Scalable representations of diseases in biomedical ontologies Stefan Schulz, Djamila Raufie, Martin Boeker Institute of Medical Biometry und Medical Informatics.

Slides:



Advertisements
Similar presentations
1 Institute of Medical Biometry und Medical Informatics, University Medical Center Freiburg, Germany 2 AVERBIS GmbH, Freiburg, Germany 3 Paediatric Hematology.
Advertisements

A plant disease extension of the Infectious Disease Ontology Ramona Walls 1, Barry Smith 2, Justin Elser 3, Albert Goldfain 4 Dennis W. Stevenson 1, Pankaj.
Catalina Martínez-Costa, Stefan Schulz: Ontology-based reinterpretation of the SNOMED CT context model Ontology-based reinterpretation of the SNOMED CT.
Schulz S, Boeker M – BioTopLite – An Upper Level Ontology for the Life Sciences – ODLS 2013 BioTopLite : An Upper Level Ontology for the Life Sciences.
Toward an Ontology for General Medical Science SSFW09 September 4, 2009 William Hogan, MD, MS Associate Professor of Biomedical Informatics University.
Toward a definition of “pathological process” Richard H. Scheuermann, Ph.D. Professor of Pathology Director of Biomedical Informatics U.T. Southwestern.
SNOMED CT’s Ontological Commitment Stefan Schulz University Medical Center, Freiburg, Germany Ronald Cornet Academic Medical Center, Amsterdam, The Netherlands.
Records and situations. Integrating contextual aspects in clinical ontologies Stefan Schulz a,b Daniel Karlsson b a Institute for Medical Informatics,
Division of Biomedical Informatics Beyond Interoperability: What Ontology Can Do for the EHR William R. Hogan, MD, MS July 30 th, 2011 International Conference.
Biomedical Informatics Some Observations on Clinical Data Representation in EHRs Christopher G. Chute, MD DrPH, Mayo Clinic Chair, ICD11 Revision, World.
VT. From Basic Formal Ontology to Medicine Barry Smith and Anand Kumar.
What is an Ontology? AmphibiaTree 2006 Workshop Saturday 8:45–9:15 A. Maglia.
Extracted from “Lmo-2 interacts with Elf-2” On the Meaning of Common Statements in Biomedical Literature Stefan Schulz Department of Medical Informatics,
HL7 RIM Exegesis and Critique Regenstrief Institute, November 8, 2005 Barry Smith Director National Center for Ontological Research.
The Ontology for General Medical Science Barry Smith 11/5/2012.
Ontology-based Convergence of Medical Terminologies: SNOMED CT and ICD-11 Institute for Medical Informatics, Statistics and Documentation, Medical University.
Developing an OWL-DL Ontology for Research and Care of Intracranial Aneurysms – Challenges and Limitations Holger Stenzhorn, Martin Boeker, Stefan Schulz,
The Foundational Model of Anatomy and its Ontological Commitment(s) Stefan Schulz University Medical Center, Freiburg, Germany FMA in OWL meeting November.
Stefan Schulz Medical Informatics Research Group
BFO and Disease Barry Smith 8/ A Chart representing how John’s temperature changes 2.
Concept Model for observables, investigations, and observation results For the IHTSDO Observables Project Group and LOINC Coordination Project Revision.
SNOMED CT's Condition / Situation approach related to OGMS Stefan Schulz Medical University of Graz (Austria) ICBO 2013 Montreal OGMS Meeting.
Ontology for General Medical Science Overview and OBO Foundry Criteria Albert Goldfain Blue Highway / University at Buffalo ICBO.
Abstraction and Visualization to Support Access to NPO’s Structure and Content Michael Halper, Vladimir Ventura, Yehoshua Perl SABOC New Jersey Institute.
The Ontology of Biological Taxa
Scalable representations of diseases in biomedical ontologies and in SNOMED CT Stefan Schulz Institute of Medical Biometry und Medical Informatics University.
A CompuGroup Software GmbH, Koblenz, Germany b IMBI, University Medical Center Freiburg, Germany c Semfinder AG, Kreuzlingen, Switzerland d OntoMed Research.
1 HL7 RIM Barry Smith
A School of Information Science, Federal University of Minas Gerais, Brazil b Medical University of Graz, Austria, c University Medical Center Freiburg,
WHO – IHTSDO: SNOMED CT – ICD-11 coordination: Conditions vs. Situations Stefan Schulz IHTSDO conference Copenhagen, Apr 24, 2012.
Ontology of Disease and the OBO Foundry Chris Mungall NCBO GO Nov 2006.
Ontological Foundations of Biological Continuants Stefan Schulz, Udo Hahn Text Knowledge Engineering Lab University of Jena (Germany) Department of Medical.
From GENIA to BioTop – Towards a Top-Level Ontology for Biology Stefan Schulz, Elena Beisswanger, Udo Hahn, Joachim Wermter, Anand Kumar, Holger Stenzhorn.
Ontological Analysis, Ontological Commitment, and Epistemic Contexts Stefan Schulz WHO – IHTSDO Joint Advisory Group First Face-to-Face Meeting Heathrow,
The ICPS: A taxonomy, a classification, an ontology or an information model? Stefan SCHULZ IMBI, University Medical Center, Freiburg, Germany.
How much formality do we need ? Stefan Schulz MedInfo 2007 Workshop: MedSemWeb 2007 What Semantics Do We Need for A Semantic Web for Medicine? University.
Naïve Approach “ABCC5 ⊑  encodes.MRP5” Critique There may be ABCC5 (sensu nucleotide chain) instances that happen to never encode any instance of the.
Some Thoughts to Consider 8 How difficult is it to get a group of people, or a group of companies, or a group of nations to agree on a particular ontology?
Towards a Top-Domain Ontology for Linking Biomedical Ontologies Holger Stenzhorn a,b Elena Beißwanger c Stefan Schulz a a Department of Medical Informatics,
OGMS Monthly Call 21OCT2015. Agenda 1.Implementation- Definition and placement of ‘injury’ (Sagar) 1.Propositions + Discussion – Cleanup of ‘process’
Pre- and Post-Coordination in Biomedical Ontologies Stefan Schulz Daniel Schober Djamila Raufie Martin Boeker Medical Informatics Research Group University.
Detection of underspecifications in SNOMED CT concept definitions using language processing 1 Federal Technical University of Paraná (UTFPR), Curitiba,
. Daniel Schober, Martin Boeker, University Medical Center Freiburg ‘Ontology Simplification’ Buzzword or real Need ? OBML 2010.
Basic Formal Ontology Barry Smith August 26, 2013.
1 Biomarkers in the Ontology for General Medical Science Medical Informatics Europe (MIE) 2015 May 28, 2015 – Madrid, Spain Werner CEUSTERS 2, MD and Barry.
Introduction to Pathophysiology Dr. Manzoor Ahmad Mir Assistant Professor (Immunopatholgy) College of Applied Medical Sciences Majmaah University.
New York State Center of Excellence in Bioinformatics & Life Sciences R T U New York State Center of Excellence in Bioinformatics & Life Sciences R T U.
Stefan Schulz, Martin Boeker, Holger Stenzhorn: How Granularity Issues Concern Biomedical Ontology Integration Stefan Schulz, Martin Boeker, Holger Stenzhorn.
1 Standards and Ontology Barry Smith
Representation of Hypersensitivity and Allergy in SNOMED CT
Representation of Hypersensitivity, Allergy and adverse reactions in SNOMED CT Bruce Goldberg, MD, PhD.
Representation of Hypersensitivity and Allergy in SNOMED CT
The “Common Ontology” between ICD 11 and SNOMED CT
Cmpe 589 Spring 2006.
Discovery Seminar UE141 PP– Spring 2009 Solving Crimes using Referent Tracking Entities and their relationships - How the homework should have been.
The BioTop Family of Upper Level Ontological Resources for Biomedicine
Original model Disorder to allergic reaction Allergic state
Can SNOMED CT be harmonized with an upper-level ontology?
February 17, 2017 Bruce Goldberg, MD, PhD
DEVELOPING AN OWL-DL ONTOLOGY FOR RESEARCH AND CARE OF
Ontology authoring and assessment
Representing Natural Kinds by Spatial Inclusion and Containment
Representing Dispositions
Structured Electronic Health Records and Patient Data Analysis: Pitfalls and Possibilities. January 7, 2013 Farber Hal G-26, University at Buffalo, South.
Hypersensitivity dispositions as clinical findings
Architecture for ICD 11 and SNOMED CT Harmonization
Lexical ambiguity in SNOMED CT
Stefan SCHULZ IMBI, University Medical Center, Freiburg, Germany
A Description Logics Approach to Clinical Guidelines and Protocols
BFO meets critics Workshop organised by Barry Smith
Presentation transcript:

Scalable representations of diseases in biomedical ontologies Stefan Schulz, Djamila Raufie, Martin Boeker Institute of Medical Biometry und Medical Informatics University Medical Center Freiburg

Ontological Nature of Disease Hucklenbroich 2007: diseases are processes, events, or states Williams 2007: diseases are dispositional entities Scheuermann, Smith 2009: diseases are dispositions, disorders are abnormal bodily components, and the manifestation of diseases are pathological processes SNOMED CT: Diseases under “Disorder”, “Finding”, “Event”, (rearrangement currently being discussed in the IHTSDO Event, condition, episode PG)

Diseases, disorders, pathological processes in disjoint BFO categories acknowledgements: Ed Cheetham Deficiency of saccadic eye movements (disorder) Abnormal optokinetic response (finding) Anterior capsule opacification (finding) Anterior capsule opacification (disorder) Azoospermia (disorder) Azoospermia (finding) Bends (disorder) Does bend (finding) Bleeding from vagina (disorder) Bleeding from vagina (finding) Blister of skin AND/OR mucosa (finding) Blistering eruption (disorder) Depressive disorder (disorder) Depression (finding) Excess skin of eyelid (finding) Cutis laxa (disorder) Alcohol intoxication (disorder) Heavy drinker (finding) Disorder of skin pigmentation (disorder) Discoloration of skin (finding) Oral dyskinesia (disorder) Dyskinesia (finding) Exposure to electric current, with passage of current through tissue (event)

Two Major Problems Being pathological is rather a result of interpretation than a categorial property – Example: bleeding, pain, depression Ontologically motivated distinctions between disease, disorder, pathological process no not match the current meaning of words like “disease”, “disorder”, “abnormality” etc.

Disease matrix pathological normal structure disposition process Genetic disease Fracture Bleeding Fracture Alopecia Allergic reaction Allergy Genetic dispo- sition Discrete and disjoint ontological categories Grading of canonicity

Disease matrix pathological normal structure disposition process Genetic disease Fracture Bleeding Fracture Alopecia Allergic reaction Allergy Genetic dispo- sition Discrete and disjoint ontological categories Grading of canonicity

Redefinition: avoiding ambiguous terms like disease, disorder Disorder Pathological Structure: a combination of bodily components of or in an organism 1.that is not part of the life plan for an organism of the relevant type (thus aging or pregnancy are not clinically abnormal), 2.that is causally linked to an elevated risk of pain or other feelings of illness or of death or dysfunction on the part of the organism, and 3.that it is such that this elevated risk exceeds a certain threshold level. Disease Pathological Disposition: disposition 1.to undergo pathological processes that 2.exists in an organism because of one or more pathological structures in that organism. Pathological Process: bodily process that is a manifestation of a pathological disposition. according to Scheuermann & Smith, 2009

Formalization of Scheuermann & Smiths definitions PathologicalDisposition ⊑ ∃ inheresIn.PathologicalStructure PathologicalProcess ⊑ ∃ hasParticipant.PathologicalStructure PathologicalProcess ⊑ ∃ realizationOf. PathologicalDisposition PathologicalDisposition ⊑ ∀ hasRealization. PathologicalProcess ?

Example 1 Allergy is a disposition of specific components of the immune system of an organism. All instances of the process type Allergic Reaction, are realizations of a disposition of this type, and have an allergen as their causative agent. Image credit:

Example 2 A specific binding of thalidomide to DNA forms a pathological structure on a molecular level This structure is the bearer of the pathological disposition realized by the misdevelopment of limbs (process) and results in a body without forearms (pathological structure) Thalidomide Image credit: Bilder/contergan-co-b/missbild-bild5g.jpg

Example 3 The fracture (process) is caused by an external force, and has a fractured bone (pathological structure) as its characteristic outcome. This event is, however, not the realization of a disposition A fractured bone (structure) has many pathological dispositions which can result in a variety of pathological processes (e.g. the development of a pseudarthrosis). Image credit:

Ontological soundness vs. engineering requirements Ontology engineering: labor-intensive, use case-driven Not realistic to implement this model – in each well-founded ontology from the very beginning – for all pathological entities to be represented Problem: how can a coarse-grained, pragmatic representation (which ignores the structure / disposition / process distinction) gracefully evolve towards a more sophisticated ontology? Can this be done in a intuitive, user-friendly, ontologically sound, computable, and scalable way?

Disjunctive class PathologicalEntity ≡ PathologicalStructure ⊔ PathologicalDisposition ⊔ PathologicalProcess Top node of disease / disorder hierarchy (as long as no distinction made between processes, structures, dispositions)

Relation to organism parts / locations … crucial for defining pathological entities Different relations (e.g. OBO RO, BioTop) Pathological Structures: part-of / located-in Pathological Dispositions: inheres- in Pathological Processes: has-participant located-in

Redesign of relation hierarchy in the BioTop domain upper level ontology … allows connection to organism parts or locations, without commitment to structure, disposition, or process part-of ⊑ has-locus has-location ⊑ has-locus inheres-in ⊑ has-locus has-participant ⊑ locus-of locus-of ≡ has-locus -1 : reflexive and transitive relation

Corollaries of relation abstraction a disposition of a part is also borne by the whole a pathological structure located in a part is also located in the whole a process located in a part is also located in the whole all participants of a process are located where the process is located

PSPDPP Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2 OS PE Construction of basic disease ontology

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE PD PS PP PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE PD PS PP PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2  inheres-in

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2  has-participant

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2  has-location

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2  has-participant

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2  has-participant  inheres-in

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2  has-participant  inheres-in INCONSISTENT !

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2  has-participant  inheres-in INCONSISTENT !  inheres-in. ⊤ ⊑ PD  has-participant. ⊤ ⊑ PP PD ⊓ PP ⊑ ⊥

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2a  has-participant  inheres-in PE2

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2a  has-participant PE2

Construction of advanced disease ontology Basic components: – Top nodes Pathological Entity Organism Structure – Disease classes (broad sense) – Organism structure classes – transitive relations  has-locus  locus-of Advanced components – PathologicalStructure – PathologicalDisposition – Pathological Process – Relations  inheres-in  has-location  has-participant OS PE OS PE OS PSPDPP PEOS PE 1 PE 2 OS 1 OS 2 OS 3 OS 5 OS 4 PE 2a  has-participant PE2 PE 2b  inheres-in

Conclusions “Disease”: ontologically polymorphic category Refinement of disease classes into pathological structures, pathological dispositions, and pathological processes often not necessary Introduction of umbrella category Pathological entity, together with the high-level relation has-locus: – construction of simple model which already supports important inferences – permits graceful evolution towards more sophisticated models in which the above distinctions are introduced where necessary Implemented in BioTop ( and under discussion at IHTSDO for SNOMED CThttp://purl.org/biotop

Scalable representations of diseases in biomedical ontologies DFG, grant agreement JA 1904/2-1, SCHU 2515/1-1 GoodOD (Good Ontology Design). German HL-7 users group IHTSDO Event, condition, episode PG Acknowledgements

DFG, grant agreement JA 1904/2-1, SCHU 2515/1-1 GoodOD (Good Ontology Design). German HL-7 users group IHTSDO Event, condition, episode PG