Abstract

Clinical Decision Support Systems (CDSS) are increasingly deployed in healthcare to assist clinicians in diagnosis, treatment planning, and patient management. The integration of predictive modeling techniques within CDSS can enhance clinical decision-making by leveraging historical patient data, medical records, and real-time health metrics. Computational intelligence (CI) methods, including machine learning, artificial neural networks, fuzzy logic, and evolutionary algorithms, provide robust frameworks for predicting patient outcomes, identifying disease risk factors, and recommending personalized interventions. This paper explores the application of computational intelligence techniques for predictive modeling in CDSS, focusing on accuracy, interpretability, and real-time applicability. Feature selection, data preprocessing, and model optimization are emphasized to handle high-dimensional, heterogeneous, and sometimes incomplete clinical data. Experimental evaluation demonstrates that CI-driven predictive models outperform traditional statistical approaches in terms of predictive accuracy, sensitivity, and specificity. The research also highlights challenges, including data privacy, model generalization, and integration with clinical workflows. Overall, computational intelligence-based predictive modeling offers a promising approach for improving clinical decision-making, reducing diagnostic errors, and enhancing patient outcomes in modern healthcare environments