Meet The Speaker
“Philosophy and reason as elements of computer-based decision support: A plea to stop searching for our keys only where the light is”
Dr. Musen is Professor of Biomedical Informatics at Stanford University, where he is Director of the Stanford Center for Biomedical Informatics Research. Dr. Musen conducts research related to intelligent systems, the Semantic Web, reusable ontologies and knowledge representations, and biomedical decision support. His long-standing work on a system known as Protégé has led to an open-source technology now used by thousands of developers to build intelligent computer systems and new computer applications for e-science. He is principal investigator of the National Center for Biomedical Ontology. He chairs the Health Informatics and Modeling Topic Advisory Group for the World Health Organization’s revision of the International Classification of Diseases (ICD-11). Dr. Musen received the Donald A. B. Lindberg Award for Innovation in Informatics from the American Medical Informatics Association in 2006. He has been elected to the American College of Medical Informatics and the Association of American Physicians. He is founding co-editor-in-chief of the journal Applied Ontology.
Seminar Abstract
“Philosophy and reason as elements of computer-based decision support: A plea to stop searching for our keys only where the light is”
Despite enormous early promise, computer-based decision-support technology has had only a modest affect on clinical practice and almost no affect on clinical outcomes. Although the conventional wisdom is that we will begin to see more benefit as increasing numbers of decision rules are rendered in computational form, this attitude ignores (1) the need to offer decision support for clinical practice guidelines whose recommendations necessarily unfold over time and (2) the desire of practitioners to follow the spirit of guidelines, even for patients who have complex medical problems and several, potentially interacting conditions. Computer systems that model guidelines as longitudinal treatment plans, rather than as myriad situation–action rules, offer major advantages for addressing the messiness of real clinical situations. I will discuss how the use of formal ontologies to represent the entities that comprise clinical guidelines coupled with flexible systems to reason about the guidelines leads to decision-support frameworks that may be more useful and more usable than the technologies that are currently promoted by the vendor community. The approach has been tested in a system called ATHENA CDS, developed collaboratively with Prof. Mary Goldstein at Stanford, and deployed at several VA medical centers.
