Artificial Intelligence and Machine Learning Algorithms in Modern Cardiology

Abstract

BACKGROUND: Recent years have witnessed the widespread adoption of machine learning (ML) and deep learning techniques in various health-care applications. Artificial intelligence and ML algorithms using big medical data make it possible to predict diseases and enable the development of personalized treatments for patients. Heart diseases are one of the most common chronic diseases affecting human health, and early detection can reduce the mortality rate. AIM: We aimed to review different types of ML techniques and their applications in heart disease risk detection. METHODS: For different cardiovascular diseases, the choice of algorithms should be tailored based on their accuracy and efficiency RESULTS: The research presented highlights the critical global issue of heart disease and its impact on public health. The urgency to address this global problem is emphasized, as heart disease has become a significant factor in the increasing mortality rate worldwide. The introduction of ML in the prognosis of heart disease is a significant step toward realizing predictive, preventive, and personalized health care and reducing health-care costs. In this study, a comparative evaluation of ML models was made: Logistic regression, decision tree, random forest, and support vector machine. The quality of the data, as well as the choice of an appropriate algorithm, is key factors in the assessment of heart diseases. CONCLUSION: Despite the impressive performance of ML, there are doubts about its robustness in traditional health-care systems due to many security and privacy issues.