Meta-Heuristic-enabled Adaptive Deep Convolutional Neural Network for Heart Disease Prediction

Heart disease has posed a severe threat to the well-being of individuals in recent years due to its widespread incidence and elevated mortality rate. As a result, early detection of heart disease using basic physical signs derived from routine physical examinations has grown into an important topic. Yet, the vast volume of information makes evaluation and forecasting process more difficult. Thus, an automated heart disease prediction framework is presented in this paper with the help of meta-heuristic-enabled deep learning. Firstly, the requisite data are collected from the available resources and the attained data are processed into the feature extraction phase. The t-Distributed Stochastic Neighbor Embedding (t-SNE) method is employed for extracting the features from the obtained data and this model supports to reduce the dimensionality of the data. Subsequently, the resultant features are forwarded to the heart disease prediction phase, where the 1-Dimensional Adaptive Convolutional Neural Network (1D-ACNN) is introduced. Moreover, the parameters are optimized in this network by the Improved Dung Beetle Optimizer Algorithm (IDBO). Thus, the heart disease-predicted data are obtained in the end with the support of the 1D-ACNN network, and the implemented model is contrasted with existing algorithms and prediction techniques to ensure the system efficacy.