Related Experiment Video
Updated: Oct 2, 2025
Hydra, a Computer-Based Platform for Aiding Clinicians in Cardiovascular Analysis and Diagnosis
Published on: September 26, 2018
Automated Cardioailment Identification and Prevention by Hybrid Machine Learning Models
K S Archana1, B Sivakumar2, Ramya Kuppusamy3
1Department of Computer Science and Engineering, Vels Institute of Science, Technology & Advanced Studies (VISTAS), Chennai, India.
Insights
Machine learning models accurately predict cardiovascular disease (CHD) risk using patient data. This approach improves early detection and supports healthcare professionals in competent patient analysis.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Healthcare
Background:
- Cardiovascular disease (CHD) is a leading global cause of death, necessitating accurate early prediction for effective treatment.
- Traditional prediction methods struggle with complex data and relationships, highlighting the need for advanced analytical techniques.
Purpose of the Study:
- To apply machine learning (ML) to predict heart disease risk from historical medical data.
- To uncover data correlations for improved prediction accuracy using various ML models.
Main Methods:
- Implementation of Naive Bayes and Random Forest algorithms, including hybrid approaches.
- Utilizing 14 key patient parameters such as age, sex, blood sugar, and chest discomfort for analysis.
Main Results:
- Achieved a high prediction accuracy of 93% for identifying heart disease.
- The system provides probability percentages for developing heart disease.
Conclusions:
- The proposed ML system effectively identifies heart disease, offering improved prediction accuracy.
- This method aids physicians in competently analyzing heart patients and potentially reducing mortality rates.
Abstract:
Accurate prediction of cardiovascular disease is necessary and considered to be a difficult attempt to treat a patient effectively before a heart attack occurs. According to recent studies, heart disease is said to be one of the leading origins of death worldwide. Early identification of CHD can assist to reduce death rates. When it comes to prediction using traditional methodologies, the difficulty arises in the intricacy of the data and relationships. This research is aimed at applying recent machine learning technology to identify heart disease from past medical data to uncover correlations in data that can greatly improve the accuracy of prediction rates using various machine learning models. Models have been implemented using naive Bayes, random forest algorithms, and the combinations of two models such as naive Bayes and random forest methods. These methods offer numerous attributes associated with heart disease. This proposed system foresees the chance of rising heart disease. The proposed system uses 14 parameters such as age, sex, quick blood sugar, chest discomfort, and other medical parameters which are used in the proposed system. Our proposed systems find the probability of developing heart disease in percentages as well as the accuracy level (accuracy of 93%). Finally, this proposed method will support the doctors to analyze the heart patients competently.
