Knowledge and Data-Driven Framework for Designing a Computerized Physician Order Entry System

A major concern related to the healthcare industry is uniformity in delivery. There considerable variation diagnosis and treatment of patients depending on experience expertise doctors. Information technology can play a role addressing this issue. Research investigating use data-driven approaches knowledge-driven clinical pathways achieve delivery ongoing. Specifically, integration data be used ensure patient care, thus avoiding inappropriate variance. The approach utilize bulk medical being generated. form incorporates evidence-based care. In context, knowledge computerized physician order entry (CPOE) systems are gaining importance systems. This paper proposes framework for CPOE system. work based base populated with disease quadruples, each which comprises list symptoms, tests, results, medications particular disease. proposed system obtained from two datasets, namely, MIMIC (Medical Mart Intensive Care) Disease-Symptom Knowledge Database, operational New York-Presbyterian Hospital (NYPH). We combined both datasets using common attribute generate more attributes aid decision making. was performed help specialists knowledge. resulting further integrated before extraction quadruples. novelty lies quadruples pathways. dynamically ranks element quadruple its association score facilitate generation prescription sets CPOE. effectiveness providing uniform care has been validated by experts. significantly improve safety quality due capability