar and turbulent simulation models to simulate hemodynamics. However, the impact of different CFD simulation models on the predictive modeling of AAA growth remains unknown and is thus the knowledge gap that motivates this study. Specifically, CFD simulations were performed for 70 AAA models derived from 70 patients' computed tomography angiography (CTA) data with known growth status (i.e., fast-growing [> 5 mm/yr] or slowly growing [< 5 mm/yr]). We used laminar and large eddy simulation (LES) models to obtain hemodynamic parameters to predict AAAs' growth status. Predicting the growth status of AAAs was based on morphological, hemodynamic, and patient health parameters in conjunction with three classical machine learning (ML) classifiers, namely, support vector machine (SVM), K-nearest neighbor (KNN), and generalized linear model (GLM). Our preliminary results estimated aneurysmal flow stability and wall shear stress (WSS) were comparable in both laminar and LES flow simulations. Moreover, computed WSS and velocity-related hemodynamic variables obtained from the laminar and LES simulations showed comparable abilities in differentiating the growth status of AAAs. More importantly, the predictive modeling performance of the three ML classifiers mentioned above was similar, with less than a 2% difference observed (p-value > 0.05). In closing, our findings suggest that two different flow simulations investigated did not significantly affect outcomes of computational hemodynamics and predictive modeling of AAAs' growth status, given the data investigated.

S-Net: a multiple cross aggregation convolutional architecture for automatic segmentation of small/thin structures for cardiovascular applications

With the success of U-Net or its variants in automatic medical image segmentation, building a fully convolutional network (FCN) based on an encoder-decoder structure has become an effective end-to-end learning approach. However, the intrinsic property of FCNs is that as the encoder deepens, higher-level features are learned, and the receptive field size of the network increases, which results in unsatisfactory performance for detecting low-level small/thin structures such as atrial walls and small arteries. To address this issue, we propose to keep the different encoding layer features at their original sizes to constrain the receptive field from increasing as the network goes deeper. Accordingly, we develop a novel S-shaped multiple cross-aggregation segmentation architecture named S-Net, which has two branches in the encoding stage, i.e., a resampling branch to capture low-level fine-grained details and thin/small structures and a downsampling branch to learn high-level discriminative knowledge. In particular, these two branches learn complementary features by residual cross-aggregation; the fusion of the complementary features from different decoding layers can be effectively accomplished through lateral connections. Meanwhile, we perform supervised prediction at all decoding layers to incorporate coarse-level features with high semantic meaning and fine-level features with high localization capability to detect multi-scale structures, especially for small/thin volumes fully. To validate the effectiveness of our S-Net, we conducted extensive experiments on the segmentation of cardiac wall and intracranial aneurysm (IA) vasculature, and quantitative and qualitative evaluations demonstrated the superior performance of our method for predicting small/thin structures in medical images.

Augmenting Prediction of Intracranial Aneurysms' Risk Status Using Velocity-Informatics: Initial Experience

Our primary goal here is to demonstrate that innovative analytics of aneurismal velocities, named velocity-informatics, enhances intracranial aneurysm (IA) rupture status prediction. 3D computer models were generated using imaging data from 112 subjects harboring anterior IAs (4-25 mm; 44 ruptured and 68 unruptured). Computational fluid dynamics simulations and geometrical analyses were performed. Then, computed 3D velocity vector fields within the IA dome were processed for velocity-informatics. Four machine learning methods (support vector machine, random forest, generalized linear model, and GLM with Lasso or elastic net regularization) were employed to assess the merits of the proposed velocity-informatics. All 4 ML methods consistently showed that, with velocity-informatics metrics, the area under the curve and prediction accuracy both improved by approximately 0.03. Overall, with velocity-informatics, the support vector machine's prediction was most promising: an AUC of 0.86 and total accuracy of 77%, with 60% and 88% of ruptured and unruptured IAs being correctly identified, respectively.

Deep-learning-based image segmentation for image-based computational hemodynamic analysis of abdominal aortic aneurysms: a comparison study

Computational hemodynamics is increasingly being used to quantify hemodynamic characteristics in and around abdominal aortic aneurysms (AAA) in a patient-specific fashion. However, the time-consuming manual annotation hinders the clinical translation of computational hemodynamic analysis. Thus, we investigate the feasibility of using deep-learning-based image segmentation methods to reduce the time required for manual segmentation. Two of the latest deep-learning-based image segmentation methods, ARU-Net and CACU-Net, were used to test the feasibility of automated computer model creation for computational hemodynamic analysis. Morphological features and hemodynamic metrics of 30 computed tomography angiography (CTA) scans were compared between pre-dictions and manual models. The DICE score for both networks was 0.916, and the correlation value was above 0.95, indicating their ability to generate models comparable to human segmentation. The Bland-Altman analysis shows a good agreement between deep learning and manual segmentation results. Compared with manual (computational hemodynamics) model recreation, the time for automated computer model generation was significantly reduced (from ∼2 h to ∼10 min). Automated image segmentation can significantly reduce time expenses on the recreation of patient-specific AAA models. Moreover, our study showed that both CACU-Net and ARU-Net could accomplish AAA segmentation, and CACU-Net outperformed ARU-Net in terms of accuracy and time-saving.

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.

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.

Can we explain machine learning-based prediction for rupture status assessments of intracranial aneurysms?

Although applying machine learning (ML) algorithms to rupture status assessment of intracranial aneurysms (IA) has yielded promising results, the opaqueness of some ML methods has limited their clinical translation. We presented the first explainability comparison of six commonly used ML algorithms: multivariate logistic regression (LR), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGBoost), multi-layer perceptron neural network (MLPNN), and Bayesian additive regression trees (BART). A total of 112 IAs with known rupture status were selected for this study. The ML-based classification used two anatomical features, nine hemodynamic parameters, and thirteen morphologic variables. We utilized permutation feature importance, local interpretable model-agnostic explanations (LIME), and SHapley Additive exPlanations (SHAP) algorithms to explain and analyze 6 Ml algorithms. All models performed comparably: LR area under the curve (AUC) was 0.71; SVM AUC was 0.76; RF AUC was 0.73; XGBoost AUC was 0.78; MLPNN AUC was 0.73; BART AUC was 0.73. Our interpretability analysis demonstrated consistent results across all the methods; i.e., the utility of the top 12 features was broadly consistent. Furthermore, contributions of 9 important features (aneurysm area, aneurysm location, aneurysm type, wall shear stress maximum during systole, ostium area, the size ratio between aneurysm width, (parent) vessel diameter, one standard deviation among time-averaged low shear area, and one standard deviation of temporally averaged low shear area less than 0.4 Pa) were nearly the same. This research suggested that ML classifiers can provide explainable predictions consistent with general domain knowledge concerning IA rupture. With the improved understanding of ML algorithms, clinicians' trust in ML algorithms will be enhanced, accelerating their clinical translation.

Impact of spatial and temporal stability of flow vortices on vascular endothelial cells

PURPOSE

Intracranial aneurysms (IAs) are pathological dilations of cerebrovascular vessels due to degeneration of the mechanical strength of the arterial wall, precluded by altered cellular functionality. The presence of swirling hemodynamic flow (vortices) is known to alter vascular endothelial cell (EC) morphology and protein expression indicative of IAs. Unfortunately, less is known if vortices with varied spatial and temporal stability lead to differing levels of EC change. The aim of this work is to investigate vortices of varying spatial and temporal stability impact on ECs.

METHODS

Vortex and EC interplay was investigated by a novel combination of parallel plate flow chamber (PPFC) design and computational analysis. ECs were exposed to laminar (7.5 dynes/[Formula: see text] wall shear stress) or low (<1 dynes/[Formula: see text]) stress vortical flow using PPFCs. Immunofluorescent imaging analyzed EC morphology, while ELISA tests quantified VE-cadherin (cell-cell adhesion), VCAM-1 (macrophage-EC adhesion), and cleaved caspase-3 (apoptotic signal) expression. PPFC flow was simulated, and vortex stability was calculated via the temporally averaged degree of (volume) overlap (TA-DVO) of vortices within a given area.

RESULTS

EC morphological changes were independent of vortex stability. Increased stability promoted VE-cadherin degradation (correlation coefficient r = [Formula: see text]0.84) and 5-fold increased cleaved caspase-3 post 24 h in stable (TA-DVO 0.736 ± 0.05) vs unstable (TA-DVO 0.606 [Formula: see text]0.2) vortices. ECs in stable vortices displayed a 4.5-fold VCAM-1 increase than unstable counterparts after 12 h.

CONCLUSION

This work demonstrates highly stable disturbed flow imparts increased inflammatory signaling, degraded cell-cell adhesion, and increased cellular apoptosis than unstable vortices. Such knowledge offers novel insight toward understanding IA development and rupture.

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.

An Integrated Model Combining Machine Learning and Deep Learning Algorithms for Classification of Rupture Status of IAs

Background

The rupture risk assessment of intracranial aneurysms (IAs) is clinically relevant. How to accurately assess the rupture risk of IAs remains a challenge in clinical decision-making.

Purpose

We aim to build an integrated model to improve the assessment of the rupture risk of IAs.

Materials and Methods

A total of 148 (39 ruptured and 109 unruptured) IA subjects were retrospectively computed with computational fluid dynamics (CFDs), and the integrated models were proposed by combining machine learning (ML) and deep learning (DL) algorithms. ML algorithms that include random forest (RF), k-nearest neighbor (KNN), XGBoost (XGB), support vector machine (SVM), and LightGBM were, respectively, adopted to classify ruptured and unruptured IAs. A Pointnet DL algorithm was applied to extract hemodynamic cloud features from the hemodynamic clouds obtained from CFD. Morphological variables and hemodynamic parameters along with the extracted hemodynamic cloud features were acted as the inputs to the classification models. The classification results with and without hemodynamic cloud features are computed and compared.

Results

Without consideration of hemodynamic cloud features, the classification accuracy of RF, KNN, XGB, SVM, and LightGBM was 0.824, 0.759, 0.839, 0.860, and 0.829, respectively, and the AUCs of them were 0.897, 0.584, 0.892, 0.925, and 0.890, respectively. With the consideration of hemodynamic cloud features, the accuracy successively increased to 0.908, 0.873, 0.900, 0.926, and 0.917. Meanwhile, the AUCs reached 0.952, 0.881, 0.950, 0.969, and 0.965 eventually. Adding consideration of hemodynamic cloud features, the SVM could perform best with the highest accuracy of 0.926 and AUC of 0.969, respectively.

Conclusion

The integrated model combining ML and DL algorithms could improve the classification of IAs. Adding consideration of hemodynamic cloud features could bring more accurate classification, and hemodynamic cloud features were important for the discrimination of ruptured IAs.