A Hybrid Model for the Detection and Classification of Cardiovascular Diseases Based on Deep Learning and Optimization Techniques

The leading global cause of death is cardiovascular disease (CVD), also referred to as heart disease. A recent World Heart Federation study shows that cardiovascular disease is responsible for one in every three deaths. Stress, alcohol, smoking, poor diet, lack of exercise, and other linked health issues including diabetes or high blood pressure can all contribute to cardiovascular disease. On the other hand, most cardiovascular disease. There is known to be a complete cure for linked disorders when detected early. Emerging novel data analysis strategies may allow for early detection of cardiovascular disease by examining a patient's medical file. Cardiovascular disease is currently predicted using clinical datasets and machine learning (ML) algorithms. However, because of the disparity in classes and high dimension, clinical datasets pose significant challenges. To address such issues, a dynamic model is proposed. In this paper, an effective decision support (or assistive) system is proposed by combining a deep learning classifier and an optimization technique to detect and classify cardiovascular diseases. It is expected that combining the two strategies will increase the effectiveness of current approaches for using clinical records to anticipate cardiovascular disease. According to clinical data and the severity level of the patient, a deep learning system can anticipate cardiovascular disease in its initial stages, lowering rates of death. The optimization method determines the best parameters for a sufficient quantity of synthesised samples to achieve the best prediction with an optimised classifier. In addition, various parameters such as accuracy, PSNR, sensitivity, specificity, and others were computed and compared with existing systems in this work.