Problem-Oriented Translational Health Informatics for Evidence Based Medicine and Privacy Enhancing

As clinical practice has become increasingly evidence-based, depending only on the controlled clinical trials approach to provide evidence becomes a bottleneck challenge. It is not only due to the tough criteria to design, execution, and analysis of the investigations associated with these clinical trials but also due to the missing link to bring more evidence through data-driven analytics. Data-driven analytics is the process of interpreting or translating the quantitative data at the bench to reveal qualitative insights, answer questions, provide evidence and identify prognosis trends. It also referenced as the translational medicine from the bench to the bedside. While this translational approach hold the promise of providing innovative and personalized medical evidences that can have important implication to enhance patient care and enforce more interoperable privacy preserving policies, it is only represent one way system that do not consider the reverse direction of including the clinical science at the bedside. Based on Lawrence Weed (MD) initiative that was introduced in early 1970, this research provide an investigation into adopting more comprehensive translational solution that provide the integration between the bench and the bedside into directions based on the notion of knowledge couplers (which is part of Weed’s POMR initiative (Problem-Oriented Medical Record)). Our proposed bridging utilizes the problem-list interconnectivity and the constraints graph. Our QL4POMR prototype has been extended to demonstrate the applicability of these knowledge couplers for achieving translational informatics providing essential evidence-based decision-making services and privacy preserving tasks based on GraphQL APIs.