The Ontologically Privileged Status of the Past

1
The Ontologically Privileged Status of the Past
Barry Smith
2
Universals vs. instances

• Assertions in scientific texts pertain to
  universals in reality
• Assertions in a lab report pertain (also) to
  instances of these universals

3
Universals are those invariants in reality

• which make possible
• the use of general terms in scientific inquiry
• the use of standardized therapies in clinical
  care
• the use of standardized procedures in business
  transactions
• ...

4
universals instances / particulars
scientific texts, dictionaries diaries, biographies, histories, journalism
medical ontologies, terminologies clinical records, lab reports, X-ray images
macroeconomic surveys credit card transaction records
databases databases
5
universals / concepts instances / particulars
scientific texts, dictionaries diaries, biographies, histories, journalism
medical ontologies, terminologies clinical records
macroeconomic surveys credit card transaction records
databases databases
relate indiscriminately to past, present and
future
6
universals / concepts instances / particulars
scientific texts, dictionaries diaries, biographies, histories, journalism
medical ontologies, terminologies clinical records, lab reports, X-ray images
macroeconomic surveys credit card transaction records

relate only to the past
7
universals instances / particulars
scientific texts, dictionaries diaries, biographies, histories, journalism
medical ontologies, terminologies clinical records
macroeconomic surveys credit card transaction records
databases databases
8
universals / concepts instances / particulars
scientific texts, dictionaries diaries, biographies, histories, journalism
medical ontologies, terminologies clinical records
macroeconomic surveys credit card transaction records
databases databases
9
UMLS Semantic Network
10
Anatomical Space
Anatomical Structure
Organ Cavity Subdivision
Organ Cavity
Organ
Serous Sac
Organ Component
Serous Sac Cavity
Tissue
Serous Sac Cavity Subdivision
Pleural Sac
Pleura(Wall of Sac)
Pleural Cavity
Parietal Pleura
Visceral Pleura
Interlobar recess
Mediastinal Pleura
Mesothelium of Pleura
Foundational Model of Anatomy
11
Anatomical Space
Anatomical Structure
Organ Cavity Subdivision
Organ Cavity
Organ
Serous Sac
Organ Component
Serous Sac Cavity
Tissue
Serous Sac Cavity Subdivision
is_a
Pleural Sac
Pleura(Wall of Sac)
Pleural Cavity
part_of
Parietal Pleura
Visceral Pleura
Interlobar recess
Mediastinal Pleura
Mesothelium of Pleura
12

• Gene
• Ontology

13

• Gene
• Ontology

14
Holy grail of biomedical informatics
integration of genomic and EHR data
Main obstacles 1. Poor facility for dealing
with time and instances / particulars in current
ontologies 2. Poor facility for dealing with
instances / particulars in current clinical
record systems

15
Current ontologies are about meanings
(concepts, conceptualizations)
16
The Ontologically Privileged Status of Universals
17
The Ontologically Privileged Status of
Universals (a.k.a. Concepts)
18
The concept diabetes mellitus becomes
associated with a diabetic patient

• concept patient concept diabetes
• what it is on the
• side of the patient

?
?
19
The concept diabetes mellitus becomes associated
with a diabetic patient

• concept patient concept diabetes
• what it is on the
• side of the patient

?
?
20
A is_a B def. A is more specific in meaning
than B

• A contains B def
• the concept A stands in a containment relation to
  the concept B
• A causes B def
• the concept A stands in a causative relation to
  the concept B

21
GALEN

• vomitus contains carrot

22

• UMLS Semantic Network
• Food causes Experimental Model of Disease
• Biomedical or Dental Material causes Mental or
  Behavioral Dysfunction
• Manufactured Object causes Disease or Syndrome

23
vomitus contains carrot

• The authors of ontologies have not paid
  attention to the question whether these are all
  or some assertions

24
because they have not paid attention to instances

• some instances of vomitus contain instances of
  carrot
• all instances of vomitus contain instances of
  carrot

25
IFOMIS proposal move from associative
relations between concepts/meanings to strictly
defined relations between the universals (types,
kinds) in realityembraced also by Gene Ontology
Consortium
26
Key idea

• Ontological relations like
• contains, part_of, causes
• are relations between universals,
• but to define them properly we need to take
  account of instances and of time

27
Three kinds of relations

• ltuniversal, universalgt is_a, part_of, ...
• ltinstance, universalgt this throb here and now
  instance_of the class throb
• ltinstance, instancegt Marys heart part_of Mary
  at t

28

• part_of
• A part_of B def.
• given any particular a and any time t,
• if a is an instance of A at t,
• then there is some instance b of B
• such that
• a is an instance-level part_of b at t
• HAS ALL-SOME FORM

29
transformation_of
mature RNA transformation_of pre-RNAadult
transformation_of child
30
transformation_of

• A transformation_of B def
• for all a, t, if a is an instance of A at t then
  there is some t earlier than t which is such
  that a is an instance of B at t

HAS ALL-SOME FORM
31
transformation_of

• in short
• A transformation_of B def. any instance of A
  was at some earlier time an instance of B
• Contrast
• A transforms_into B
• child transforms_into adult
• The ontologically privileged status of relations
  pointing towards the past

32
embryological development
33
tumor development
C1
C c at t
c at t1
34
Advantages of the methodology of enforcing
commonly accepted coherent definitions

• promote quality assurance (better coding)
• guarantee automatic reasoning across ontologies
  and across data at different granularities
• yields direct connection to times and instances
  in EHR

35
The story of Jane Smith(with thanks to Werner
Ceusters)
36
July 4th, 1990 Jane goes shopping
37
A visit to the hospital

• City Health Centre Dr. Peters
• (City HC) Dr. Longley

38
Diagnosis a severe spiral fracture of the femur
39
The City HCs medical record

• captures in a structured form all of the
  clinically significant information in the
  narrative notes

Rector AL, Nowlan WA, Kay S, Goble CA, Howkins
TJ. A framework for modelling the electronic
medical record. Methods Inf Med. 1993
Apr32(2)109-19.
40
Structured Medical Record
41
CityHCs representation formalismfor statements
in records
Occurrences are specific occurrences of
individuals and must be situated in space and
time. The most important group of occurrences are
observations i.e. agents observations of
individuals.
42
City HCs EHR model
43

• Rector et al
• Every occurrence level statement concerning the
  Jane Smiths Fracture of the Femur is an
  observation of the corresponding individual.
• The existence of the individual Jane Smiths
  Fracture of Femur does not imply that Jane Smith
  has, or has ever had, a fracture of the femur,
  but merely that some observation has been made
  about Jane Smith regarding a fracture of the
  femur.
• (The only observation recorded about Jane
  Smiths Fracture of the Femur might be that she
  did not have it.)

44
Problems
45
Main problems of EHRs

• Statements refer only implicitly to the concrete
  entities about which they give information.
• Codes are general they tell us only that some
  instance of the class the codes refer to, is
  referred to in the statement, but not what
  instance precisely.
• Mixing up the act of observation and the thing
  observed.
• Mixing up statements and the entities these
  statements refer to.

46
Consequences

• Difficult to
• count the number of (numerically) different
  diseases
• Bad statistics on incidence, prevalence, ...
• Bad basis for health cost containment
• relate (numerically the same or different) causal
  factors to disorders
• Dangerous public places (specific work floors,
  swimming pools), HIV contaminated blood from
  donors, food from unhygienic source, ...
• Hampers prevention

47
Proposed solutionReferent Tracking

• Purpose
• explicit reference to the concrete individual
  entities relevant to the accurate description of
  each patients condition, therapies, outcomes,
  ...
• Method
• Introduce an Instance Unique Identifier (IUI) for
  each relevant particular / instance

48

• CUI (coo-ey) Concept Unique Identifier (e.g. a
  SNOMED code)
• UUI (oo-ey) Universal Unique Identifier
• IUI (you-ey) Instance Unique Identifier (e.g. a
  Social Security Number)

49
Referent tracking

• a response to the hard NLP problem of reference
  resolution in running text

50
Ontology
An ontology is a representation of some
pre-existing domain of reality which (1) reflects
the properties of the objects within its domain
in such a way that there obtains a systematic
correlation between reality and the
representation itself, (2) is intelligible to a
domain expert (3) is formalized in a way that
allows it to support automatic information
processing
51
Basic Formal Ontology
Continuants Occurrents / Events
endure identically through time while undergoing changes, including gaining and losing parts unfold themselves through time in successive temporal phases
52
Basic Formal Ontology
Continuants Occurrents
Independent things, substances Always dependent on their bearers (participants/agents)
Dependent functions, qualities, shapes, roles ... Always dependent on their bearers (participants/agents)
53
An ontological analysis
continuants
54
Essentials of Referent Tracking

• Deciding what particulars should receive a IUI
• Finding out whether or not a particular has
  already been assigned a IUI
• Using IUIs in the EHR, resolve issues concerning
  the syntax and semantics of statements containing
  IUIs
• Correcting errors in the assignment of IUIs
• Dealing with relation between IUI-identified
  instances and corresponding universals

55
Architecture of aReferent Tracking System (RTS)

• Ideally set up to be as geographically broad in
  scope as possible
• Services
• IUI generator
• IUI repository statements about assignments and
  reservations
• Referent Tracking Database (RTDB) statements
  relating instances to instances and universals

56
IUI generation

• Universally Unique IDs
• recently standardized through ISO/IEC
  9834-82004,
• specifies format and generation rules enabling
  users to produce 128-bit identifiers that have a
  very high probability of being globally unique
• Meaningless strings

57
IUI assignment

• an act of labelling carried out by the first
  cognitive agent needing to acknowledge the
  existence of a particular it has information
  about
• cognitive agent
• A person
• An organisation
• A device or software agent, e.g.
• Bank note printer
• Image analysis software
• Credit card transaction reader

58
Criteria for IUI assignment (1)

• The particulars existence must be determined
• Easy for persons in front of you, for body parts,
  for X-ray images
• More difficult for subjective symptoms
• No need to know what the particular exactly is,
  i.e. which universal it instantiates
• No need to be able to point to it precisely
• One bee out of a particular swarm that stung the
  patient

59
Criteria for IUI assignment (2)

• The particular should not have been already
  determined as something else
• Morning star / evening star
• May not have already been assigned a IUI.
• Must be salient/relevant/significant
• Personal decision, (scientific) community
  guideline, ...
• Reflects a possibility offered by the EHR system
• Once a IUI has been assigned, everybody making
  statements about this particular should use it

60
IUI assignments

• The act of IUI assignment can be represented as
• ltda , Ai , tdgt
• da IUI of the registering agent
• Ai ltpa, pp, tap, cgt
• pa IUI of the author of the assertion
• pp IUI of the particular
• tap time of assignment
• c optional description
• td time of registering Ai in the
  IUI-repository
• Neither td nor tap give any information about
  when pp began to exist.

61
Management of IUI-repository

• Adequate safety and security provisions
• Access authorisation, control, read/write, ...
• Pseudonymisation
• Deletionless but with facilities for correcting
  mistakes.
• Central management with adequate search
  facilities.

62
Representation in the EHR

• Relevant particulars referred to using IUIs
• Relationships that obtain between particulars at
  time t expressed using strictly defined relations
  from an ontology
• Statements describing for each particular at time
  t of what universal from an ontology it is an
  instance

63
PtoP (particular to particular) statements
ordered sextuples of the form ltsa, ta, r, o, P,
trgt
sa the IUI of the author of the statement, ta
the time when the statement is made, r a
relationship obtaining between the particulars
referred to in P, o the ontology from which r is
taken, P an ordered list of the IUIs of the
particulars between which r obtains, tr the
time at which r obtains.
64
PtoCL (particular to class) statements
ltsa, ta, inst, o, p, cl, trgt sa the IUI of the
author of the statement, ta the time when the
statement is made, inst an instance
relationship available in o obtaining between p
and cl, o the ontology from which inst and cl
are taken, p the IUI of the particular whose
inst relationship with cl is asserted, cl
the class in o to which p enjoys the inst
relationship, tr the time at which the
relationship obtains.
65
PtoCO (particular to concept code) statements
ltsa, ta, cbs, p, co, trgt sa the IUI of the
author of the statement, ta the time when the
statement is made, cbs the concept-based
system from which co is taken, p the IUI of the
particular which the author associates with
co, co the concept-code in cbs which the author
associates with p, tr a reference to the time
at which the author considers the association
appropriate,
66
Interpretation of PtoCO statements

• Such statements tell us that within the
  linguistic and scientific community in which cbs
  is used, the terms associated with co may be used
  to denote p

67
A SNOMED-CT example

• ltIUI-0945, 18/04/2005, SNOMED-CT v0301,
  IUI-1921, 367720001, forevergt
• IUI-0945 author of the statement
• IUI-1921 the left testicle of patient
  IUI-78127
• 367720001 the SNOMED concept-code to which left
  testis is (in SNOMED) attached as term
• So we can denote IUI-1921 by means of
• that left testis
• that entire left testis
• that testicle, that male gonad, that testis
• that genital structure
• that physical anatomical entity
• BUT NOT that SNOMED-CT concept

68
Pragmatics of IUIs in EHRs

• IUI assignment requires (just a bit) more effort
  compared to current use of general codes from
  concept-based systems
• A search for concept-codes is replaced by a
  search for the appropriate IUI using exactly the
  same mechanisms
• Browsing
• Code-finder software
• Auto-coding software (CLEF NLP software Andrea
  Setzer)
• With some IUIs there comes a wealth of already
  registered information

69
IUIs in structured EHRs
70
Advantage betterreality representation
IUI-003
71
Other Advantages

• Mappings between ontologies and coding systems
  created as by-product of tracking
• Descriptions about the same particular using
  different systems e.g. in different hospitals
• Quality control of ontologies and concept-based
  systems
• Systematically inconsistent descriptions within
  or across terminologies may indicate poor
  definition of the respective terms

72
Other advantages

• credit card transaction records already
  constitute a referent tracking database
• can give a global picture of economic patterns
  in a given society
• Our proposal will provide for something
  analogous in the realm of healthcare

73
Other Advantages

• Referent tracking can be used in decision support
  when making diagnoses
• We can consider the results of assignment of
  different clinical codes to one and the same
  collection of IUIs assembled over a period (and
  thereby uncover new patterns of symptoms, e.g. in
  a case of multiple sclerosis)

74
Conclusion

• Referent tracking can solve a number of problems
  in an elegant way.
• Existing (or emerging) technologies can be used
  for the implementation.
• Old technologies can play an interesting role.
• Big Brother feeling is to be expected, but with
  adequate measures easy to fight.
• Pilot is being established

75
Generalizing beyond healthcare

• 1) intelligence/security tracking movements of
  people and goods
• 2) tracking copies of papers, music-files, over
  the internet. Numbers can be assigned by
  producers of the files, but also by the people
  who forward them (buying and selling numbers)
• 3) creation of tag-technology for all forms of
  hardware/collectibles
• 4) gaming (turning spam into a game)
  collaborative, distributed story writing
• 5) gambling/play mixture a question is asked,
  and every tenth, twentieth, ... person who calls
  the TV studio is allowed to answer. But calling
  costs you 1. Referent tracking allows this idea
  to be realized over email/internet.