Speaker: Laurie Nelsen, Sr. Manager – Ontologist, Mayo Clinic
[These are my notes from the KMWorld 2014 Conference. Since I’m publishing them as soon as possible after the end of a session, they may contain the occasional typographical or grammatical error. Please excuse those. To the extent I’ve made any editorial comments, I’ve shown those in brackets.]
Session Description: Mayo Clinic’s delivery of high-quality, affordable healthcare depends on integrating knowledge to promote innovation across patient care, research, and education. Providing the best current health information and services requires an agile and responsive content management ecosystem for creating and managing content as well as meeting the emerging needs for the delivery of “smart” content. The Clinic’s solution was to extend traditional CM technologies with a semantic services layer to support standards-based knowledge interoperability within and between organizations. Nelson shares the technical architecture and design choices made to build and deploy its Knowledge Content Management System (KCMS). KCMS’s solution to the problem of knowledge integration and flexible access is twofold: First, it utilizes the capabilities of the CMS to author, manage, and deliver the information. Secondly, it tightly integrates the CMS with a semantic services layer that provides the intelligence that enables users to find the right information, no matter who authored it or how it is stored.
- Start by defining the problem: Content management system (CMS) technology provides content, authority and delivery functionality, but is fundamentally different than vocabulary and annotation (tagging) management technology, which provides the semantic context of the content to support findability.
- Then learn to tell the story well: Create success story about how the problem could be solved and then told it, over and over again. Their story illustrated “semantics in action.” (For an example, see their MayoClinic.com guide on Alzheimer’s disease.
- Their approach: Their approach involved creating a pattern with a semantic overlayer to the content manager. This could be used to create one or one thousand disease guides. They were also able to replace manual links with new dynamically generated links that were organized by the semantic layer. As a result, the organization banned all manual links.
- Vocabularies: While they try to use as many of the standardized vocabularies, they found that there were not great standard patient-facing or consumer-facing vocabularies. So they had to create those themselves. They have a series of ontologies: people, organization, medical condition, clinical studies, etc. Once they started identified the connections among these ontologies, they found powerful relationships.
- Next stage: They are working on integrating their systems into a single system. They have learned that innovation does not end with implementation of the technical solution. You need upgrades and you need to continuous improve. They also need to find and tell new stories.
- True adoption is a really long process: you need to keep nurture the tool and you need to keep telling potential users about how it works and how it can help.
- Understand and exploit your tools and systems: Why drive a Honda when you have a Maserati in the garage?
- Biggest Lesson Learned: It’s really about the story. Read The Leader’s Guide to Storytelling by Stephen Denning.