ebfcc1d768ed74c1baa5c60ed227db69.ppt
- Количество слайдов: 16
The Inter. Med. TM Collaboratory –the early years (1994 -1996) o Biomedical informatics researchers & systems developers at 5 sites: n Harvard/Brigham and Women’s Hospital n Harvard/Massachusetts General Hospital n Columbia Presbyterian Medical Center n University of Utah School of Medicine n Stanford University School of Medicine
The initial Inter. Med concept o Biomedical application development too slow o Silos, limited reuse o Internet could provide common resources (middleware) n n n Domain-specific Seven-tiered model Built on infrastructure “service” layer
Inter. Med’s Seven-Tiered Model Policies & Procedures Applications Composition Methods Component Tools Data Bases & Knowledge Bases Vocabulary / Taxonomy Infrastructure Classes & Services
Collaborative Model Development UMLS Input/Output Specifications and Protocols Browsing & Maintenance Requirements Implemented using convenient representation language SNOMED (Columbia) LOINC IVORY (WARP) Inter. Med Vocabulary Model Inter. Med Vocabulary Server Local vocabs for new applications The. Net. Sys. . . Map into local vocabularies for existing applications • Define meta-terminology • Keep implementation-independent • Provide generic solution Populated with sample domain-specific content
Intended applications o Linking DXplain to a clinical information system o Internet-based clinical guidelines o Internet-based clinical research queries
Component development work o DXplain access (MGH) - tiers 3, 4 o Guideline tools (MGH, BWH versions) - tiers 3, 4 o Semantic net tool: The. Net. Sys - tier 2, 4 o Semantic model for CXR reports - tier 3 o Form tool (a la HTML 3) using semantic model - tiers 3, 4 o Development of import/export (The. Net. Sys/Ontolingua) o Image tool (with pan, zoom, grayscale adjust) - tier 4 o Tree walker for state maps & protocols - tier 4 o WWW access modules - tier 4
Application construction tools (Tier 5) o Interface builder n n Use of visual embedding/ compound document architectures (Open Doc, OLE-2) Script-based communication among components o Work-flow-management-oriented framework n An approach to integration based on clinical guidelines: the GEODE-CM project
Early findings ~ 1995 o Activities weren’t converging o Progress in the different labs was not aligned n Each was working on what interested it o Vision was too abstract n Hard to identify measures of success o We decided to select a single application focus to drive the project n Needed to be something all of the labs cared about
Common guideline representation as a second year focus o Multiple purposes of guidelines n education (reference) n training (how to do something) n problem-solving n responding to events (alerts) n protocolized care/data capture n care plans/critical paths -- UR and QA n consultation n workflow management (GEODE)
Multiple representation approaches o Flow charts o Narratives o Decision tables o Rule-base o MLM procedures o State model database
Collaborative Model Development AHCPR Guidelines Input/Output Specifications and Protocols Browsing & Maintenance Requirements Implemented using convenient representation language T-Helper Protocols CPMC Clinical Pathways & Guidelines UTI and Other Guidelines at MGH BWH GEODE Clin. Mgt. Pathways. . . Inter. Med Guidelines Model Inter. Med Guidelines Server Map into local representation for existing applications Local versions for new applications • Define meta-terminology • Keep implementation-independent • Provide generic solution Populated with sample domain-specific content
Goals to explore: o Whether a common representation could be devised that maintains all of the information needed to support all of the applications o How the guideline knowledge should interact with vocabulary knowledge o What components and tools would be needed
Work was then able to be partitioned o Requirements analysis o Model development o Tool building o Testing and refinement o Application exploration o Cognitive studies o SDO participation, conference planning, community development
Difficulties still occurred when: o Groups disagreed about emphasis or approach o No one signed up for a needed task o Need for continuing work on maintaining a shared vision
Collaboratory issues o Ideal C’y needs shared vision and buy-in n n strong business case intellectual and practical appeal o Tight project management n funding dependent on performance o Can peer-based C’y be productive? o Can research-based C’y (vs. product- focused) be successful?
ebfcc1d768ed74c1baa5c60ed227db69.ppt