AUTHOR=Wang Kai , Shi Qianqian , Sun Chao , Liu Wencai , Yau Vicky , Xu Chan , Liu Haiyan , Sun Chenyu , Yin Chengliang , Wei Xiu’e , Li Wenle , Rong Liangqun TITLE=A machine learning model for visualization and dynamic clinical prediction of stroke recurrence in acute ischemic stroke patients: A real-world retrospective study JOURNAL=Frontiers in Neuroscience VOLUME=Volume 17 - 2023 YEAR=2023 URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2023.1130831 DOI=10.3389/fnins.2023.1130831 ISSN=1662-453X ABSTRACT=Purpose: To establish and validate a prediction model predicting risk of stroke recurrence within one year for patients with acute ischemic stroke (AIS) based on a machine learning (ML) algorithm Methods: A total of 645 AIS patients at the Second Affiliated Hospital of Xuzhou Medical University were screened, included and followed up for 1 year for comprehensive clinical data. Univariate and multivariate logistic regression were used to screen the risk factors of stroke recurrence. The data set was randomly divided into training set and test set according to the ratio of 7:3, and the following six prediction models were established by machine algorithm: Random forest(RF), Naive Bayes model(NBC), Decision tree(DT), XGB(Extreme gradient boosting), Gradient boosting machine(GBM) and Logistic regression(LR). The best prediction model was selected by 10-fold cross validation and receiver operating characteristic (ROC) curve, and the variables were visualized by interpretable model SHAP. Finally, a web-based calculator was constructed. Results: Logistic regression analysis showed that right hemisphere, homocysteine(HCY), C-reaction protein(CRP), and stroke severity (SS) were independent risk factors for the development of stroke recurrence in AIS patients. In 10-fold cross-validation, AUC ranked from 0.777 to 0.959. In ROC curve analysis, AUC ranged from 0.887 to 0.946. RF model has the best ability to predict stroke recurrence, and HCY has the largest contribution to the model. A Web-based calculator (https://mlmedicine-re-stroke2-re-stroke2-baylee.streamlitapp.com/) has been developed accordingly. Conclusions: The current study presented a well-performing clinical prediction model based on machine learning on 1-year recurrence for stroke, especially the RF model.