Enhancing Clinical Decision Support Systems Through Patient Risk Level Prediction Using Machine Learning Techniques

Clinical Decision Support Systems (CDSS) assist healthcare professionals by providing data-driven insights to support diagnosis and treatment decisions. However, many existing CDSS frameworks face limitations in adaptability, scalability, and interpretability across diverse clinical datasets. This paper presents an enhanced CDSS framework that applies supervised machine learning algorithms—Support Vector Machine (SVM), Random Forest (RF), and AdaBoost—for patient risk level prediction. The dataset undergoes preprocessing steps including normalization and feature selection to improve model performance. Evaluation metrics such as accuracy, precision, recall, F1-score, and ROC-AUC are used to assess model effectiveness. Among the implemented algorithms, the Random Forest classifier achieved the highest predictive accuracy of 99%, outperforming other models. The trained model is integrated into a Django-based web application for real-time patient risk assessment. The proposed system improves predictive reliability and clinical usability, thereby supporting timely and informed medical decision-making.