A Comparative Study of Cardiac Disease Classification Methods, Systematic Literature Reviews, Open Challenges and Future Directions

In recent years, Cardiovascular Diseases (CVDs) have appeared as one of the primary causes of death worldwide. The human heart is essential for circulating blood and ensuring its distribution throughout the parts of the body. The symptoms such as fatigue, difficulty in breathing, swelling in the ankles, fluid retention, and other related issues are the causes of CVD. Diagnosing CVDs poses a challenge for healthcare professionals due to the extensive amount of data in each scan and the potential for low contrast. Deep learning and machine learning models are utilized in various medical fields, including CVD diagnosis, achieving promising outcomes. This review examines the latest techniques and progress in the classification of cardiac diseases through machine learning and deep learning methods. In this review, different classification approaches based on the types of data utilized, Magnetic Resonance Imaging (MRI) are categorized. This survey explores the software platforms and tools frequently used for model implementation and their effectiveness in clinical applications. Additionally, the performance of various classification models using relevant metrics such as accuracy, sensitivity, and specificity are reviewed in this work. Finally, the challenges encountered in existing cardiac disease classification models are analyzed and potential areas for improvement are discussed to enhance the accuracy and reliability of the categorization systems in future clinical settings.