Related Experiment Video
Updated: May 21, 2025
In Silico Clinical Trials for Cardiovascular Disease
Published on: May 27, 2022
A survey on open challenges in heart disease prediction models
Chetan Vikram Andhare1, D R Ingle1
1Bharati Vidyapeeth College of Engineering, Navi Mumbai, Maharashtra 400614, India.
Insights
This study critically analyzes 68 papers on heart disease (HD) prediction. It identifies key strategies and performance levels to improve early and accurate diagnosis, aiming to reduce mortality.
Area of Science:
- Cardiology and Medical Informatics
- Computational Health
- Biomedical Data Science
Background:
- Heart disease (HD) is a significant cause of mortality, influenced by genetic and lifestyle factors.
- Accurate and early diagnosis of heart disease is crucial for effective intervention and prevention.
- Automated prediction models for heart disease diagnosis are essential for clinical decision-making.
Purpose of the Study:
- To critically analyze existing literature on heart disease prediction strategies.
- To identify high-performing heart disease prediction models and their utilized features.
- To highlight challenges and suggest improvements for future heart disease prediction research.
Main Methods:
- Systematic review and critical analysis of 68 published studies on heart disease prediction.
- Evaluation of employed strategies, performance levels, and characteristics within the selected papers.
- Synthesis of findings to identify areas for enhancement in heart disease prediction methodologies.
Main Results:
- Comprehensive analysis of diverse heart disease prediction strategies and their performance metrics.
- Identification of top-performing heart disease prediction models and commonly used features.
- Detailed examination of methodological adherence and feature traceability in existing studies.
Conclusions:
- Enhanced performance metrics, stronger methodological support, and improved feature traceability are key to advancing heart disease prediction.
- The study provides insights for researchers to improve the accuracy and reliability of automated heart disease diagnosis.
- Addressing identified concerns will guide future efforts in developing more effective heart disease prediction systems.
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.
