Clinical Decision Support Systems (CDSS) are tools used in a diverse range of healthcare environments to support physicians and improve overall healthcare quality. Electronic Medical Records (EMRs) play a crucial role in enhancing CDSS performance by providing them with sophisticated, well-organized digital data. This enables these systems to operate seamlessly in different medical contexts. Without the support of these integrated tools, medical staff may face difficulties in their daily tasks, which can have adverse consequences for patient health. In this study, we propose a CDSS that enables better management of drug iatrogenicity while improving patient safety and treatment precision. The system features management and prediction functionalities designed to improve nephrology service quality. The prediction subsystem exploits EMRs data and uses machine learning approaches capable of identifying drug interactions. In addition, it personalizes drug prescriptions for renal failure patients, ensuring a tailored and efficient approach to treatment management. Drug interaction data come from the pharmacology lab at the University Hospital of Oran, Algeria, while EMRs are from the nephrology service at the same establishment. The aim of the CDSS is to improve drug safety, optimize treatment precision, and personalize drug management in nephrology. By integrating EMRs with CDSS, our system plays a key role in proactive healthcare management in nephrology. It enhances drug interaction detection and treatment personalization.