Hybrid Machine Learning Framework for Early Prediction of Cardiovascular Disease Using Logistic Regression and Artificial Neural Networks

Abstract

Cardiovascular diseases are one of the major reasons for mortality worldwide, and hence there is a need to develop efficient diagnostic tools. With the advancement in artificial intelligence and machine learning, efficient intelligent healthcare systems can be designed to support the decision-making process in the medical community. This study proposes a machine learning model using a combination of logistic regression and neural networks for the early prediction of cardiovascular disease. The proposed model makes use of the clinical data available, including the attributes of the patients. The data preprocessing steps such as handling missing values, feature normalization, and feature selection are performed on the data to increase the accuracy of the model. The model makes use of logistic regression as a baseline model and neural networks to efficiently handle the complex relationships in the data. The performance metrics used to evaluate the model include accuracy, precision, recall, F1-score, and area under the ROC curve. This is further supported by the experimental results, which show that the performance of the neural network model is better than the traditional models, with an accuracy of about 86% and AUC-ROC of 0.90. The importance of the features also indicates the clinical factors such as age, cholesterol level, blood pressure, and smoking status as the major factors affecting the risk of cardiovascular disease. The proposed framework proves the effectiveness of machine learning methods for the early detection of the disease and the diagnostic decisions of the clinicians. The research also proves the effectiveness of AI-based predictive models for the prevention of cardiovascular diseases.