Ontological analysis of SNOMED CT

Document Sample
Ontological analysis of SNOMED CT Powered By Docstoc
					                  Copyright Notice
    Usage of semanticHEALTH public presentations
You are encouraged to use slides from this collection
   for your presentations on the condition that:
    Each slide is copied1 exactly as it is presented by the
    speaker. This means with the same colors and
    backgrounds, and including all headers and footers.
    It is not allowed to add any items including your styles
    and logos without written permission of the author(s).
    Your presentation should acknowledge the author(s) of
    the semanticHEALTH public presentation

 02.10.2006              Gergely Héja - SMCS2006               1
   Ontological analysis of
        SNOMED CT
             Gergely Héja MSc., György Surján MD., Péter Varga
              National Institute for Strategic Health Research,
                             Budapest, Hungary

02.10.2006                    Gergely Héja - SMCS2006             2
    ESKI needs a reference terminology
        To represent of classification systems (ICD,
         ICPM) and public health indicators
        To provide pre co-ordinated code lists for
         enabling semantic interoperability in Hungary
         (and potentially in the EU)
    SNOMED CT seems to be a first choice

02.10.2006              Gergely Héja - SMCS2006          3
Needs for classification systems
    Combinatory representation of categories
    based on a reference ontology to enable
        supporting statistical analysis
        formal consistency checking
        (semi)automatic interconnection of different
         classification systems (e.g. ICD and ICPM)
    A high-quality formal ontology is needed,
    but it needs not to be very detailed

02.10.2006              Gergely Héja - SMCS2006         4
             Needs for interoperability
    Detailed common terminology covering the
    whole domain of medicine
    Mapping the concepts of the HIS to the
    concepts of the common terminology and
    vice versa
    A common, consistent, comprehensive
    and decidable ontology is needed

02.10.2006            Gergely Héja - SMCS2006   5
             Computational issues
    Computability vs. comprehensiveness
    Clean hierarchies
    Less emphasis on coverage (the multitude
    of non-defined leaf categories) and greater
    emphasis on rich and well-organized high-
    level categories is needed
    SNOMED CT contains too much concepts

02.10.2006         Gergely Héja - SMCS2006    6
 Based on DOLCE formal top-level ontology
 Review of high and low-level concepts
 needed for the representation of
 classification systems and public health
      Is the subsumption relation valid?
      Meaning derived from the FSN vs.
       meaning derived from synonyms vs.
       meaning derived from formal definition
02.10.2006             Gergely Héja - SMCS2006   7
                   Error types - 1
 Misplacing concepts in the hierarchy
        smoker (an agent) is subsumed by tobacco smoking
         behaviour – finding (a role)
        severe asthma is not a kind of asthma, but a kind of
         asthma finding.
 Mixing the subsumption relation with other roles
  (typically part of)
        haemoglobin subsumes haemin (instead of
         constitutional part)
        exacerbation of asthma attack is subsumed by
         asthma (instead of temporal part)

02.10.2006               Gergely Héja - SMCS2006            8
                 Error types - 2
 Hierarchy violating medical thinking and
  biomedical knowledge
        Disease, observation and finding are
         subsumed by clinical finding
        acute on chronic, which is both subsumed by
         acute and chronic
        polycarbonate is a polymer (instead of
         synthetic polymer)

02.10.2006             Gergely Héja - SMCS2006     9
                             Error types - 3
 Contracting disjoint entities into one concept
      Smoker (an agent) and smoker (finding) (a description
       of a situation)
      additional pathologic finding in tumor specimen
       (observable entity) and additional pathologic finding
      Function is classified as an observable entity
              Ontological definition: ability of an object to play a certain role in a certain
              kind of activity
              functions (e.g. gene function, adaptation)
              measures (quality) that evaluate the realisation of a function (e.g.
              respiratory rhythm, excretory rate)
      Inflammation (morphological abnormality) (a physical
       object) and inflammatory reaction (perdurant)

 02.10.2006                           Gergely Héja - SMCS2006                            10
             Additional problems - 1
    Categories taken form classification systems
        pneumonia in other diseases classified elsewhere
         (marked as “ConceptStatus Limited”)
        The danger of taking over concepts from other
         conceptual systems: the context of the concept is lost.
         What is meant by “other diseases classified
    relations (such as part of) are represented also
    as concepts
        It prohibits the direct conversion to any formalism
         based on first order logic, thus to any DL formalism

02.10.2006                Gergely Héja - SMCS2006               11
             Additional problems - 2
      roles are not quantified (existential / universal)
      criteria are not specified (necessary / sufficient)
      conversion to DL: do we have to decide in each particular
       case, or can it be done universally?
 Multiple hierarchy
      alcoholic beverage (through its parent ingestible alcohol) is
       subsumed by central depressant, ethyl alcohol and psychoactive
       substance of abuse – non-pharmaceutical. Alcoholic drinks contain
       ethyl alcohol, which plays a role of depressant and substance of
       abuse (with respect to human beings)
      Is this a general phenomenon in SNOMED?
      Which relations are asserted and which are inferred?

02.10.2006                   Gergely Héja - SMCS2006                  12
                  Discussion - 1
    The intended meaning of the categories is
    not always clear: possible translation
    Is it reasonable to import categories from
    medical classifications?
        Size
        Artificial concepts
        Consistency errors

02.10.2006              Gergely Héja - SMCS2006   13
                  Discussion - 2
    Real world entities listed heterogeneously
        Mars bar and Kit Kat (chocolate candy would
        UFO is subsumed by transport vehicle
        tendon pulley reconstruction is represented,
         but tendon pulley not

02.10.2006             Gergely Héja - SMCS2006          14
 Use SNOMED as a plain or loosely
  structured list of terms (with extending the
  coverage). Not appropriate for intelligent
 Restructure SNOMED into a high-quality
 Build a new medical ontology from scratch
  (partial reuse of the existing ones), and to
  restrict   the    use   of    SNOMED        for
  interoperability by mapping concepts to it.

 02.10.2006        Gergely Héja - SMCS2006    15
             Restructuring SNOMED
 A formal top level ontology (e.g. DOLCE).
 A high level core reference ontology of
 general medical knowledge (e.g. anatomy,
 physiology, pathology, medical procedures).
      Logic-based formalism
      Single hierarchies with formal definitions
 (sub)domain ontologies of specialities
      Compound entities (e.g. tonsillitis)
      Manual assertion (e.g. autism)
02.10.2006              Gergely Héja - SMCS2006     16

02.10.2006     Gergely Héja - SMCS2006   17

Shared By: