Predictive Analytics in Cardiology: Using Machine Learning to Identify Heart Failure Early

Abstract

The creation of an algorithm using machine learning and natural language processing methods is the main goal of this study. Predicting cardiac disease is a tough problem in medical data analysis, given that it is now one of the leading causes of mortality globally. The varity of types such as health records, electronic health records, network monitoring (body), and patients diagnosing data conditions by projecting medical sensors and wearable technology onto the human body are used. Machine learning (ML) has demonstrated its value in aiding decision-making and predicting outcomes from the extensive datasets provided by the healthcare industry. The study presents the innovative MLP-EBMDA (Multi-Layer Perceptron for Enhanced Brownian Motion-based Dragonfly Algorithm) for heart disease prediction and uses an efficient unsupervised method for feature selection. The process begins with obtaining input from the dataset, followed by preprocessing and the proposed feature selection technique, which adeptly chooses relevant features. Subsequently, The MLP-EBMDA is a novel technique is used to classify cardiac disease, making early-stage heart disease prediction easier. The suggested method successfully classifies cardiac disease as normal or abnormal with an excellent impressive accuracy rate of 96.4%.