The LASSO machine learning model predicted stroke risk in patients with blunt cerebrovascular injury with an area under the curve of 0.79, achieving 66% sensitivity and 74% specificity.
Cohort (n=1,197)
No
Can a machine-learning-based tool predict stroke risk in patients with blunt cerebrovascular injury?
A machine-learning model was developed to predict stroke risk in patients with blunt cerebrovascular injury, which may help optimize antithrombotic therapy.
Effect estimate: AUC 0.79 (95% CI 0.57-0.95)
We developed and evaluated a preliminary predictive model for personalized stroke risk assessment in patients with BCVI using key risk factors. The integration of patient-specific risk-benefit assessments into clinical decision-making could optimize and reduce variability in AT therapy. External validation is warranted to prepare this tool for broad clinical applicability.
Wagner et al. (Fri,) conducted a cohort in Blunt cerebrovascular injury (BCVI) (n=1,197). LASSO machine learning model vs. Other machine learning models was evaluated on Stroke prediction (ROC AUC) (AUC 0.79, 95% CI 0.57-0.95). The LASSO machine learning model predicted stroke risk in patients with blunt cerebrovascular injury with an area under the curve of 0.79, achieving 66% sensitivity and 74% specificity.