biostudies-literatureZeng ZMajor Projects of Institute of Chinese Materia Medica China Academy of Chinese Medical SciencesJiangxi Provincial Education Department ProjectHospital-level project of the Second Affiliated Hospital of Nanchang UniversityJiangxi Provincial Science and Technology Department Projecte0279706https://www.ebi.ac.uk/biostudies/studies/S-EPMC9794063biostudies-literatureUnknown17(12)<h4>Objective</h4>Ischemic stroke (IS) with subsequent cerebrocardiac syndrome (CCS) has a poor prognosis. We aimed to investigate electrocardiogram (ECG) changes after IS with artificial intelligence (AI).<h4>Methods</h4>We collected ECGs from a healthy population and patients with IS, and then analyzed participant demographics and ECG parameters to identify abnormal features in post-IS ECGs. Next, we trained the convolutional neural network (CNN), random forest (RF) and support vector machine (SVM) models to automatically detect the changes in the ECGs; Additionally, We compared the CNN scores of good prognosis (mRS ≤ 2) and poor prognosis (mRS > 2) to assess the prognostic value of CNN model. Finally, we used gradient class activation map (Grad-CAM) to localize the key abnormalities.<h4PloS oneAssessing electrocardiogram changes after ischemic stroke with artificial intelligence.PMC9794063No. 2021efyC02No. 20208031No. GJJ200133No. ZZ15-WT-04Xu LYan LChen QYi YZeng ZLou WDeng LCheng ZFan GWang QZhang YWang YYu YfalseAssessing electrocardiogram changes after ischemic stroke with artificial intelligence.<h4>Objective</h4>Ischemic stroke (IS) with subsequent cerebrocardiac syndrome (CCS) has a poor prognosis. We aimed to investigate electrocardiogram (ECG) changes after IS with artificial intelligence (AI).<h4>Methods</h4>We collected ECGs from a healthy population and patients with IS, and then analyzed participant demographics and ECG parameters to identify abnormal features in post-IS ECGs. Next, we trained the convolutional neural network (CNN), random forest (RF) and support vector machine (SVM) models to automatically detect the changes in the ECGs; Additionally, We compared the CNN scores of good prognosis (mRS ≤ 2) and poor prognosis (mRS > 2) to assess the prognostic value of CNN model. Finally, we used gradient class activation map (Grad-CAM) to localize the key abnormalities.<h42022-01-01T00:00:00Z20222025-04-04T12:09:44.643Z2025-04-04T12:09:44.643ZS-EPMC97940633657442710.1371/journal.pone.0279706