Abstract

The detection and treatment of heart disorders caused by the impact of COVID-19 are improved by intensive work. Mining and recording the data in the medical field provides potential development in maintaining patients’ details. This is possible with the convergence of improved technology and medical diagnostics models. It is essential to analyze the interconnection of threat factors in the patient's clinical history to achieve the respective heart disease diagnosis. A meticulous analysis of multiple mechanisms in patient data is used to predict heart disease before COVID-19 infection. The main required attributes for detecting cardiac disorders due to COVID-19 are acquired by applying the feature selection model. The critical patient history details such as age, smoking habits, physical activity, stress levels, gender, previous chest pain occurrences, diabetes, electrocardiogram (ECG) readings, dietary patterns, chest pain type, and troponin levels are considered for predicting heart disease. Different AI technologies, such as deep neural networks with SVM (d-SVM), were used, and the results were compared between two datasets from the heart disease database. These methodologies are employed to select features from the database and are also employed on all features of the data repository. The enhanced accuracy rate of 95% is acquired through our proposed model, which uses selected features as input. Early heart disease prediction is achieved through our proposed technique's assisting structure. Successful deployment of an AI model as computationally intensive as this, from the laboratory setting to real-time actual clinical practice, however, relies on an architecture for scalable and fault-tolerant deployment. To this end, we introduce a Cloud Computing and Service Deployment paradigm that is specifically tailored for executing the d-SVM model. It leverages the cloud's elasticity as well as high availability to deal with fluctuating diagnosis loads in healthcare networks. By converting the d-SVM into an Internet-based Service through an Application Programming Interface (API), this work maximizes the potential of the model for large-scale, sustainable, and cost-saving clinical utility.