Development of Multi Model Algorithms for Cardiac Arrest Prediction
Main Article Content
Abstract
Cardiac arrest or myocardial infarctions pose a significant health risk globally, necessitating effective monitoring and early detection strategies. In this study, we propose a framework comprising two complementary methods for enhancing cardiovascular health management: a text-based prediction system and an image-based analysis approach. The text-based prediction system provides individualized risk estimates for cardiovascular events by analyzing user-input health factors using machine learning algorithms. With the use of a well- chosen dataset and thorough preprocessing, our models show strong predictive capabilities for heart disease. Convolutional neural networks (CNNs) are used in tandem with other image-based analytic techniques to identify anomalies and arrhythmias from electrocardiogram (ECG) information. Through data augmentation and rigorous model training, our approach achieves high accuracy in classifying various cardiac rhythms and arrhythmias. The framework presents a promising approaches for advancing cardiovascular health management, leveraging both text and image-based techniques. Through rigorous evaluation, we demonstrate the efficacy and reliability of our proposed methodology in addressing the challenges of heart attack detection and risk assessment.