Machine Learning Models for Brain Arteriovenous Malformations Presenting with Hemorrhage Based on Clinical and Angioarchitectural Characteristics

•We validated the application of R2eD AVM score to predict the likelihood of rupture presentation of brain arteriovenous malformations in China.•By comparing five supervised machine learning algorithms, multivariable logistic regression, and R2eDAVM scoring system, we found that the multilayer perceptron model showed the best performance.•We conducted a complementary systematic review and meta-analysis of previous studies to explore the potential predictors for bAVMs rupture.