Time in C|inlcal Dedsi6n Support Systems: Temporal Reasoning in ONCOCIN and ONYX Michael (3. Kahn 1, Lawrence M. Fagan 2, and Edward H. Shortliffe 2 1 Section on Medical Information Science University of California, San Francisco CA 2 Medical Computer Science Group, Knowledge Systems Laboratory Stanford University School of Medicine, Stanford CA Time is an essential part of our environment. We use temporal reasoning to analyze the past for recurring patterns or trends in order to predict and plan for future events. Temporal relationships allow us to understand how events are related and what events are likely to occur in the future. As time proceeds, new information may force a re-evaluation of our expectations made with previous data. This re-evaluation may result in altered plans or predictions based on the new expectations, it is this special property of continual change that makes time an interesting but difficult subject for computer-based modeling. Time has a prominent role in medical reasoning. Diagnoses are based on the patterns of signs and symptoms that evolve over time. Once a diagnosis emerges, prognostic factors provide insight into possible future clinical events. Management decisions are greatly influenced by the set of expected future gains
/lp/association-for-computing-machinery/time-in-clinical-decision-support-systems-temporal-reasoning-in-AsUCejS9QN