Stroke prediction using 1DCNN with ANOVA

Abstract

Stroke and heart disease are among the most common outcomes of hypertension. Each year, heart disease, stroke, and other cardiovascular disorders claim the lives of more than 877,500 people in the United States, making them the first and fifth leading causes of death, so being able to predict them early helps save lives. A lot of research has been done to reach this goal. Machine learning models are mostly used for this purpose. For the first time in this study, we have used the Deep Learning (DL) model, i.e., one dimensional convolutional neural network (1D CNN) . In this study, first we extracted important features using the Analysis of variance (ANOVA) method. Then the data set with the new features that came up was given to the model. Then we compare all machine learning algorithms—K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Logistic Regression (LR), Random Forest Classifier (RF), Gradient Boosting Clas-sifier (XGB), and LoLight gradient boosting machine classifier (LGBM)—with 1DCNN. Recall, the F1 score, accuracy, and precision are some of the confusion metrics used to assess the effectiveness of the results.The results show that when used on reprocessed data, the proposed model performs best and is more than 98% accurate.