Ensemble of Convolutional Neural Networks for StrokePrediction: Towards Improved Diagnostic Accuracy

The paper presents an intelligent system that predicts stroke risk using eleven clinical features and evaluates seven supervised machine learning algorithms. Ensemble methods—Random Forest, Stacking Classifier, and Bagging Classifier—achieved the highest accuracies, reaching 99.52%, while other models such as KNN, TabNet, and a custom feedforward network also performed well. The study demonstrates that ensemble approaches are particularly effective for stroke classification tasks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.