We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Toward a Model to Predict Cardiovascular Disease Risk Using a Machine Learning Approach.
- Authors
Slime, Khaoula; Maizate, Abderrahim; Hassouni, Larbi; Mouine, Najat
- Abstract
Cardiovascular diseases (CVD) remain a major global health concern, contributing significantly to both death and morbidity. To avoid premature mortality, people with heart diseases who are more likely to have a crisis should be informed as soon as feasible. In the meanwhile, the main difficulties in determining the risks of heart disease diagnosis are related to inadequate information, poor data quality, imprecision, and uncertainty. In this research, we examined a wide range of variables, with particular attention to critical risk factors identified by medical professionals, such as age, body mass index (BMI), diastolic and systolic blood pressure (BP), alcohol usage, smoking, and physical activity. Furthermore, a range of feature selection techniques is used to evaluate their utility in the prediction of CVD. To help with the creation of a predictive model that can identify and evaluate cardiovascular risk and enable the early prediction of heart attacks, we also recommend conducting a comparison analysis. Using a combination of numerous dataset attributes, we empirically assess the performance of our system and attain 70% accuracy in both training and testing data, with the best score difference of 0.15%. When 99% accuracy in training data and 63% accuracy in test data are combined, the worst difference score comes out to 36%.
- Subjects
MACHINE learning; CARDIOVASCULAR diseases risk factors; DIASTOLIC blood pressure; CARDIOVASCULAR diseases; SYSTOLIC blood pressure; CARDIOVASCULAR fitness; BLOOD pressure testing machines
- Publication
IAENG International Journal of Computer Science, 2024, Vol 51, Issue 5, p519
- ISSN
1819-656X
- Publication type
Article