Poster L13 - Considering Size, Shape, and Hemodynamics: A Synergistic Approach in Risk of Rupture Assessment of Cerebral Aneurysms

Introduction: Amidst the complexity of cerebral anomalies, assessing the status of Intracranial Aneurysms (IAs) proves challenging for clinicians. Conventional metrics such as size and location alone lack precision in predicting severity. Our objective is to elucidate the intricate interplay between 3D Shape, Size, Hemodynamics, and Rupture risk. We aim to provide a reliable analytic tool to redefine clinical decision-making and enhance patient outcomes by addressing this cerebrovascular threat. This study emphasizes to elucidate the importance of assessing each feature set and its collective impact on risk assessment, constituting an essential basis for informed, patient-centric decisions.

Materials and

Methods: We investigated 735 aneurysms from 596 patients (474 unruptured, 261 ruptured), capturing 20 morphological and 6 hemodynamic features across 10 anatomical locations. Hemodynamics have been computed using Ansys Fluent 2023. Wall Shear Stress (WSS)-derived indices Oscillatory Shear Index (OSI), Relative Resident Time (RRT), and Time-Averaged Wall Shear Stress (TAWSS) have been computed for analysis. We identified significant differences between ruptured and unruptured groups by employing the Jarque-Bera test for normality check and the Wilcoxon rank-sum test (data is not normally distributed). Utilizing Receiver Operating Characteristic (ROC) analysis, we determined effective thresholds and Area Under the ROC Curve (AUC) values, demonstrating the discriminative power of morphometric models in classifying rupture status.

Results, Conclusions, and Discussions: Features like Non-Sphericity Index (NSI) (AUC= 0.75), Writhe metrics like ¯W_mean^(L_1 ) (AUC=0.75), Ellipticity Index (EI) (AUC=0.73), WSS (AUC=0.79), OSI (AUC=0.80) demonstrated a fair association with the rupture status. Incorporating shape features revealed heightened rupture risk among smaller aneurysms, initially classified as benign based solely on size features. Location (AUC=0.85) emerged as a significant predictor, individually and in combination with other features. Additionally, hemodynamic factors such as Average WSS, Maximum intra-aneurysmal WSS, Low WSS Area, Average OSI, and RRT achieved statistical significance (p < 0.02). Multivariate analysis highlighted the independent significance of EI, NSI, WSS, and OSI in predictive models, underscoring their role in aneurysm prognosis.

Acknowledgements (Optional):