Rupture Risk Assessment for Cerebral Aneurysm Using Interpretable Machine Learning on Multidimensional Data.

Background: Assessment of cerebral aneurysm rupture risk is an important task, but it remains challenging. Recent works applying machine learning to rupture risk evaluation presented positive results. Yet they were based on limited aspects of data, and lack of interpretability may limit their use in clinical setting. We aimed to develop interpretable machine learning models on multidimensional data for aneurysm rupture risk assessment. Methods: Three hundred seventy-four aneurysms were included in the study. Demographic, medical history, lifestyle behaviors, lipid profile, and morphologies were collected for each patient. Prediction models were derived using machine learning methods (support vector machine, artificial neural network, and XGBoost) and conventional logistic regression. The derived models were compared with the PHASES score method. The Shapley Additive Explanations (SHAP) analysis was applied to improve the interpretability of the best machine learning model and reveal the reasoning behind the predictions made by the model. Results: The best machine learning model (XGBoost) achieved an area under the receiver operating characteristic curve of 0.882 [95% confidence interval (CI) = 0.838-0.927], significantly better than the logistic regression model (0.779; 95% CI = 0.729-0.829; P = 0.002) and the PHASES score method (0.758; 95% CI = 0.713-0.800; P = 0.001). Location, size ratio, and triglyceride level were the three most important features in predicting rupture. Two typical cases were analyzed to demonstrate the interpretability of the model. Conclusions: This study demonstrated the potential of using machine learning for aneurysm rupture risk assessment. Machine learning models performed better than conventional statistical model and the PHASES score method. The SHAP analysis can improve the interpretability of machine learning models and facilitate their use in a clinical setting.

Comparison of polyvinyl alcohol copolymer with detachable balloons for the embolisation of direct carotid cavernous fistula: a single-centre experience.

OBJECTIVES: To characterise the safety, efficacy and cost of direct carotid-cavernous fistula (CCF) treatment using polyvinyl alcohol copolymer or detachable balloons. METHODS: We reviewed retrospectively patients with direct CCFs treated with either a detachable balloon or polyvinyl alcohol copolymer at our hospital from 2005 to 2015 and identified 94 patients with 105 CCFs. All patients had follow-up angiograms. The CCF occlusion rate, procedure complication rate, treatment expense and operation time were recorded. RESULTS: With a mean of 5.4 months of angiographic follow-up, the complete occlusion rate and recanalisation rate of the polyvinyl alcohol copolymer group was not significantly different from that of the detachable balloon group. The treatment expense was much higher and the operation time was much longer in the polyvinyl alcohol copolymer group than the detachable balloon group (P < 0.001). CONCLUSIONS: Embolisation of CCF with polyvinyl alcohol copolymer is as safe and effective as detachable balloon but has a much higher cost and longer operation time. KEY POINTS: • Carotid-cavernous fistula results from a damaged carotid artery. • Detachable balloons have been used with success for many years. • Some reported excellent outcomes after embolisation with polyvinyl alcohol copolymer. • Treatment expense is much higher in the polyvinyl alcohol copolymer group.