PT - JOURNAL ARTICLE AU - Seyedeh Fatemeh Salimi Ashkezari AU - Fernando Mut AU - Martin Slawski AU - Boyle Cheng AU - Alexander K Yu AU - Tim G White AU - Henry H Woo AU - Matthew J Koch AU - Sepideh Amin-Hanjani AU - Fady T Charbel AU - Behnam Rezai Jahromi AU - Mika Niemelä AU - Timo Koivisto AU - Juhana Frosen AU - Yasutaka Tobe AU - Spandan Maiti AU - Anne M Robertson AU - Juan R Cebral TI - Prediction of bleb formation in intracranial aneurysms using machine learning models based on aneurysm hemodynamics, geometry, location, and patient population AID - 10.1136/neurintsurg-2021-017976 DP - 2022 Oct 01 TA - Journal of NeuroInterventional Surgery PG - 1002--1007 VI - 14 IP - 10 4099 - http://jnis.bmj.com/content/14/10/1002.short 4100 - http://jnis.bmj.com/content/14/10/1002.full SO - J NeuroIntervent Surg2022 Oct 01; 14 AB - Background Bleb presence in intracranial aneurysms (IAs) is a known indication of instability and vulnerability.Objective To develop and evaluate predictive models of bleb development in IAs based on hemodynamics, geometry, anatomical location, and patient population.Methods Cross-sectional data (one time point) of 2395 IAs were used for training bleb formation models using machine learning (random forest, support vector machine, logistic regression, k-nearest neighbor, and bagging). Aneurysm hemodynamics and geometry were characterized using image-based computational fluid dynamics. A separate dataset with 266 aneurysms was used for model evaluation. Model performance was quantified by the area under the receiving operating characteristic curve (AUC), true positive rate (TPR), false positive rate (FPR), precision, and balanced accuracy.Results The final model retained 18 variables, including hemodynamic, geometrical, location, multiplicity, and morphology parameters, and patient population. Generally, strong and concentrated inflow jets, high speed, complex and unstable flow patterns, and concentrated, oscillatory, and heterogeneous wall shear stress patterns together with larger, more elongated, and more distorted shapes were associated with bleb formation. The best performance on the validation set was achieved by the random forest model (AUC=0.82, TPR=91%, FPR=36%, misclassification error=27%).Conclusions Based on the premise that aneurysm characteristics prior to bleb formation resemble those derived from vascular reconstructions with their blebs virtually removed, machine learning models can identify aneurysms prone to bleb development with good accuracy. Pending further validation with longitudinal data, these models may prove valuable for assessing the propensity of IAs to progress to vulnerable states and potentially rupturing.Data are available upon reasonable request. The data that support the findings of this study are available from the corresponding author, upon reasonable request.