A survey on open challenges in heart disease prediction models.
Journal:
Computational biology and chemistry
Published Date:
Mar 8, 2025
Abstract
Heart disease (HD), is a deadly serious disease, that has received a great deal of consideration in medical study. A lot of prompting factors like professional, as well as personal behaviors and genetic nature, are accountable for the occurrence of HD. Accurate, early, and efficient medical diagnosis of HD is essential for taking precautionary measures to prevent deaths. The analysis of HD is a demanding task that could offer automatic prediction regarding the condition of the patient's heart so that the required diagnosis could be made effectually. This study analyses 68 papers that discuss HD prediction critically. Data were gathered for this survey study from previously published studies. As a result, thorough analyses of the strategies employed are done and succinctly described. Additionally, a thorough analysis of the highest performance levels of HD prediction models and a variety of characteristics used in each work is conducted. Lastly, the improved HD prediction is expanded with the identification of various concerns, which may guide the analysts in enhancing future studies. To assess the accuracy of HD methods, the main outcomes of this study include improved performance metrics, stronger methodological support and adherence, as well as enhanced features and traceability in the work.
