Integrating Medication Recommendation and Lab Test Response Prediction for Enhanced Clinical Decision Support

Abstract

The rich data available in Electronic Health Records has led to the development of numerous systems for disease inference, mortality prediction, and personalized medication recommendations. However, integrating these diverse systems to enhance clinical decision-making in disease management remains an under-explored territory. This work describes a unified system that melds two distinct clinical tasks, namely medication recommendation and lab test response prediction, to bolster clinical decision support. This system could assist clinicians in personalizing dosage titrations and add-on medications. We present empirical studies from real-world datasets to demonstrate the potential of such a system.