Machine Learning Approaches for Personalized Treatment Planning in Healthcare Systems

Abstract

Machine learning techniques have the potential to support healthcare professionals by analyzing patient clinical data and recommending appropriate treatment strategies. This study focuses on applying machine learning approaches to support personalized treatment planning for heart disease patients. In this research, a dataset consisting of 1000 patient records was prepared, including various clinical and demographic attributes such as age, gender, blood pressure, cholesterol level, lifestyle factors, and other heart disease–related health indicators. The dataset was synthetically generated to simulate realistic clinical conditions for research purposes. The target variable used in this study is "medication_category", which represents different treatment strategies including Combination Cardiac Therapy, Early-stage Cardiac Therapy, Intensive Cardiac Management, Lifestyle-only Intervention, Preventive Risk-factor Management, and Risk-based Preventive Therapy. To evaluate the effectiveness of machine learning in personalized treatment recommendation, four classification models were implemented: Logistic Regression, Random Forest, Support Vector Machine (SVM), and XGBoost. The dataset was divided into 80% training data and 20% testing data for model evaluation. Experimental results indicate that Logistic Regression achieved the highest accuracy of 87%, followed by XGBoost with 82%, Random Forest with 80%, and SVM with 79% accuracy. Model performance was further evaluated using precision, recall, and F1-score metrics. The findings of this study demonstrate that machine learning techniques can effectively analyze patient clinical features to support treatment decision-making for heart disease management. Among the evaluated models, Logistic Regression provided the most consistent performance for the given dataset. Therefore, machine learning–based decision-support systems have strong potential to assist healthcare professionals in providing data-driven and personalized treatment planning for heart disease patients.