Date of Presentation
5-4-2023 12:00 AM
College
School of Osteopathic Medicine
Poster Abstract
The objective of this project is to utilize the UCI Heart Disease dataset to identify physiological biomarkers that are highly correlated with heart disease incidence. A predictive model can then be developed using these biomarkers to estimate the likelihood of someone having or developing a heart-related condition. This study compares the efficacy of predicting cardiovascular disease as an outcome using three machine learning algorithms: Support Vector Machine, Gaussian Naive Bayes, and logistic regression. Support Vector Machine works by creating hyperplanes between data points to conduct classification. Gaussian Naive Bayes works by using the conditional probabilities of events to classify the target. In logistic regression, the independent variables included all features in the data set except for “target,” which is a categorical variable that indicates whether the patient has cardiovascular disease. The dependent variable included the “target” variable. The findings suggest that the logistic regression model had the highest accuracy in predicting cardiovascular disease. The results of this study can be beneficial to healthcare professionals in developing new preventative protocols for assessing and treating cardiovascular disease.
Keywords
Biomarkers, Heart Diseases, Statistical Models, Theoretical Models, Machine Learning, Preventive Medicine
Disciplines
Biomedical Informatics | Community Health and Preventive Medicine | Disease Modeling | Health Services Research | Investigative Techniques | Medicine and Health | Medicine and Health Sciences | Other Analytical, Diagnostic and Therapeutic Techniques and Equipment | Other Public Health
Document Type
Poster
Included in
Biomedical Informatics Commons, Community Health and Preventive Medicine Commons, Disease Modeling Commons, Health Services Research Commons, Investigative Techniques Commons, Medicine and Health Commons, Other Analytical, Diagnostic and Therapeutic Techniques and Equipment Commons, Other Public Health Commons
Cardiovascular Disease Prediction Modelling: A Machine Learning Approach
The objective of this project is to utilize the UCI Heart Disease dataset to identify physiological biomarkers that are highly correlated with heart disease incidence. A predictive model can then be developed using these biomarkers to estimate the likelihood of someone having or developing a heart-related condition. This study compares the efficacy of predicting cardiovascular disease as an outcome using three machine learning algorithms: Support Vector Machine, Gaussian Naive Bayes, and logistic regression. Support Vector Machine works by creating hyperplanes between data points to conduct classification. Gaussian Naive Bayes works by using the conditional probabilities of events to classify the target. In logistic regression, the independent variables included all features in the data set except for “target,” which is a categorical variable that indicates whether the patient has cardiovascular disease. The dependent variable included the “target” variable. The findings suggest that the logistic regression model had the highest accuracy in predicting cardiovascular disease. The results of this study can be beneficial to healthcare professionals in developing new preventative protocols for assessing and treating cardiovascular disease.