Our objective for this project is to analyze a heart dataset and establish a correlation among number of cardiac test variables to heart disease.
• Data Source: Original data is from a clinical study conducted in 1989 and published in the American Journal of Cardiology. The title is: "International Application of a New Probability Algorithm for the Diagnosis of Coronary Artery" by Detrano et. al. 1989. The data is also available on Kaggle Heart Disease Dataset (kaggle.com) https://www.kaggle.com/datasets/johnsmith88/heart-disease-dataset
Major observations are listed below.
• Sex was a major determinant for heart disease. Females were less likely to develop heart disease than males.
• Correlation between fasting blood sugar level (less than 120 mg/dL or more than 120 mg/dL) was not statistically significant. However, data indicates a positive correlation between fasting sugar level (>120 mg/dL) and heart disease.
• Cardiac fluoroscopy to measure calcification of major vessels was a strong indicator of development of heart disease.
Major limitations of analyses are listed below.
• The dataset was relatively small and with lower percentage of females compared to males.
• A number of related predictors of heart disease were not present in this dataset, such as lifestyle, race, diet, socio-economic status etc.
• Analysis could not establish interdependence among number of variables and heart disease outcome.
• The group presentation is available at: https://docs.google.com/presentation/d/1quj41RWc7e0ZAL_RzmAEYelfGfuSSD2WnHmeOPNvNVc/edit#slide=id.p