GLINDA: GuideLine INteraction Detection Architecture







Guideline Interaction Ontology

Related Links


Welcome to the GLINDA Project Web site

As the baby boomers age, increasingly clinicians have to manage patients who have multiple chronic conditions. However, clinical practice guidelines, which set the standards of care, almost always focus on the management of single diseases. Similarly, clinical decision support systems for guideline-based care do not deal with the problems of multiple interacting guidelines. Studies have shown that one cannot blindly apply multiple guidelines in the management of a single patient. In the GLINDA project, we seek to build a computational architecture that allows multiple agents to generate relevant recommendations from multiple guidelines, to detect interactions among them, to repair these interactions, and to integrate, constrain, and prioritize the final management plan. In this project, we combine and extend the agent-based BioSTORM architecture (developed at Stanford) with the ATHENA clinical decision support system (created by VA and Stanford researchers using the EON technology) . We chose hypertension, diabetes mellitus, hyperlipidemia, chronic kidney disease, and heart failure as the clinical domains. As real life patient cases on which to apply our methods, we extracted anonymized patient data from Stanford's STRIDE database.

This work has been funded by a National Library of Medicine "Computational Thinking" contract.

Any views expressed here are those of the project members and are not necessarily those of the NLM or of the Department of Veterans Affairs