Accurate Prediction of Myocardial Infarction By Comparing Logistic Regression Algorithm with CatBoost Classifier

Abstract

Aim: The forecast of Myocardial Infarction for humans employing a Machine learning model by corresponding a Logistic Regression Algorithm with a CatBoost Classifier. The accuracy is enhanced by utilizing the novel LR Classifier. Materials and Methods: The study utilized a total of 20 sample iterations, with 10 samples per group. Group 1 was analyzed using a logistic regression algorithm, while Group 2 was analyzed using a decision tree classifier. The statistical power was set at 80%, and the confidence level was set at 95%. Results: The accuracy of the outcome with logistic regression is 94.61% and CatBoost Classifier is 79.516%, both the groups are statistically significant as p = 0.015 (<0.05) is the significant value in the independent sample T-test between LR and CB Classifier. Conclusion: This research concludes that the logistic regression algorithm gives the most accurate mortality with the difference of 15.1%, compared to the CatBoost Classifier.