Predicting Heart Disease Algorithm Using DNN and MNN in Deep Learning

Abstract

Heart diseases remain a significant global health concern, necessitating advanced predictive models for early detection and intervention.  Based on an extensive collection of patient data, this study suggests a method for predicting cardiac disease using deep learning algorithms. The architecture employs a multi-layered neural network and deep neural network to capture intricate patterns within the data, optimizing predictive accuracy. The data set downloaded from Kaggle with 14 attributes is used in this study. The deep learning model multilayer neural network and deep neural network are chosen for this study. These models undergo rigorous evaluation using standard performance metrics, demonstrating their efficacy in discriminating between individuals with and without heart disease. Results indicate that the deep learning algorithm exhibits promising predictive capabilities, outperforming traditional methods. Integrating interpretable elements contributes to the model's clinical utility, facilitating better-informed decision-making for healthcare professionals.