Cardiovascular Disease Prediction Using Machine Learning

Machine Learning Abstract Machine Learning is an important part to this area. Prediction of cardiovascular diseases but not so much for perforation. Mance should be considered within its relation to a particular dataset or application domain. Definition of definition and Validation process. This study demonstrates And evaluates a model for predicting cardiovascular risks from prototypes to models. The prediction app was created by developing it on streamlit website. There are 1000 synthetic records of demographics for this project. physiological, laboratory, clinical and lifestyle variables and can Extract 4 more relevant to risks for this model. . Data is divided by means of an 80-20 stratified train test split. After that Logistic Regression, Random Forest and Gradient Boosting. Classifier boosting models are initialized from an arbitrary starting point of a given random seed value. The model performance was evaluated on a 200 record dataset where it had an accuracy rate of. 0.8800, 0.9050, and 0.8900 with an ROC-AUC score of 0.9542. 0.9615 and 0.9641 respectively. Random Forest achieved the Highest accuracy, precision, specificity, and F1-score compared to others. Gradient Boosting achieved the highest ROC-AUC. Analysis of The implementation shows an overall risk for this product. GradientBoosting score is shown instead of an average value for it. true ensemble prediction. While these models are good at predicting outcomes. On a dataset generated from an identical artificial intelligence model, these are. Results are not reliable for diagnosis nor of clinical value. in real patients. Therefore, it is better to consider this as an app. an educational machine-learning prototype and a reproducible Foundations of future studies on ethically obtained medical data. Data analysis, calibration evaluation, and external verification.