A Model Utilizing Stacking Classifiers for Identifying Heart Irregularities and Forecasting Cardiovascular Disease

Résumé fourni par la source

The medical sector generates vast amounts of patient data, covering areas such as liver health, diabetes, heart conditions, and maternal health. Among these, cardiovascular diseases (CVDs) are considered the deadliest, representing a significant health risk due to their silent and non-communicable nature. Preventing CVDs requires thorough examination of health records, which can be quite challenging. So, it is necessary to use machine learning (ML) technology to identify early indications of CVD. Here, we used four different types of boosting methods, such as adaptive boosting, gradient boost, eXtreme gradient boosting, and stacked boosting, used with and without hyperparameter fine-tuning to predict CVD-affected patients. The results demonstrate that the stacked boosting algorithm has a very high recall score equated to the single boosting model, which is achieving a remarkable 93% recall subsequently using hyperparameter tuning. This will support the hospital staff to take precautionary action and thus help the hospital managers to dynamically allocate resources, which in turn will reduce the financial cost for the patient.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.