Search

This is an Accepted Manuscript of an article published by Taylor & Francis in Computer Methods in Biomechanics and Biomedical Engineering : Imaging & Visualization on 17.03.2020, available online: https://www.tandfonline.com/doi/full/10.1080/21681163.2020.1728579, The morphological assessment of anatomical structures is clinically relevant, but often falls short of quantitative or standardised criteria. Whilst human observers are able to assess morphological characteristics qualitatively, the development of robust shape features remains challenging. In this study, we employ psychometric and radiomic methods to develop quantitative models of the perceived irregularity of intracranial aneurysms (IAs). First, we collect morphological characteristics (e.g. irregularity, asymmetry) in imaging-derived data and aggregated the data using rank-based analysis. Second, we compute regression models relating quantitative shape features to the aggregated qualitative ratings (ordinal or binary). We apply our method for quantifying perceived shape irregularity to a dataset of 134 IAs using a pool of 179 different shape indices. Ratings given by 39 participants show good agreement with the aggregated ratings (Spearman rank correlation ρSp=0.84). The best-performing regression model based on quantitative shape features predicts the perceived irregularity with R2:0.84±0.05.

Erworben im Rahmen der Schweizer Nationallizenzen (http://www.nationallizenzen.ch), Background: Morphological irregularity is linked to intracranial aneurysm wall instability and manifests in the lumen shape. Yet there is currently no consent on how to assess shape irregularity. The aims of this work are to quantify irregularity as perceived by clinicians, to break down irregularity into morphological attributes, and to relate these to clinically relevant factors such as rupture status, aneurysm location, and patient age or sex. Methods: Thirteen clinicians and 26 laypersons assessed 134 aneurysm lumen segmentations in terms of overall perceived irregularity and five different morphological attributes (presence/absence of a rough surface, blebs, lobules, asymmetry, complex geometry of the parent vasculature). We examined rater agreement and compared the ratings with clinical factors by means of regression analysis or binary classification. Results: Using rank-based aggregation, the irregularity ratings of clinicians and laypersons did not differ statistically. Perceived irregularity showed good agreement with curvature (coefficient of determination R2 = 0.68 ± 0.08) and was modeled very accurately using the five morphological rating attributes plus shape elongation (R2 = 0.95 ± 0.02). In agreement with previous studies, irregularity was associated with aneurysm rupture status (AUC = 0.81 ± 0.08); adding aneurysm location as an explanatory variable increased the AUC to 0.87 ± 0.09. Besides irregularity, perceived asymmetry, presence of blebs or lobules, aneurysm size, non-sphericity, and curvature were linked to rupture. No association was found between morphology and any of patient sex, age, and history of smoking or hypertension. Aneurysm size was linked to morphology. Conclusions: Irregular lumen shape carries significant information on the aneurysm’s disease status. Irregularity constitutes a continuous parameter that shows a strong association with the rupture status. To improve the objectivity of morphological assessment, we suggest examining shape through six different morphological attributes, which can characterize irregularity accurately.

Erworben im Rahmen der Schweizer Nationallizenzen (http://www.nationallizenzen.ch) Purpose: Incidental aneurysms pose a challenge to physicians who need to decide whether or not to treat them. A statistical model could potentially support such treatment decisions. The aim of this study was to compare a previously developed aneurysm rupture logistic regression probability model (LRM) to other machine learning (ML) classifiers for discrimination of aneurysm rupture status. Methods: Hemodynamic, morphological, and patient-related information of 1631 cerebral aneurysms characterized by computational fluid dynamics simulations were used to train support vector machines (SVMs) with linear and RBF kernel (RBF-SVM), k-nearest neighbors (kNN), decision tree, random forest, and multilayer perceptron (MLP) neural network classifiers for predicting the aneurysm rupture status. The classifiers’ accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were evaluated and compared to the LRM using 249 test cases obtained from two external cohorts. Additionally, important variables were determined based on the random forest and weights of the linear SVM. Results: The AUCs of the MLP, LRM, linear SVM, RBF-SVM, kNN, decision tree, and random forest were 0.83, 0.82, 0.80, 0.81, 0.76, 0.70, and 0.79, respectively. The accuracy ranged between 0.76 (decision tree,) and 0.79 (linear SVM, RBF-SVM, and MLP). Important variables for predicting the aneurysm rupture status included aneurysm location, the mean surface curvature, and maximum flow velocity. Conclusion: The performance of the LRM was overall comparable to that of the other ML classifiers, confirming its potential for aneurysm rupture assessment. To further improve the predictions, additional information, e.g., related to the aneurysm wall, might be needed.

Erworben im Rahmen der Schweizer Nationallizenzen (http://www.nationallizenzen.ch) Background: Hemodynamic patterns have been associated with cerebral aneurysm instability. For patient-specific computational fluid dynamics (CFD) simulations, the inflow rates of a patient are typically not known. The aim of this study was to analyze the influence of inter- and intra-patient variations of cerebral blood flow on the computed hemodynamics through CFD simulations and to incorporate these variations into statistical models for aneurysm rupture prediction. Methods: Image data of 1820 aneurysms were used for patient-specific steady CFD simulations with nine different inflow rates per case, capturing inter- and intra-patient flow variations. Based on the computed flow fields, 17 hemodynamic parameters were calculated and compared for the different flow conditions. Next, statistical models for aneurysm rupture were trained in 1571 of the aneurysms including hemodynamic parameters capturing the flow variations either by defining hemodynamic “response variables” (model A) or repeatedly randomly selecting flow conditions by patients (model B) as well as morphological and patient-specific variables. Both models were evaluated in the remaining 249 cases. Results: All hemodynamic parameters were significantly different for the varying flow conditions (p

OBJECTIVEIncidental 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.METHODSPatient 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.RESULTSThe 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.CONCLUSIONSDeveloping 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.

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.

Background and Purpose— The aim of this study is to assess whether the PHASES score allows to (1) match decisions taken by multidisciplinary team whether to observe or intervene, (2) classify patients being diagnosed with a ruptured versus unruptured intracranial aneurysm (UIA), and (3) discriminate patients at low risk of rupture from the population of patients diagnosed with intracranial aneurysm. Methods— Population-based prospective and consecutive data were collected between 2006 and 2014. Patients (n=841) were stratified into 4 groups: stable UIA; growing observed UIA; immediately treated UIA; and aneurysmal subarachnoid hemorrhage (aSAH). All patients initially observed were pooled in a follow-up UIA group; patients from growing observed UIA, immediately treated UIA, and aSAH were pooled in a high risk of rupture group. Results are expressed as median [quartile 1, quartile 3]. Results— PHASES scores of immediately treated UIA patients were significantly higher than follow-up UIA group (5 [3, 7] versus 2 [1, 4]). Patients diagnosed with UIA and PHASES score of >3 were more likely to be treated, and the score ≤3 was predictive for observation (areas under these curves=0.74). Odds of being diagnosed with an aSAH were associated with PHASES score of >3 (UIA, 4 [2, 6]; aSAH, 5 [4, 8]; areas under these curves=0.66). Scores of stable UIA patients were significantly lower than high risk of rupture group (2 [1, 4] versus 5 [4, 7]; stable UIA outcome prediction by PHASES score of ≤3: areas under these curves=0.76). Conclusions— There is a progression of PHASES score between stable UIA, growing observed UIA, immediately treated UIA, and aSAH groups. PHASES score of ≤3 is associated with a low but not negligible likelihood of aneurysm rupture, and specificity of the classifier is low.

The disease resulting in the formation, growth, and rupture of intracranial aneurysms is complex. Research is accumulating evidence that the disease is driven by many different factors, some constant and others variable over time. Combinations of factors may induce specific biophysical reactions at different stages of the disease. A better understanding of the biophysical mechanisms responsible for the disease initiation and progression is essential to predict the natural history of the disease. More accurate predictions are mandatory to adequately balance risks between observation and intervention at the individual level as expected in the age of personalized medicine. Multidisciplinary exploration of the disease also opens an avenue to the discovery of possible preventive actions or medical treatments. Modern information technologies and data processing methods offer tools to address such complex challenges requiring 1) the collection of a high volume of information provided globally, 2) integration and harmonization of the information, and 3) management of data sharing with a broad spectrum of stakeholders.Over the last decade an infrastructure has been set up and is now made available to the academic community to support and promote exploration of intracranial disease, modeling, and clinical management simulation and monitoring.The background and purpose of the infrastructure is reviewed. The infrastructure data flow architecture is presented. The basic concepts of disease modeling that oriented the design of the core information model are explained. Disease phases, milestones, cases stratification group in each phase, key relevant factors, and outcomes are defined. Data processing and disease model visualization tools are presented. Most relevant contributions to the literature resulting from the exploitation of the infrastructure are reviewed, and future perspectives are discussed.

Transcatheter aortic valve implantation (TAVI) is a minimally invasive procedure to treat severe aortic stenosis in patients with a high risk for conventional surgery. In-silico experiments of stent deployment within patient-specific models of the aortic root have created an opportunity to predict stent behavior during the intervention. Current limitations in procedure planning are a primary motivator for these simulations. The virtual stent placement preceding the deployment phase of such experiments has major influence on the outcome of the simulation, but only received little attention in literature up to now. This work presents a methodical approach to patient-specific planning of placement of self-expanding stent models by analyzing experimental outcomes of different sets of boundary conditions constraining the stent. As a results, different paradigms for automated or expert guided stent placement are evaluated, which demonstrate the benefits of virtual stent deployment for intervention planning. To build a predictive planning pipeline for TAVI we use an automatic segmentation of the aorta, aortic root and left ventricle, which is converted to a finite element mesh. The virtual stent is then placed along a guide wire model and deployed at multiple locations around the aortic root. The simulation has been evaluated using pre- and post-interventional CT scans with an average relative circumferential error of 4.0% (±2.5%), which is less than half of the average difference in circumference between individual stent sizes (8.6%). Our methods are therefore enabling patient-specific planning and provide better guidance during the intervention.

International audience; We propose a machine learning-based framework using oblique random forests for 3-D vessel segmentation. Two different kinds of features are compared. One is based on orthogonal subspace filtering where we learn 3-D eigenspace filters from local image patches that return task optimal feature responses. The other uses a specific set of steerable filters that show, qualitatively,similarities to the learned eigenspace filters, but also allow for explicit parametrization of scale and orientation that we formally generalize to the 3-D spatial context. In this way, steerable filters allow to efficiently compute oriented features along arbitrary directions in 3-D. The segmentation performance is evaluated on four 3-D imaging datasets of the murine visual cortex at a spatial resolution of 0.7μm. Our experiments show that the learning-based approach is able to significantly improve the segmentation compared to conventional Hessian-based methods. Features computed based on steerable filters prove to be superior to eigenfilter-based features for the considered datasets. We further demonstrate that random forests using oblique split directions outperform decision tree ensembles with univariate orthogonal splits

The cerebrovascular system continuously delivers oxygen and energy substrates to the brain, which is one of the organs with the highest basal energy requirement in mammals. Discontinuities in the delivery lead to fatal consequences for the brain tissue. A detailed understanding of the structure of the cerebrovascular system is important for a multitude of (patho-)physiological cerebral processes and many noninvasive functional imaging methods rely on a signal that originates from the vasculature. Furthermore, neurodegenerative diseases often involve the cerebrovascular system and could contribute to neuronal loss. In this review, we focus on the cortical vascular system. In the first part, we present the current knowledge of the vascular anatomy. This is followed by a theory of topology and its application to vascular biology. We then discuss possible interactions between cerebral blood flow and vascular topology, before summarizing the existing body of the literature on quantitative cerebrovascular topology.