The goal of this system is to provide physicians evidence based clinical knowledge to enhance diabetic patient care. The clinical knowledge is in the form of guideline recommendations. There are various guidelines for diabetes treatment including ADA (American Diabetes Association), AACE (American Association of Clinical Endocrinologists), Diabetes Canada and so on. These guidelines, although they have several overlapping recommendations, they also have subtle differences in recommendations therefore, providing the physician with the source of the suggested treatment plan is vital to ensure the physician makes the most informed decision for any patient.
We are developing an ontology for modeling the provenance of SWRL rules and that ontology is called G-PROV or guideline provenance, it provides the physician with information about the rule that fired to generate the suggestion, and we started our research using the ADA guidelines and annotation of DMTO (Diabetes Mellitus Treatment Ontology) SWRL rules. The ontology will also include information from other guidelines, thus providing the physician with multiple sources/evidence for the suggested treatment.
A detailed description of all the resources and how they can be used to reproduce the research represented on this site can be found on the indicidual page of the resource. Click on the resource of interest to you to open the GitHub page containing the resource
Provenance can be used to enhance transparency and trust of recommendations generated by clinical decision support systems, thereby assisting healthcare practitioners by providing them with ample information to make more informed decisions. While some general provenance ontologies exist, they lack provenance terms aimed at guidelines. Provenance can be used to resolve ambiguity and conflicts between various guideline sources. It can also enable the decision support system to be easily updated when new guidelines are released. We have developed our guideline provenance ontology, by extending existing provenance ontologies, to enable accurate encoding of the source of the reasoning rules that decision support systems rely on to generate diagnosis and treatment suggestions. Our ontology enables provenance representation at different granularity levels, and can be used to annotate rules with citation backed evidence sentences and other sources of knowledge such as figures and tables. We have further developed an application to show the wide range of use cases for our ontology. We demonstrate our work using clinical practice guidelines for diabetes mellitus and discuss how our approach is easily used in diverse clinical practice guideline settings.