Initial Clinical Experience with AView-A Clinical Computational Platform for Intracranial Aneurysm Morphology, Hemodynamics, and Treatment Management

Deep learning-based cerebral aneurysm segmentation and morphological analysis with three-dimensional rotational angiography

BACKGROUND

The morphological assessment of cerebral aneurysms based on cerebral angiography is an essential step when planning strategy and device selection in endovascular treatment, but manual evaluation by human raters only has moderate interrater/intrarater reliability.

METHODS

We collected data for 889 cerebral angiograms from consecutive patients with suspected cerebral aneurysms at our institution from January 2017 to October 2021. The automatic morphological analysis model was developed on the derivation cohort dataset consisting of 388 scans with 437 aneurysms, and the performance of the model was tested on the validation cohort dataset consisting of 96 scans with 124 aneurysms. Five clinically important parameters were automatically calculated by the model: aneurysm volume, maximum aneurysm size, neck size, aneurysm height, and aspect ratio.

RESULTS

On the validation cohort dataset the average aneurysm size was 7.9±4.6 mm. The proposed model displayed high segmentation accuracy with a mean Dice similarity index of 0.87 (median 0.93). All the morphological parameters were significantly correlated with the reference standard (all P<0.0001; Pearson correlation analysis). The difference in the maximum aneurysm size between the model prediction and reference standard was 0.5±0.7 mm (mean±SD). The difference in neck size between the model prediction and reference standard was 0.8±1.7 mm (mean±SD).

CONCLUSION

The automatic aneurysm analysis model based on angiography data exhibited high accuracy for evaluating the morphological characteristics of cerebral aneurysms.

Stability of unruptured intracranial aneurysms in the anterior circulation: nomogram models for risk assessment

OBJECTIVE

The probable stability of the lesion is critical in guiding treatment decisions in unruptured intracranial aneurysms (IAs). The authors aimed to develop multidimensional predictive models for the stability of unruptured IAs.

METHODS

Patients with unruptured IAs in the anterior circulation were prospectively enrolled and regularly followed up. Clinical data were collected, IA morphological features were assessed, and adjacent hemodynamic features were quantified with patient-specific computational fluid dynamics modeling. Based on multivariate logistic regression analyses, nomograms incorporating these factors were developed in a primary cohort (patients enrolled between January 2017 and February 2018) to predict aneurysm rupture or growth within 2 years. The predictive accuracies of the nomograms were compared with the population, hypertension, age, size, earlier rupture, and site (PHASES) and earlier subarachnoid hemorrhage, location, age, population, size, and shape (ELAPSS) scores and validated in the validation cohort (patients enrolled between March and October 2018).

RESULTS

Among 231 patients with 272 unruptured IAs in the primary cohort, hypertension, aneurysm location, irregular shape, size ratio, normalized wall shear stress average, and relative resident time were independently related to the 2-year stability of unruptured IAs. The nomogram including clinical, morphological, and hemodynamic features (C+M+H nomogram) had the highest predictive accuracy (c-statistic 0.94), followed by the nomogram including clinical and morphological features (C+M nomogram; c-statistic 0.89), PHASES score (c-statistic 0.68), and ELAPSS score (c-statistic 0.58). Similarly, the C+M+H nomogram had the highest predictive accuracy (c-statistic 0.94) in the validation cohort (85 patients with 97 unruptured IAs).

CONCLUSIONS

Hemodynamics have predictive values for 2-year stability of unruptured IAs treated conservatively. Multidimensional nomograms have significantly higher predictive accuracies than conventional risk prediction scores.

Rebleeding of Ruptured Intracranial Aneurysm After Admission: A Multidimensional Nomogram Model to Risk Assessment

Objective

Rebleeding is recognized as the main cause of mortality after intracranial aneurysm rupture. Though timely intervention can prevent poor prognosis, there is no agreement on the surgical priority and choosing medical treatment for a short period after rupture. The aim of this study was to investigate the risk factors related to the rebleeding after admission and establish predicting models for better clinical decision-making.

Methods

The patients with ruptured intracranial aneurysms (RIAs) between January 2018 and September 2020 were reviewed. All patients fell to the primary and the validation cohort by January 2020. The hemodynamic parameters were determined through the computational fluid dynamics simulation. Cox regression analysis was conducted to identify the risk factors of rebleeding. Based on the independent risk factors, nomogram models were built, and their predicting accuracy was assessed by using the area under the curves (AUCs).

Result

A total of 577 patients with RIAs were enrolled in this present study, 86 patients of them were identified as undergoing rebleeding after admission. Thirteen parameters were identified as significantly different between stable and rebleeding aneurysms in the primary cohort. Cox regression analysis demonstrated that six parameters, including hypertension [hazard ratio (HR), 2.54; P = 0.044], bifurcation site (HR, 1.95; P = 0.013), irregular shape (HR, 4.22; P = 0.002), aspect ratio (HR, 12.91; P < 0.001), normalized wall shear stress average (HR, 0.16; P = 0.002), and oscillatory stress index (HR, 1.14; P < 0.001) were independent risk factors related to the rebleeding after admission. Two nomograms were established, the nomogram including clinical, morphological, and hemodynamic features (CMH nomogram) had the highest predicting accuracy (AUC, 0.92), followed by the nomogram including clinical and morphological features (CM nomogram; AUC, 0.83), ELAPSS score (AUC, 0.61), and PHASES score (AUC, 0.54). The calibration curve for the probability of rebleeding showed good agreement between prediction by nomograms and actual observation. In the validation cohort, the discrimination of the CMH nomogram was superior to the other models (AUC, 0.93 vs. 0.86, 0.71 and 0.48).

Conclusion

We presented two nomogram models, named CMH nomogram and CM nomogram, which could assist in identifying the RIAs with high risk of rebleeding.

Optimal Woven EndoBridge (WEB) Device Size Selection Using Automated Volumetric Software

INTRODUCTION

Selecting the appropriate Woven EndoBridge (WEB) device sizing for the treatment of wide-neck bifurcation aneurysms (WNBAs) remains challenging. The aim of this study was to evaluate different volumetric-based imaging methodologies to predict an accurate WEB device size selection to result in a successful implantation.

METHODS

All consecutive patients treated with WEB devices for intracranial aneurysms from January 2019 to June 2020 were included. Aneurysm dimensions to calculate aneurysm volumes were measured using three different modalities: automated three-dimensional (3D) digital subtraction angiography (DSA), manual 3D DSA, and two-dimensional (2D) DSA. The device-aneurysm volume (DAV) ratio was defined as device volume divided by the aneurysm volume. WEB volumes and the DAV ratios were used for assessing the device implantation success and follow-up angiographic outcomes at six months. Pearson correlation, Wilcoxon Rank Sum test, and density approximations were used for estimating the WEB volumes and the imaging modality volumes for successful implantation.

RESULTS

A total of 41 patients with 43 aneurysms were included in the study. WEB device and aneurysm volume correlation coefficient was highest for 3D automatic (r = 0.943), followed by 3D manual (r = 0.919), and 2D DSA (r = 0.882) measurements. Measured median volumes were significantly different for 3D automatic and 2D DSA (p = 0.017). The highest rate of successful implantation (87.5%) was between 0.6 and 0.8 DAV ratio.

CONCLUSION

Pre-procedural assessment of DAV ratios may increase WEB device implantation success. Our results suggest that volumetric measurements, especially using automated 3D volumes of the aneurysms, can assist in accurate WEB device size selection.

Extending statistical learning for aneurysm rupture assessment to Finnish and Japanese populations using morphology, hemodynamics, and patient characteristics

OBJECTIVE

Incidental aneurysms pose a challenge for physicians, who need to weigh the rupture risk against the risks associated with treatment and its complications. A statistical model could potentially support such treatment decisions. A recently developed aneurysm rupture probability model performed well in the US data used for model training and in data from two European cohorts for external validation. Because Japanese and Finnish patients are known to have a higher aneurysm rupture risk, the authors' goals in the present study were to evaluate this model using data from Japanese and Finnish patients and to compare it with new models trained with Finnish and Japanese data.

METHODS

Patient and image data on 2129 aneurysms in 1472 patients were used. Of these aneurysm cases, 1631 had been collected mainly from US hospitals, 249 from European (other than Finnish) hospitals, 147 from Japanese hospitals, and 102 from Finnish hospitals. Computational fluid dynamics simulations and shape analyses were conducted to quantitatively characterize each aneurysm's shape and hemodynamics. Next, the previously developed model's discrimination was evaluated using the Finnish and Japanese data in terms of the area under the receiver operating characteristic curve (AUC). Models with and without interaction terms between patient population and aneurysm characteristics were trained and evaluated including data from all four cohorts obtained by repeatedly randomly splitting the data into training and test data.

RESULTS

The US model's AUC was reduced to 0.70 and 0.72, respectively, in the Finnish and Japanese data compared to 0.82 and 0.86 in the European and US data. When training the model with Japanese and Finnish data, the average AUC increased only slightly for the Finnish sample (to 0.76 ± 0.16) and Finnish and Japanese cases combined (from 0.74 to 0.75 ± 0.14) and decreased for the Japanese data (to 0.66 ± 0.33). In models including interaction terms, the AUC in the Finnish and Japanese data combined increased significantly to 0.83 ± 0.10.

CONCLUSIONS

Developing an aneurysm rupture prediction model that applies to Japanese and Finnish aneurysms requires including data from these two cohorts for model training, as well as interaction terms between patient population and the other variables in the model. When including this information, the performance of such a model with Japanese and Finnish data is close to its performance with US or European data. These results suggest that population-specific differences determine how hemodynamics and shape associate with rupture risk in intracranial aneurysms.

Relationship between haemodynamic changes and outcomes of intracranial aneurysms after implantation of the pipeline embolisation device: a single centre study

BACKGROUND

Intracranial aneurysms are increasingly being treated by the placement of flow diverters; however, the factors affecting the outcome of aneurysms treated using flow diverters remain unclarified.

METHODS

The present study investigated 94 aneurysms treated with pipeline embolisation device placement, and used a computational fluid dynamics method to explore the factors influencing the outcome of aneurysms.

RESULTS

Seventy-six completely occluded aneurysms and 18 incompletely occluded aneurysms were analysed. Before treatment, inflow jets were found in 13 (72.2%) aneurysms in the incompletely occluded group and 34 (44.7%) in the completely occluded group (P = 0.292). After deployment of the pipeline embolisation device, inflow jets remained in nine (50%) aneurysms in the incompletely occluded group and nine (11.8%) in the completely occluded group (P = 0.001). In the incompletely occluded group, regions with inflow jets after treatment corresponded with the patent areas shown on follow-up digital subtraction angiography. The mean reduction ratios of velocity in the whole aneurysm and on the neck plane were lower in the incompletely occluded than in the completely occluded group (P = 0.003; P = 0.017). Multivariate analysis revealed that the only independent risk factors for incomplete aneurysm occlusion were the reduction ratios of velocity (in the whole aneurysm, threshold 0.362, P = 0.005; on the neck plane, threshold 0.273, P = 0.015).

CONCLUSIONS

After pipeline embolisation device placement, reduction ratios of velocity in the whole aneurysm of less than 0.362 and on the neck plane of less than 0.273 are significantly associated with a greater risk of aneurysm incomplete occlusion. In addition, the persistence of inflow jets in aneurysms is associated with incomplete occlusion in the inflow jet area.

Ostium Ratio and Neck Ratio Could Predict the Outcome of Sidewall Intracranial Aneurysms Treated with Flow Diverters

BACKGROUND AND PURPOSE

Incompletely occluded flow diverter treated aneurysms remain at risk of rupture and thromboembolic complications. Our aim was to identify the potential for incomplete occlusion of intracranial aneurysms treated by flow diverters. We investigated whether aneurysm ostium size in relation to parent artery size affects angiographic outcomes of flow diverter-treated sidewall aneurysms.

MATERIALS AND METHODS

Flow diverter-treated sidewall aneurysms were divided into "occluded" and "residual" (incomplete occlusion) groups based on 6-month angiographic follow-up. We calculated the ostium ratio, a new parameter defined as the aneurysm ostium surface area versus the circumferential surface area of the parent artery. We also calculated the neck ratio, defined as clinical aneurysm neck diameter versus parent artery diameter from pretreatment 2D DSA, as a 2D surrogate. We compared the performance of these ratios with existing aneurysm morphometrics (size, neck diameter, volume, aspect ratio, size ratio, undulation index, nonsphericity index, ellipticity index, bottleneck factor, aneurysm angle, and parent vessel angle) and flow diverter-related parameters (metal coverage rate and pore density). Statistical tests and receiver operating characteristic analyses were performed to identify significantly different parameters between the 2 groups and test their predictive performances.

RESULTS

We included 63 flow diverter-treated aneurysms, 46 occluded and 17 residual. The ostium ratio and neck ratio were significantly higher in the residual group than in the occluded group (P < .001 and P = .02, respectively), whereas all other parameters showed no statistical difference. As discriminating parameters for occlusion, ostium ratio and neck ratio achieved areas under the curve of 0.912 (95% CI, 0.838-0.985) and 0.707 (95% CI, 0.558-0.856), respectively.

CONCLUSIONS

High ostium ratios and neck ratios could predict incomplete occlusion of flow diverter-treated sidewall aneurysms. Neck ratio can be easily calculated by interventionists to predict flow-diverter treatment outcomes.

External validation of cerebral aneurysm rupture probability model with data from two patient cohorts

BACKGROUND

For a treatment decision of unruptured cerebral aneurysms, physicians and patients need to weigh the risk of treatment against the risk of hemorrhagic stroke caused by aneurysm rupture. The aim of this study was to externally evaluate a recently developed statistical aneurysm rupture probability model, which could potentially support such treatment decisions.

METHODS

Segmented image data and patient information obtained from two patient cohorts including 203 patients with 249 aneurysms were used for patient-specific computational fluid dynamics simulations and subsequent evaluation of the statistical model in terms of accuracy, discrimination, and goodness of fit. The model's performance was further compared to a similarity-based approach for rupture assessment by identifying aneurysms in the training cohort that were similar in terms of hemodynamics and shape compared to a given aneurysm from the external cohorts.

RESULTS

When applied to the external data, the model achieved a good discrimination and goodness of fit (area under the receiver operating characteristic curve AUC = 0.82), which was only slightly reduced compared to the optimism-corrected AUC in the training population (AUC = 0.84). The accuracy metrics indicated a small decrease in accuracy compared to the training data (misclassification error of 0.24 vs. 0.21). The model's prediction accuracy was improved when combined with the similarity approach (misclassification error of 0.14).

CONCLUSIONS

The model's performance measures indicated a good generalizability for data acquired at different clinical institutions. Combining the model-based and similarity-based approach could further improve the assessment and interpretation of new cases, demonstrating its potential use for clinical risk assessment.

Development and internal validation of an aneurysm rupture probability model based on patient characteristics and aneurysm location, morphology, and hemodynamics

PURPOSE

Unruptured cerebral aneurysms pose a dilemma for physicians who need to weigh the risk of a devastating subarachnoid hemorrhage against the risk of surgery or endovascular treatment and their complications when deciding on a treatment strategy. A prediction model could potentially support such treatment decisions. The aim of this study was to develop and internally validate a model for aneurysm rupture based on hemodynamic and geometric parameters, aneurysm location, and patient gender and age.

METHODS

Cross-sectional data from 1061 patients were used for image-based computational fluid dynamics and shape characterization of 1631 aneurysms for training an aneurysm rupture probability model using logistic group Lasso regression. The model's discrimination and calibration were internally validated based on the area under the curve (AUC) of the receiver operating characteristic and calibration plots.

RESULTS

The final model retained 11 hemodynamic and 12 morphological variables, aneurysm location, as well as patient age and gender. An adverse hemodynamic environment characterized by a higher maximum oscillatory shear index, higher kinetic energy and smaller low shear area as well as a more complex aneurysm shape, male gender and younger age were associated with an increased rupture risk. The corresponding AUC of the model was 0.86 (95% CI [0.85, 0.86], after correction for optimism 0.84).

CONCLUSION

The model combining variables from various domains was able to discriminate between ruptured and unruptured aneurysms with an AUC of 86%. Internal validation indicated potential for the application of this model in clinical practice after evaluation with longitudinal data.

Morphological parameters and anatomical locations associated with rupture status of small intracranial aneurysms

Characterization of the rupture risk factors for small intracranial aneurysms (SIAs, ≤5 mm) is clinically valuable. The present study aims to identify image-based morphological parameters and anatomical locations associated with the rupture status of SIAs. Two hundred and sixty-three patients with single SIAs (128 ruptured, 135 unruptured) were included, and six morphological parameters, including size, aspect ratio (AR), size ratio (SR), height–width ratio (H/W), flow angle (FA) and aneurysm width–parent artery diameter ratio, and the aneurysm locations were evaluated using three-dimensional geometry, and were used to identify a correlation with aneurysm rupture. Statistically significant differences were observed between ruptured and unruptured groups for AR, SR, H/W, FA, and aneurysm locations, from univariate analyses. Logistic regression analysis further revealed that AR (p = 0.034), SR (p = 0.004), H/W (p = 0.003), and FA (p < 0.001) had the strongest independent correlation with ruptured SIAs after adjustment for age, gender and other clinical risk factors. A future study on a larger SIA cohort need to establish to what extent the AR, SR, H/W and FA increase the risk of rupture in patients with unruptured SIAs in terms of absolute risks.