Impact of Random Forest and XGBoost Algorithms on Improving Patient Outcomes Compared to Standard Decision-Making Methods in Healthcare Predictive Analytics

In the field of healthcare, it is important to bring quick, accurate and prompt predictions in order to serve the needs of patients better, avoid re-hospitalizations, and better allocate resources. With this, this paper proposes an ML framework for enhancing decision-making in healthcare predictive analytics employing Random Forest and XGBoost algorithms. The first and major problem solved is the inefficiency of the existing decision-making approaches which is caused by the impossibility to analyze and process big and big complex data. The proposed solution uses these machine learning models for activities such as estimating mortality, critical illness, readmission, and length of stay using massive healthcare data set. Due to the utilization of sophisticated mechanisms of machine learning, the models provide much more precise predictions to reach better results in terms of early interventions and the usage of resources. The outcomes reveal that Random Forest and XGBoost improve the prediction accuracy and XGBoost can improve most of the measures compared with Random Forest. The findings of the models suggest their usefulness for providing daily operational decision making in clinical contexts, helping enhance the delivery of care and management of resources in healthcare facilities.