An Efficient Application of Hybrid Optimization with Deep Learning Approach in the Prediction of Cardiovascular Disease

The incidence of arteries and cardiovascular diseases, which result in heart failure and early deaths in the form of myocardial infarction, stroke, and fainting, is currently one of the primary causes of mortality in every country in the globe. Deep Learning (DL) has lately attracted research interest in a variety of fields, especially medical treatment, where it is being used to diagnose the cardiovascular disease early on via image processing. The methodology proposed in this work is revolutionary using a deep learning method for the premature heart disease identification and diagnosis. To reduce noise and improve image quality, the dataset must first be pre-processed using the Weiner-Frost filter. Then, Feature extraction is carried out using k-means clustering technique to retrieve the exact features for further process. The following phase involves selecting the most crucial features for predicting CVD using a hybrid lion-grey-wolf optimization algorithm, which incorporates the advantages of both techniques. A deep learning model which incorporates a convolutional neural network (CNN) and bidirectional long short-term memory (Bi-LSTM) architecture completes the categorization of CVD. The suggested method is highly accurate at identifying CVD and is a dependable technique for identifying and diagnosing CVD early.