A novel framework for event stream processing of clinical practice guidelines

Clinical Decision Support Systems (CDSSs) play important roles aiding in patient care; they provide accurate data analysis and timely evidence-informed recommendations. Although the availability of biomedical data continues to flourish, there have been limited translations of this type of data to information in real-time at the bedside. Existing systems have either focused on providing process-oriented or knowledge-modeled frameworks, often relying on retrospective data analysis. We have developed a framework capable of providing clinicians the ability to represent existing knowledge and processes in realtime. This framework presents a real-time environment for modeling clinical workflow processes abstracted from clinical guidelines, while applying existing knowledge to produce intelligent evidence-informed recommendations. In this paper we provide a framework to support the detection of neonatal hypoglycaemia using a design supporting the automated realtime, evidence-informed enactment of complex business processes existing in clinical practice guidelines.