1 / 38

Consistency between Metathesaurus and Semantic Network

Olivier Bodenreider Lister Hill National Center for Biomedical Communications Bethesda, Maryland - USA. Workshop on The Future of the UMLS Semantic Network NLM, April 8, 2005. Consistency between Metathesaurus and Semantic Network. Overview. Defining consistency

primo
Download Presentation

Consistency between Metathesaurus and Semantic Network

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Olivier Bodenreider Lister Hill National Centerfor Biomedical CommunicationsBethesda, Maryland - USA Workshop onThe Future of the UMLS Semantic Network NLM, April 8, 2005 Consistency between Metathesaurus and Semantic Network

  2. Overview • Defining consistency • What does inconsistency mean? • Testing consistency • Comparing Metathesaurus relations to SN relations • Aligning Metathesaurus concepts and semantic types • Semantic type distribution of sets of descendants of Metathesaurus concepts • Suggestions • Enforcement mechanism • Ontology of relationships • CVF

  3. Two levels in the UMLS

  4. Semantic Network SemanticType b SemanticType a SemanticType c Concept 2 Concept 1 Metathesaurus The UMLS: a two-level structure

  5. Semantic Types Anatomical Structure Fully Formed Anatomical Structure Embryonic Structure Disease or Syndrome Body Part, Organ or Organ Component Semantic Network Pharmacologic Substance Population Group Metathesaurus Medias-tinum Saccular Viscus 4 Angina Pectoris 97 Esophagus 12 Heart Cardiotonic Agents 225 Left PhrenicNerve Tissue Donors Heart Valves Fetal Heart 22 9 31 Concepts

  6. Fully Formed Anatomical Structure Biologic Function location of isa Pathologic Function isa Body Part, Organ, or Organ Component isa Disease or Syndrome Adrenal Cortex Adrenal Cortical hypofunction location of Relationships can inherit semantics Semantic Network Metathesaurus

  7. Defining consistency

  8. Semantic Network SemanticType b SemanticType a Concept 2 Concept 1 Metathesaurus The consistency “square”

  9. Salmonella American Medical Association The categorization link Semantic Network Organism isa Bacterium Professional Society isa is an instance of Metathesaurus

  10. Fully Formed Anatomical Structure Biologic Function location of isa Pathologic Function isa isa Semantic Network relations • 54 types of relationships • 558 asserted relations (SRSTR) • 6703 fully expanded relations (SRSTRE*) Semantic Network Body Part, Organ, or Organ Component Disease or Syndrome

  11. REL vs. RELA Not always labeled 106 additional types of relationships ~7 M symbolic relations Metathesaurus Medias-tinum Saccular Viscus 4 Angina Pectoris 97 Esophagus 12 Heart Cardiotonic Agents 225 Left PhrenicNerve Tissue Donors Heart Valves Fetal Heart 22 9 31 Concepts Metathesaurus relations

  12. has_finding_site Oat cell carcinoma of lung Carcinoma, Small Cell SCLC Lung structure Lung Pulmonary has_finding_site Metathesaurus relations • Recorded • at the term level: from source vocabularies • at the concept level: from Metathesaurus editors • Aggregated at the concept level Oat cell carcinoma of lung Carcinoma, Small Cell SCLC Lung structure Lung Pulmonary

  13. Not all relationships in hierarchies are isa (1) Autoimmune Diseases is generally a Addison’s disease Tuberculous Addison’s disease Addison’s disease due to autoimmunity

  14. Not all relationships in hierarchies are isa (2) Environment and Public Health [G03] Public Health [G03.850] Accidents [G03.850.110] Accident Prevention [G03.850.110.060] + Accidental Falls [G03.850.110.085] Accidents, Aviation [G03.850.110.185] […] Drowning [G03.850.110.500] +

  15. SN rel. and Meta rel. must have the same direction SN rel. and Meta rel. must be of the same type (both hierarchical or associative) Meta rel. must be the same as SN rel. or one of its descendants Semantic Network SemanticType b SemanticType a Concept 2 Concept 1 Metathesaurus Defining consistency

  16. Pneumonia Lung Examples of consistent relations Disease orSyndrome Body Part, Organ,or Organ Component has_location has_location

  17. Semantic Network SemanticType b SemanticType a Concept 2 Concept 1 Metathesaurus Examples of consistent relations affects treats

  18. What does inconsistency mean?

  19. Semantic Network Semantic Network SemanticType b SemanticType b SemanticType a SemanticType a Concept 2 Concept 2 Concept 1 Concept 1 Metathesaurus Metathesaurus The consistency “square” revisited ? ? ? ?

  20. What does inconsistency mean? • Inaccurate/missing Semantic Network relation • Inaccurate (/missing?) categorization • Inaccurate Metathesaurus relation

  21. Testing consistency

  22. (A) Consistency of associative relations [McCray& Bodenreider, 2002]

  23. Results • 6894 pairs of related concepts • 4496 (65%): a SN relation can be inferred unambiguously • Validity confirmed in 1981 cases • 2515 not labeled in the Metathesaurus • 1491 (22%): multiple possible SN relationships • multiple possible Metathesaurus relationships • 907 (13%): inconsistency SN/Meta relationships • 372: no SN relationship between the STs • 415: inconsistent SN/Meta relationship type (REL) • 120: inconsistent SN/Meta relationship attribute (RELA)

  24. (B) Consistency of hierarchical relations • Relations used • SN: isa • Categorization: isa • Metathesaurus: PAR/CHD + RB/RN • Hypothesis • For a pair of (ST, C), the concepts categorized by ST (and its descendants) correspond to the descendants of the concept C • In the set of descendants of C, expected STs are the ST of C (and its descendants) ➊ ➋

  25. Semantic type List of all conceptshaving this semantic type Concept List of all descendants Comparing the 2 sets Intersection of the 2 sets ST-based classes vs. descendants ➊ [Bodenreider& Burgun, 2004]

  26. Invertebrate Class Reptilia Tadpole Amphibians and Reptiles PharmacologicSubstance Toad licking Ranaunclassified Analyzing inconsistencies Miscategor-ization (?) Amphibian Amphibia Miscategor-ization Wronghierarchicalrelation Missinghierarchicalrelation 1124in common 1135concepts 1126descendants

  27. Concept Set of all descendants Distribution of semantic types in the set Allowable STs: ST of C and its descendants (strict) or ST from the same semantic group (loose) Semantic types of descendants ➋ [Mougin& Bodenreider, 2005]

  28. Mental Process Finding Hostility Reaction belligerent Analyzing inconsistencies • 26,584 concepts studied • 59% of their descendants have a semantic type incompatible with that of the original concept

  29. Analyzing inconsistencies # ------------------------------------------------------------ # C0597249 Neoplasm of placenta (disorder) (neop) # * B: 190 C0597249|ST|acab| 5.50|incp C0597249|ST|anab| 1.50|incp C0597249|ST|cgab| 76.50|incp C0597249|ST|dsyn| 27.50|incp C0597249|ST|inpo| 1.00|incp C0597249|ST|neop| 76.50|comp C0597249|ST|patf| 1.50|incp C0597249|SG|DISO| 190.00|comp # ------------------------------------------------------------

  30. Suggestions

  31. Aligning SN and Meta relationships • 54 types of SN relationships • 106 additional types of Metathesaurus relationships • Some are simply synonymous(caused_by / due_to; follows / temporally_follows) • Some are specialized relationships(manifestation_of / definitional_manifestation_of) • Many types of mapping relationships, not in SN

  32. Add classification information to SN • Explicit classificatory principles (in addition to textual definition and examples) • Abandon economy principle and return to JEPD (jointly exhaustive/pairwise disjoint) approach

  33. Metathesaurus editing environment • Use SN/Meta relation consistency as a constraint for assigning semantic types • Use SN relations to suggest labels for unspecified Meta relations • Use SN/Meta relation consistency to guide the review by the Metathesaurus editors • Inaccurate categorization? • Inaccurate Metathesaurus relation?

  34. Conclusions

  35. Conclusions • Simultaneously • Improve SN • Improve categorization • ST assignment can be automated in part

  36. Some references • McCray AT, Bodenreider O.A conceptual framework for the biomedical domain.In: Green R, Bean CA, Myaeng SH, editors. The semantics of relationships: an interdisciplinary perspective. Boston: Kluwer Academic Publishers; 2002. p. 181-198. • Bodenreider O, Burgun A.Aligning knowledge sources in the UMLS: Methods, quantitative results, and applications.Medinfo 2004:327-331.

  37. Some references • Burgun A, Bodenreider O.Aspects of the taxonomic relation in the biomedical domain.In: Welty C, Smith B, editors. Collected papers from the Second International Conference "Formal Ontology in Information Systems": ACM Press; 2001. p. 222-233. • Mougin F, Bodenreider O.Approaches to eliminating cycles in the UMLS Metathesaurus: Naive vs. formal.Proceedings of AMIA Annual Symposium 2005:(submitted).

  38. MedicalOntologyResearch Contact: Web: olivier@nlm.nih.gov mor.nlm.nih.gov Olivier Bodenreider Lister Hill National Centerfor Biomedical CommunicationsBethesda, Maryland - USA

More Related