Predicting Stroke Disease Based on Recurrent Neural Network and Optimization techniques

Abstract

Stroke disease is one of the most prevalent diseases all over the world. This paper presents a powerful early stroke prediction system that uses medical records that describe whether a person is infected or not. We proposed an optimized DeepRNN based on different layers of A Recurrent Neural Network (RNN) and KerasTuner optimization technique for predication stroke disease. The proposed model is compared with other ML models: Decision Tree (DT), Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), K Nearest Neighbors (K-NN), and Naive Bayes (NB). The GridsearchCV technique optimized ML models. The results showed that DeepRNN was the highest performance model compared with ML models.