Machine learning approach for hemorrhagic transformation prediction: Capturing predictors' interaction
ConclusionCerebral microbleeds, NIHSS, and infarction size were identified as HT predictors. The best predicting models were RFC and GBC capable of capturing nonlinear interaction between predictors. Predictor interaction suggests a dynamic, rather than, fixed cutoff risk value for any of these predictors.
Source: Frontiers in Neurology - Category: Neurology Source Type: research
More News: Brain | Hemorrhagic Stroke | Ischemic Stroke | Learning | Neurology | Statistics | Stroke | Study | Training | Universities & Medical Training