This study assessed the accuracy of artificial intelligence (AI) models in predicting stroke prognosis. Seven studies were included in a meta-analysis. The pooled area under the curve (AUC) was 0.872 (95% CI 0.862–0.881), indicating good predictive accuracy. Subgroup analysis showed varying AUCs across different AI algorithms (SVM, RF, LR, DL, Xgboost), with SVM and Xgboost performing best. AI models offer a valuable assistive tool for physicians in predicting stroke outcomes.
Publisher
Frontiers in Neuroscience
Published On
Sep 07, 2023
Authors
Yujia Yang, Li Tang, Yiting Deng, Xuzi Li, Anling Luo, Zhao Zhang, Li He, Cairong Zhu, Muke Zhou, Ni Zhang, Xia Zhang, Qifu Li
Tags
artificial intelligence
stroke prognosis
predictive accuracy
meta-analysis
SVM
Xgboost
AI algorithms
Related Publications
Explore these studies to deepen your understanding of the subject.
