Machine Learning-Based Classification Algorithms for the Prediction of Coronary Heart Diseases

TL;DR

This study compares various machine learning algorithms for predicting coronary heart disease, finding that logistic regression performs best when applied to well-processed datasets, aiding early diagnosis and treatment.

Contribution

It demonstrates the effectiveness of logistic regression with data balancing and feature selection in predicting coronary heart disease.

Findings

Logistic regression achieved the highest performance score.

SMOTE and feature selection improved model accuracy.

LR outperformed other ML algorithms on the dataset.

Abstract

Coronary heart disease, which is a form of cardiovascular disease (CVD), is the leading cause of death worldwide. The odds of survival are good if it is found or diagnosed early. The current report discusses a comparative approach to the classification of coronary heart disease datasets using machine learning (ML) algorithms. The current study created and tested several machine-learning-based classification models. The dataset was subjected to Smote to handle unbalanced classes and feature selection technique in order to assess the impact on two distinct performance metrics. The results show that logistic regression produced the highest performance score on the original dataset compared to the other algorithms employed. In conclusion, this study suggests that LR on a well-processed and standardized dataset can predict coronary heart disease with greater accuracy than the other algorithms.