Evolutionary algorithms and decision trees for predicting poor outcome after endovascular treatment for acute ischemic stroke

Despite the large overall beneficial effects of endovascular treatment in patients with acute ischemic stroke, severe disability or death still occurs almost one-third patients. These patients, who might not benefit from treatment, have been previously identified traditional logistic regression models, which may oversimplify relations between characteristics and outcome, machine learning techniques, be difficult to interpret. We developed evaluated a novel evolutionary algorithm for fuzzy decision trees accurately identify poor outcome after was defined as having modified Rankin Scale score (mRS) higher equal 5. The created being comprehensible, easily interpretable making its predictions easy explain practitioners. Insights reason predicted can encourage acceptance adaptation practice help manage expectations treatment. compared our proposed method CART, benchmark tree algorithm, on classification accuracy interpretability. significantly outperformed CART: using 5-fold cross-validation average 1090 training set 273 test set, misclassified 77 (standard deviation 7) 83 (±7) CART. mean number nodes (decision leaf nodes) 11 (±2) 26 (±1) CART trees. With an 72% much fewer than used gain insights into predictive value patient contribute development more accurate medical prediction methods improved clarity practitioners