ICBO Tutorial Introduction to Referent Tracking July 22, 2009 112 Norton Hall, UB North Campus

New York State Center of Excellence in Bioinformatics & Life Sciences

R T U

ICBO Tutorial

Introduction to Referent TrackingJuly 22, 2009

112 Norton Hall, UB North Campus

Werner CEUSTERSCenter of Excellence in Bioinformatics and Life Sciences

Ontology Research Group

University at Buffalo, NY, USA


R T U

?

Short personal history

1959 - 20091977

1989

1992

1998

2002

2004

2006

19931995


R T U

House keeping rules

• Feel free to ask clarifications at any time if you don’t understand something I just said (but not more than three slides earlier);

• Please do not interrupt me if you ‘just’ disagree with something I say until:– near beginning of the break,

– near end of the tutorial;

• Everybody in the audience may sleep except those students who are here for credit,– I’ll test them

– redundancy in my slides serves thus a purpose: to help them !


R T U

Tutorial overview

• Setting the scene: a rough description of what Referent Tracking is and why it is important

• Review the basics of BFO relevant to RT

• The crucial distinction between representations and what they represent

• Implementation of RT systems

• Examples of use


R T U

Prologue:

Referent Tracking:What and Why ?


R T U

When did Weiss kill Senator Long ?time

Senator Long’s living

Weiss’ shooting of Long

Carl Weiss’ living

Bodyguards’shooting of Weiss

Weiss’s path. body reactions

Long’s pathological body reactions


R T U

What is Referent Tracking ?

• A paradigm under development since 2005,– based on Basic Formal Ontology,

– designed to keep track of relevant portions of reality and what is believed and communicated about them,

– enabling adequate use of realism-based ontologies, terminologies, thesauri, and vocabularies,

– originally conceived to track particulars on the side of the patient and his environment denoted in his EHR,

– but since then studied in and applied to a variety of domains,

– and now evolving towards tracking absolutely everything, not only particulars, but also universals.


R T U

‘The spectrum of the Health Sciences’

http://www.uvm.edu/~ccts

?Turning data in knowledge


R T U

Source of all data

Reality !


R T U

Ultimate goal of Referent Tracking

A digital copy of the world


R T U

Requirements for this digital copy

• R1: A faithful representation of reality• R2 … of everything that is digitally registered,

what is generic scientific theories

what is specific what individual entities exist and how they relate

• R3: … throughout reality’s entire history,• R4 … which is computable in order to …

… allow queries over the world’s past and present,

… make predictions,

… fill in gaps,

… identify mistakes,

...


R T U

In fact … the ultimate crystal ball


R T U

The ‘binding’ wall

How to do it right ?

I don’t want a cartoon of the world


R T U

Distinction between Ontologies and Information Models

• Ontologies should represent only what is always true about the entities of a domain (whether or not it is known to the person that reports),

• Information models (or data structures) should only represent the artifacts in which information is recorded.– Such information may be incomplete and error-laden

which needs to be accounted for in the information model rather than in the ontology itself.


R T U

Perfect ‘semantic’ tools are useless …

• … if data captured at the source is not of high quality

• Prevailing EHR systems don’t allow data to be stored at acceptable quality level:– No formal distinction between disorders and diagnosis– Messy nature of the notions of ‘problem’ and ‘concern’– No unique identification of the entities about which

data is stored• Unique IDs for data-elements cannot serve as unique IDs for

the entities denoted by these data-elements


R T U

Terminologies for ‘unambiguous representation’ ???

5572 04/07/1990 26442006 closed fracture of shaft of femur

5572 04/07/1990 81134009 Fracture, closed, spiral


5572 12/07/1990 9001224 Accident in public building (supermarket)

5572 04/07/1990 79001 Essential hypertension

0939 24/12/1991 255174002 benign polyp of biliary tract



47804 03/04/1993 58298795 Other lesion on other specified region


298 22/08/1993 2909872 Closed fracture of radial head




PtID Date ObsCode Narrative

0939 20/12/1998 255087006 malignant polyp of biliary tract


R T U

















If two different fracture codes are used in relation to

observations made on the same day for the same patient, do they

denote the same fracture ?



R T U

















If the same fracture code is used for the

same patient on different dates, can

these codes denote the same fracture?



R T U

















Can the same fracture code used in relationto two different patients denote the same fracture?



R T U

















Can two different tumor codes usedin relation to observations made on differentdates for the same patient, denote the same tumor ?



R T U

















Do three references of ‘hypertension’ for the same patient denote three times the same disease?



R T U


















Can the same type of location code used in relation to three different events denote the same location?


R T U

















How will we ever know ?


R T U

The problem in a nutshell

• Generic terms used to denote specific entities do not have enough referential capacity– Usually enough to convey that some specific entity is denoted,

– Not enough to be clear about which one in particular.

• For many ‘important’ entities, unique identifiers are used:– UPS parcels

– Patients in hospitals

– VINs on cars

– …


R T U

1. explicit reference to the concrete individual entities relevant to the accurate description of some portion of reality, ...

Fundamental goals of ‘our’ Referent Tracking

Ceusters W, Smith B. Strategies for Referent Tracking in Electronic Health Records. J Biomed Inform. 2006 Jun;39(3):362-78.


R T U

Method: numbers instead of words

Ceusters W, Smith B. Strategies for Referent Tracking in Electronic Health Records. J Biomed Inform. 2006 Jun;39(3):362-78.

– Introduce an Instance Unique Identifier (IUI) for each relevant particular (individual) entity

78


R T U

2. Use these identifiers in expressions using a language that acknowledges the structure of reality

e.g.: a yellow ball:

#1: the ball #2: #1’s yellow

Then not:

ball(#1) and yellow(#2) and hascolor(#1, #2)

But:

instance-of(#1, ball, since t)

instance-of(#2, yellow, since t)

inheres-in(#1, #2, since t)

Fundamental goals of ‘our’ Referent Tracking

Strong foundationsin realism-based

ontology


R T U

















IUI-001

IUI-001

IUI-001

IUI-003

IUI-004

IUI-004

IUI-005

IUI-005

IUI-005

IUI-007

IUI-007

IUI-007

IUI-002

IUI-012

IUI-006

7 distinct disorders

Codes for ‘types’ AND identifiers for instances


R T U

‘Principles for Success’• Evolutionary change• Radical change:

• Principle 6: Architect Information and Workflow Systems to Accommodate Disruptive Change

» Organizations should architect health care IT for flexibility to support disruptive change rather than to optimize today’s ideas about health care.

• Principle 7: Archive Data for Subsequent Re-interpretation » Vendors of health care IT should provide the capability of

recording any data collected in their measured, uninterpreted, original form, archiving them as long as possible to enable subsequent retrospective views and analyses of those data.NOTE

Willam W. Stead and Herbert S. Lin, editors; Committee on Engaging the Computer Science Research Community in Health Care Informatics; National Research Council. Computational Technology for Effective Health Care: Immediate Steps and Strategic Directions (2009)


R T U

‘Principles for Success’ (continued)

• The NOTE:– ‘See, for example, Werner Ceusters and Barry Smith,

“Strategies for Referent Tracking in Electronic Health Records” Journal of Biomedical Informatics 39(3):362-378, June 2006.’

Willam W. Stead and Herbert S. Lin, editors; Committee on Engaging the Computer Science Research Community in Health Care Informatics; National Research Council. Computational Technology for Effective Health Care: Immediate Steps and Strategic Directions (2009)


R T U

Words, words, words, …

• A paradigm under development since 2005,– based on Basic Formal Ontology,

– designed to keep track of relevant portions of reality and what is believed and communicated about them,

– enabling adequate use of realism-based ontologies, terminologies, thesauri, and vocabularies,

– originally conceived to track particulars on the side of the patient and his environment denoted in his EHR,

– but since then studied in and applied to a variety of domains,

– and now evolving towards tracking absolutely everything, not only particulars, but also universals.


R T U

Therefore:Part 1: the Basics

No (good) Referent Trackingwithout (good) Realism-based Ontology


R T U

1. There is an external reality which is ‘objectively’ the way it is;

2. That reality is accessible to us;

3. We build in our brains cognitive representations of reality;

4. We communicate with others about what is there, and what we believe there is there.

Basic axioms

Smith B, Kusnierczyk W, Schober D, Ceusters W. Towards a Reference Terminology for Ontology Research and Development in the Biomedical Domain. Proceedings of KR-MED 2006, Biomedical Ontology in Action, November 8, 2006, Baltimore MD, USA


R T U

What is there ?

The parts of BFO relevant for Referent Tracking (1)

some particular

some universal

instanceOf …


R T U

The shift envisioned• From:

– ‘this man is a 40 year old patient with a stomach tumor’• To (something like):

– ‘this-1 on which depend this-2 and this-3 has this-4’, where• this-1 instanceOf human being …• this-2 instanceOf age-of-40-years …• this-2 qualityOf this-1 …• this-3 instanceOf patient-role …• this-3 roleOf this-1 …• this-4 instanceOf tumor …• this-4 partOf this-5 …• this-5 instanceOf stomach …• this-5 partOf this-1 …• …


R T U




denotators for particulars


R T U




denotators for appropriate relations


R T U




denotators for universals or particulars


R T U




something I’ll come to later


R T U

Relevance: the way RT-compatible systems ought to interact with representations of generic portions of reality

instance-of at t

#105caused

by


R T U

What is there ?


some particular

some universal

instanceOf …

entities on either site

cannot ‘cross’ this boundary

every particular is

an instance of at least one universal

for every universal

there is or has been at least one instance


R T U

My terminology (1)

• ‘entity’: – denotes either a universal or a particular

• ‘instance’: – denotes a particular to which I refer in the context of

some universal:• If A instanceOf B … then

– ‘B is a universal’

– ‘A is a particular’

– ‘A is an instance’


R T U

My terminology (1)

• ‘entity’: – denotes either a universal or a particular

• ‘instance’: – denotes a particular to which I refer in the context of

some universal:• If A instanceOf B … then

– ‘B is a universal’

– ‘A is a particular’

– ‘A is an instance’

do not denote isa !!!


R T U

My terminology (2)• ‘entity’:

– denotes either a universal or a particular• ‘instance’:

– denotes a particular to which I refer in the context of some universal:

• If A instanceOf B … then– ‘B is a universal’– ‘A is a particular’– ‘A is an instance’

• ‘denotes’: (roughly for now) a relation between an entity and a representational construct (sign, symbol, term,…) such that the latter stands for the former in descriptions about reality.


R T U

What is there ?


some particular

some universal

instanceOf …

?


R T U

instanceOf

What is there ?


some continuantparticular

some continuantuniversal

instanceOf at t

some occurrentparticular

some occurrentuniversal


R T U

instanceOf at t2 instanceOf at t1

instanceOf at t2

The importance of temporal indexing

this-1’s stomach

benigntumor

instanceOf at t1

this-4

malignanttumor

partOf at t1

stomach

partOf at t2


R T U

Use of the CEN Time Standard for HIT


R T UThings do change indeed

child adult

caterpillar butterfly

t

person

animal

Livingcreature

vampire


R T U

The continuants relevant for Referent Tracking

spatial region

independentcontinuant

genericallydependentcontinuant

specificallydependentcontinuant

dependentcontinuant

information content entity

material object

site

ontologyterminology…


R T U

My terminology (3)

• ‘ontology’: – denotes an information artifact whose representational

elements denote universals - either directly or indirectly - and whose structure is intended to mimic the structure of reality

• ‘terminology’: – denotes an information artifact whose representational

elements are terms from some language that are defined in terms of other terms and that are structured independent of the structure of reality


R T UMeSH-2008: give me 666 reasons why this is not an ontology under my terminology.

Wolfram Syndrome

All MeSH Categories

Diseases Category

Nervous System Diseases

Cranial Nerve Diseases

Optic Nerve Diseases

Optic Atrophy

Optic Atrophies,Hereditary

NeurodegenerativeDiseases

HeredodegenerativeDisorders,

Nervous System

Eye Diseases

Eye Diseases, Hereditary


Male UrogenitalDiseases

Urologic Diseases

Kidney Diseases

Diabetes Insipidus

Female Urogenital Diseasesand Pregnancy Complications

Female Urogenital Diseases


R T UWhat would it mean if used in the context of a patient ?

Wolfram Syndrome

All MeSH Categories

Diseases Category

Nervous System Diseases

Cranial Nerve Diseases


Optic Atrophy

Optic Atrophies,Hereditary

has

NeurodegenerativeDiseases

HeredodegenerativeDisorders,

Nervous System

Eye Diseases

Eye Diseases, Hereditary


Female Urogenital Diseasesand Pregnancy Complications

Female Urogenital Diseases

Male UrogenitalDiseases

Urologic Diseases

Kidney Diseases

Diabetes Insipidus

???

…

has

???


R T U

Snomed CT (July 2007):Why not an ontology ?


R T U

Cause: coding / classification confusion

‘A patient with a fractured nasal bone’

‘A patient with a broken nose’

‘A patient with a fracture of the nose’

means the same thing as

means the same thing as

note: doesn’t say what it means


R T U

A patient with a fractured nasal bone

A patient with a broken nose

A patient with a fracture of the nose

=

=

Cause: coding / classification confusion

A patient with a fractured nasal bone

A patient with a broken nose

A patient with a fracture of the nose

=

=


R T U

The occurrents relevant for Referent Tracking

spatiotemporal region

contiguoustemporal

regionhistory

process

time instant

timeinterval

temporal region

scatteredtemporal

region


R T U

Sorts of relations

U1 U2

P1 P2

UtoU: isa, partOf(UU), …

PtoU: instanceOf,

lacks, denotes(PU)…

PtoP: partOf, denotes, …


R T U

t t tinstanceOf

Putting the pieces together: what is there to track?

materialobject

spacetimeregion

me some temporal

region

my life

my 4D STR

some spatialregion

historyspatialregion

temporalregion

dependent continuant

some quality located-in at t

… at t

partic

ipan

tOf a

t t

occupies

projectsOnprojectsOn

at t


R T U

Part 2:Let’s get more serious about

‘representation’(in general)

Beware !!!

Colors don’t really matter

but in what follows

I used them in different ways than before.


R T U

‘Marriage’

marriage ofBill and Hillary

BillClinton Hillary

Clinton

humanbeing

marriage

husbandInspouseIn

husbandOf

spouseOf

instanceOf

instanceOf

instanceOf

createdBy…


R T U

Time and the Bill-Hillary marriage:what about the various some t’s ?


BillClinton Hillary

Clinton

humanbeing

marriage

husbandInspouseIn

husbandOf

spouseOf

instanceOf

instanceOf

instanceOf

createdBy…exists at some t

exists at some t

exists at some t

exists at some t

exists at some t

exists at some t

at some t

at some t

at some t

at some t

at some t at some t

at some t

at some t


R T U

Representation of the Bill-Hillary marriage

‘marriage ofBill and Hillary’

‘BillClinton’ ‘Hillary

Clinton’

‘humanbeing’

‘marriage’

husbandInspouseIn

husbandOf

spouseOf

instanceOf

instanceOf

instanceOf

createdBy…exists at some t

exists at some t

exists at some t

exists at some t

exists at some t

exists at some t

at some t

at some t

at some t

at some t

at some t at some t

at some t

at some t

‘ ,

‘ ,‘ ,


R T U

Representation and what it is about

? at some t


R T U

instanceOf

instanceOf

Representations as first order entities (1)

?2

?1?3

isa

isa

at some t

at some t


R T U

Representations as first order entities (2)

instanceOf

instanceOf

at some t

at some t

ontology

about

aboutL1 R


R T U

Two sorts of representations

L1R

L2 L3

beliefssymbolizations

‘about’


R T U

Three levels of reality


R T U

Diseases : L1 Diagnoses L2/L3

Diagnosis:• A configuration of

representational units;• Believed to mirror the

person’s disease;• Believed to mirror the

disease’s cause;• Refers to the universal

of which the disease is believed to be an instance.

#56John’s

Pneumonia

#78John’s portionof pneumococs

Pneumococcal pneumonia

causedby

Instance-of at t1

Disease

isa


R T U

Some motivations and consequences (1)

• The same diagnosis can be expressed in various forms.

#56#78


causedby

Instance-of at t1

Instance-of at t1

#56 #78

Pneumonia

causedby

Portion of pneumococs

Instance-of at t1

Disease

isa

isa

causedby


R T U


• A diagnosis can be of level 2 or level 3, i.e. either in the mind of a cognitive agent, or in some physical form.

• Allows for a clean interpretation of assertions of the sort ‘these patients have the same diagnosis’: The configuration of representational units is such

that the parts which do not refer to the particulars related to the respective patients, refer to the same portion of reality.


R T U

Distinct but similar diagnoses

#56John’s

Pneumonia

#78John’s portionof pneumococs


causedby

#956Bob’s

pneumonia

#2087Bob’s portion

of pneumococs

causedby

Instance-of at t1 Instance-of at t2


R T U


• Allows evenly clean interpretations for the wealth of ‘modified’ diagnoses:– With respect to the author of the representation:

• ‘nursing diagnosis’, ‘referral diagnosis’

– When created:• ‘post-operative diagnosis’, ‘admitting diagnosis’, ‘final

diagnosis’

– Degree of the belief:• ‘uncertain diagnosis’, ‘preliminary diagnosis’


R T U Important to differentiate betweenLexical, semantic and ontological relations

‘gall’

‘gallbladder’

‘urinarybladder’

‘urine’

‘urinary bladderinflammation’

‘gallbladderinflammation’

‘inflammation’

gall

gall bladderbladder

inflammation

urine

cystitis

biliary cystitis

gallbladderinflammation urinary

bladder


R T U

The three levels applied to diabetes management

1.First-order

reality

2. Beliefs (knowledge)

Generic Specific

DIAGNOSIS

INDICATION

my doctor’swork plan

my doctor’sdiagnosis

MOLECULE

PERSON

DISEASE

PATHOLOGICALSTRUCTURE

PORTION OFINSULIN

DRUG

me

my blood glucose level

my NIDDM

my doctor my doctor’s computer

3. Representation ‘person’ ‘drug’ ‘insulin’ ‘W. Ceusters’ ‘my sugar’


R T U

Referent TrackingBasic Formal Ontology

The three levels applied to C2

1.First-order

reality

2. Beliefs (knowledge)

Generic Specific

GOAL

ATTACK STRATEGY

John Doe’s plan SACEUR’s

strategy

TANK

PERSON

CORPSE

building

SOLDIER

WEAPON

John Doe’s

platoon

Tank with serial numberTH1280A44V

John Doe’s gun

Private John Doe

3. Representation ‘weapon’ ‘person’ ‘tank’ ‘John Doe’ ‘Enola Gay’


R T U

Terminology is too reductionistWhat concepts do we need?

How do we name concepts properly?


R T U

And often confuse L3 with L1

‘Head’ in the NCIT


R T U

The power of realism in ontology design

Reality as benchmark !

1. Is the scientific ‘state of the art’consistent with biomedical reality ?


R T U



2. Is my doctor’s knowledge up to date?


R T U



3. Does my doctor have an accurateassessment of my health status?


R T U


Reality as benchmark !4. Is our terminology rich enough

to communicate about all three levels?


R T U



5. How can we use case studies betterto advance the state of the art?


R T U

Central mechanism in RT: ‘denotation’• Something like a marriage between an L3-entity and an L1-entity


BillClinton

HillaryClinton

husbandIn spouseIn

husbandOf

spouseOf

createdBy…

this particulardenotation

‘This greensquare’

hasReference

referentOf

denotes

denotedBy

createdBy…

referenceOf

hasReferent


R T U

Denotation and time: some axioms



hasReferenceat t1 referentOf

at t1

denotes at t1

denotedBy at t1

createdBy…

S1

referenceOfat t1

hasReferentat t1

• D cannot exist if S or R never existed

• D can continue to exist even when S does not exist anymore

• the existence of R and S are not sufficient for D to exist

• D ceases to exist when R ceases to exist

• …

D

R S


R T U

Denotators with distinct ‘meanings’



hasReferenceat t1

denotes at t1

createdBy at …

S1

hasReferentat t1

A1this otherparticulardenotation

S2

createdBy at …A2


R T U

Changes in reality



hasReferenceat t2

denotes at t2

createdBy at …

S1

hasReferentat t2

A1this otherparticulardenotation

S2

createdBy at …A2

S1

(imagine S1 turned red, yet still being that very same square on the very same spot)

• ‘at’ as defined in CEN:TSHSP

• thus t2 is the ‘coContinuation’ of t1


R T U

Changes in representations

representationOfat t


R T U

Reality and representations

representationOfat t1



R T U

Reality and representations



gain in understanding


R T UChanges in SNOMED


R T U

Reality and representation: both in evolution

IUI-#3

O-#2

O-#1

tU1

U2

p3Reality

Repr.O-#0

= “denotes” = what constitutes the meaning of representational units …. Therefore: O-#0 is meaningless


R T U

Reality versus representations, both in evolution

tU1

U2

p3

IUI-#3

O-#2

O-#1

L1

L2O-#0

Several types of mismatches between reality and representations


R T U

Mistakes, discoveries, being lucky, having bad luck

tU1

U2

p3

IUI-#3

O-#2

O-#1

O-#0

Mistakes

L1

L2


R T U


tU1

U2

p3

IUI-#3

O-#2

O-#1

O-#0

discoveries

L1

L2


R T U


tU1

U2

p3

IUI-#3

O-#2

O-#1

O-#0

L1

L2


R T U

Changes over time

• In John Smith’s Electronic Health Record:– At t1: “male” at t2: “female”

• What are the possibilities ?• Change in reality:

• transgender surgery• change in legal self-identification

• Change in understanding: it was female from the very beginning but interpreted wrongly

• Correction of data entry mistake: it was understood as male, but wrongly transcribed

• (Change in word meaning)


R T U

Part 3:Representation inReferent Tracking


R T U

Representations in ReferentTracking

Portion of Reality

Entity

ParticularUniversal

Defined class

Representation

Non-referringparticular

Denotator

IUI

RT-tuplecorresponds-to

Configuration represents

CUI UUI

denotes

denotes

is about

Representational unit

denotes

contains

class

Extension

……

…

Relation

RUI

denotes

Information content ent.


R T U

Extensions – Defined Classes


R T U

Further distinctions amongst PORs in RT


R T U

Referent Tracking System


R T U

Architecture of aReferent Tracking System (RTS)

• RTS: system in which all statements referring to particulars contain the IUIs for those particulars judged to be relevant.

• Ideally set up as broad as possible:– some metrics:

• % of particulars referred to by means of IUI• % of HCs active in a region

– Geographic region– functional region: defined by contacts amongst patients

• % of patients referred to within a region

• Services:– IUI generator– IUI repository: statements about assignments and reservations– Referent Tracking ‘Database’ (RTDB): index (LSID) to statements relating

instances to instances and classes


R T U

Referent Tracking System Components

• Referent Tracking SoftwareManipulation of assertions about L1

• Referent Tracking Datastore:• IUI repository

A collection of globally unique singular identifiers denoting particulars

• Referent Tracking Database

A collection of assertions about the particulars denoted in the IUI repository

Manzoor S, Ceusters W, Rudnicki R. Implementation of a Referent Tracking System. International Journal of Healthcare Information Systems and Informatics 2007;2(4):41-58.


R T U

Essentials of Referent Tracking • Generation of universally unique identifiers;• deciding what particulars should receive a IUI;• finding out whether or not a particular has already been

assigned a IUI (each particular should receive maximally one IUI);

• using IUIs in the EHR, i.e. issues concerning the syntax and semantics of statements containing IUIs;

• determining the truth values of statements in which IUIs are used;

• correcting errors in the assignment of IUIs.


R T U

Elementary RTS tuple types (1.0)Template Name Abstract Syntax RDFS class

Description A-template Ai = < IUIp, IUIa, tap> ParticularRepresentation Captures the assignment of an IUIp to a particular at time tap by the particular referred to by author IUIa. PtoP – template Ri = <IUIa, ta, r, o, P, tr> PtoP The particular referred to by author IUIa asserts at time ta that the relationship r from ontology o obtains between the particulars referred to in the set of IUIs P at time tr. PtoU-template Ui = <IUIa, ta, inst, o, IUIp, u, tr> PtoU The particular referred to by author IUIa asserts at time ta that the particular referred to by IUIp instantiate inst relation from ontology o with the universal u at time tr. PtoCo-template Coi = <IUIa, ta, cbs, IUIp, co, tr> PtoCo The particular referred to by author IUIa asserts at time ta that the particular referred to by IUIp is annotated by concept code co from terminology system cbs at time tr. PtoU—template U

i = <IUIa, ta, r, o, IUIp, u, tr> PtoLackU The particular referred to by author IUIa asserts at time ta that the relation r of ontology o does not obtain at time tr between the particular referred to by IUIp and any of the instances of the class u at time tr PtoN-template Ni=< IUIa, ta, ntj, ni, IUIp, tr> PtoN The particular referred to by author IUIa asserts at time ta that ni is the name of the nametype ntj assigned to the particular referred to by IUIp at tr. Meta-template Di = <IUId, Xi, td> Publication of a description of a portion of reality in the RTS where IUId is the IUI of the entity registering Xi in the system, Xi is the information-unit in question (in the form of any other template above), and td is a reference to the time the registration was carried out.


R T U

IUI assignment

• = an act carried out by the first ‘cognitive agent’ feeling the need to acknowledge the existence of a particular it has information about by labeling it with a UUID.

• ‘cognitive agent’:– A person;– An organization;– A device or software agent, e.g.

• Bank note printer,• Image analysis software.


R T U

Criteria for IUI assignment (1)

• The particular’s existence must be determined:– Easy for persons in front of you, body parts, ...– Easy for ‘planned acts’: they do not exist before the plan is

executed !• Only the plan exists and possibly the statements made about the

future execution of the plan

– More difficult: subjective symptoms• But the statements the patient makes about them do exist !

– However: • no need to know what the particular exactly is, i.e. which universal

it instantiates• Not always a need to be able to point to it precisely

– One bee out of a particular swarm that stung the patient, one pain out of a series of pain attacks that made the patient worried

– But: this is not a matter of choice, not ‘any’ out of ...


R T U

Criteria for IUI assignment (2)

• May not have already been assigned a IUI.• Morning star and evening star

• Himalaya

• Multiple sclerosis

• It must be relevant to do so:• Personal decision, (scientific) community guideline, ...

• Possibilities offered by the EHR system

• If a IUI has been assigned by somebody, everybody else making statements about the particular should use it


R T U

Assertion of assignments

• IUI assignment is an act of which the execution has to be asserted in the IUI-repository:– Di = <IUId, Ai, td> (1.0)

• IUId IUI of the registering agent

• Ai the assertion of the assignment < IUIp, IUIa, tap>

» IUIa IUI of the author of the assertion

» IUIp IUI of the particular

» tap time of the assignment

• td time of registering Ai in the IUI-repository

• Neither td or tap give any information about when # IUIp started to

exist ! That might be asserted in statements providing information

about # IUIp .


R T U

D-tuples 2.0: dealing with mistakes

Validity and availability of information

Tuple name Attributes Description

D-tuple < IUId, IUIA, td, E, C, S >

The particular referred to by IUId registers the particular referred to by IUIA

(the IUI for the corresponding A-tuple) at time td. E is either the symbol ‘I’

(for insertion) or any of the error type symbols as defined in [1]. C is the reason for inserting the A-tuple. S is a list of IUIs denoting the tuples, if any, that replace the retired one.

A D-tuple is inserted: (1) to resolve mistakes in RTS, and (2) whenever a new tuple other than a D-tuple is inserted in the RTS.

[1] Ceusters W. Dealing with Mistakes in a Referent Tracking System. In: Hornsby KS (eds.) Proceedings of Ontology for the Intelligence Community 2007 (OIC-2007), Columbia MA, 28-29 November 2007;:5-8.


R T U

Types of matches and mismatches


R T U

Management of the IUI-repository

• Adequate safety and security provisions– Access authorisation, control, read/write, ...– Pseudonymisation

• Deletionless but facilities for correcting mistakes.

• Registration of assertion ASAP after IUI assignment

• (virtual, e.g. LSID) central management with adequate search facilities.


R T U

PtoP statements - particular to particular • ordered sextuples of the form Ri = <IUIa, ta, r, o, P, tr>

IUIa is the IUI of the author of the statement,

ta a reference to the time when the statement is made,

r a reference to a relationship (available in o) obtaining between the particulars referred to in P,o a reference to the ontology from which r is taken,P an ordered list of IUIs referring to the particulars between which

r obtains, and,tr a reference to the time at which the relationship obtains.

• P contains as much IUIs as required by the arity of r. In most cases, P will be an ordered pair such that r obtains between the particular represented by the first IUI and the one referred to by the second IUI. • As with A statements, these statements must also be accompanied by a meta-statement capturing when the sextuple became available to the referent tracking system.


R T U

PtoU statements – particular to universal

Ui = <IUIa, ta, inst, o, IUIp, u, tr>



inst a reference to an instance relationship available in o obtaining between p and cl,

o a reference to the ontology from which inst and u are taken,

IUIp the IUI referring to the particular whose inst

relationship with u is asserted,u the universal in o to which p enjoys the inst relationship, and,tr a reference to the time at which the relationship obtains.


R T U

PtoN-statements

Ni=< IUIa, ta, ntj, ni, IUIp, tr, IUIc>

• The person referred to by IUIa asserts at time ta

that ni is the name of the nametype ntj that

designates in the context IUIC in the real world the particular referred to by IUIp at tr. This template

will further be referred to as PtoN template.

• Would assert that “Werner” is my first name, and “Ceusters” is my last name.


R T U

U--tuples: “negative findings”

Relation type

Type of Negative Finding Examples %

C1 <p, u> * A particular is not related in a specific way to any instance of a universal at some given time

he denies abdominal pain; no alcohol abuse; no hepatosplenomegaly; he has no children, without any cyanosis

85.4

C2 <p, u> A particular is not the instance of a given class at some given time

which ruled out primary hyperaldosteronism, nontender, in no apparent distress, Romberg sign was absent , no palpable lymph nodes

12.4

C3 <p, p> A particular is not related to another particular in a specific way at some given time

this record is not available to me; it is not the intense edema she had before; he has not identified any association with meals.

2.2

* ‘p’ ranges over particulars, ‘u’ over universals

U

i = <IUIa, ta, r, o, IUIp, u, tr> The particular referred to by IUIa asserts at time ta that the relation r

of ontology o does not obtain at time tr between the particular

referred to by IUIp and any of the instances of the universal u at

time tr


R T U

PtoCO statements: particular to concept code

Coi = <IUIa, ta, cbs, IUIp, co, tr>



cbs a reference to the concept-based system from which co is taken,

IUIp the IUI referring to the particular which the author

associates with co,co the concept-code in cbs which the author associates with p, and,tr a reference to the time at which the author considers the

association appropriate,


R T U

Interpretation of PtoCO statements

• must be interpreted as simple indexes to terms in a dictionary.

• All that such a statement tells us, is that within the linguistic and scientific community in which cbs is used, the terms associated with co may - i.e. are acceptable to - be used to denote p in their determinative version.


R T U

A SNOMED-CT example • <IUI-0945, 18/04/2005, SNOMED-CT v0301, IUI-1921,

367720001, forever>• #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


R T U

Referent Tracking System Environment

Referent Tracking Server (Peers)

Referent Tracking System

Referent Tracking Data Access Server

ExternalInformationSystem

Reasoning Server

Referent Tracking System User Interface(s)

UserUser

Terminology Server

Vocabulary Thesaurus Nomenclature Concept SystemRealism-basedOntology

or ororor

Referent Tracking Data Store

RTS ProxyPeer

RTSServer ProxyPeer

Internal Ontology

IUI Component


R T U

Networks of Referent Tracking systems

RTS ProxyPeer

RTSServer ProxyPeer

Referent Tracking Server C2


…

RTS ProxyPeer

RTSServer ProxyPeer

Referent Tracking Server B2 Referent Tracking Server B3

…

RTS ProxyPeer

RTSServer ProxyPeer

Referent Tracking Server A2


…

Information System A Information System C

Information System B

Referent Tracking Server B1


Referent Tracking System C


Referent Tracking System A

Referent Tracking System B

RTS ProxyPeer

RTSServer ProxyPeer



…

RTS ProxyPeer

RTSServer ProxyPeer

Referent Tracking Server B2 Referent Tracking Server B3

…

RTS ProxyPeer

RTSServer ProxyPeer



…

Information System A Information System C

Information System B

Referent Tracking Server B1


Referent Tracking System C


Referent Tracking System A

Referent Tracking System B


R T U

Data store


R T U

Pragmatics of IUIs in EHRs • IUI assignment requires an additional effort• In principle no difference qua (or just a little bit more) effort

compared to using directly 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 that IUI comes a wealth of already registered information– If for the same patient different IUIs apply, the user must make the

decision which one is the one under scrutiny, or whether it is again a new instance

• A transfert or reference mechanism makes the statements visible through the RTDB


R T U

< PtSession > < PtsInfo m_ PtL astName ="John" m_PtDOB ="01/01/1985 /> < PtVisitInfo m_PtTimeIn ="02/27/2007 02:44 PM"> … < Level1 m_TemplateName ="Fracture - femur" m_TemplateGUID="{13792543 - C66D - 4B47 - A055 - CEA1A0A53C87} > < Item m_Text=”Examination”> …. < Level4 m _TemplateName =”

” >

< Item m_Text=" strength of left foot plantar flexion is 3/5; strength of left foot dorsi flexion is 2/5 ; "

m_GUID="{65B26952 - 81A1 - 4291 - B26F - 344EBFD2B56B}" / > </ Level4 > …… </ Item > </ Level1 > < / PtVisitInfo > < / PtSessi on >

MedtuityEMR Patient’s Encounter Document


R T U

Decomposing EHR Statements into Particulars • Information units in EHR statements

are phrases.• For each phrase, e.g. strength of left

foot plantar flexion is 3/5, a list of templates containing references to defined classes and universals are stored in a database called Terms Configuration Database, describing the correct decomposition

• The decomposition of a phrase is based on our work described elsewhere*.

*Rudnicki R., Ceusters W., Manzoor S and Smith B. What Particulars are Referred to in EHR Data? A Case Study in Integrating Referent Tracking into an Electronic Health Record Application. Accepted for American Medical Informatics Association 2007 Annual Symposium (AMIA 2007) Proceedings, Chicago IL, 10-14 November 2007.

U1: The universal Person

DC1: MMT scale data value 3.

DC2: defined class whose members are a persons’ left foot plantar muscle group

DC3: defined class whose members are the disposition of persons’ right foot plantar muscle groups to attain a certain performance on the heel-rise test

DC4: defined class of persons who perform members of DC5

DC5: defined class whose members are acts of assessing the performance of heel-rise tests.

DC6: defined class whose members are acts of left foot heel test carried out by a person.

U2: clinical encounter


R T U

Decomposing EHR Statements into Particulars

• Middleware component iterates through the XML document to retrieve the phrases.– For each phrase, e.g. strength

of left foot plantar flexion is 3/5, middleware contacts with Terms Configuration Database to retreive the list of templates containing references to defined classes and universals .

U1: The universal Person

DC1: MMT scale data value 3.

DC2: defined class whose members are a persons’ left foot plantar muscle group

DC3: defined class whose members are the disposition of persons’ right foot plantar muscle groups to attain a certain performance on the heel-rise test

DC4: defined class of persons who perform members of DC5

DC5: defined class whose members are acts of assessing the performance of heel-rise tests.

DC6: defined class whose members are acts of left foot heel test carried out by a person.

U2: clinical encounter


R T U

RTS example graph

rts:IUI-1 rts:IUI-1012/01/2006

rts:pton_1

Name

John

12/01/2006

12/01/2006

rts:IUI-3

rts:ptou_4

rts:r//OBO_REL/instance_of

rts:u//FMA/Left+forearm

rts:r//OBO_REL/has_part

rts:ptop_5

12/01/2006

12/01/2006

rts:plist_6

12/01/2006

12/01/2006

12/01/2006

rts:PtoN

rts:PoPrts:PoU

rts:ParticularRepresentation

rdf:type

rts:nt

rts:n

rts:tr

rts:ta

rts:iup

rts:tap

rts:iuiardf:_1

rdf:type

rts:tap

rts:iuiprts:u

rts:inst

rtsta

rtstr

rdf:typerdf:type

rts:PListrdf:type

rts:iuiprts:Prts:ta

rts:tr

rts:r


R T U

Part 4:Applications & Projects


R T U

eyeGENE (June 2008 - …)


R T U

Ontology for Risks Against Patient Safety


R T U

Representing particular adverse event cases

• Is the generic representation of the portion of reality adequate enough for the description of particular cases?

• Example: a patient – born at time t0

– undergoing anti-inflammatory treatment and physiotherapy since t2

– for an arthrosis present since t1

– develops a stomach ulcer at t3. 133


R T U

Anti-inflammatory treatment with ulcer developmentIUI Description of particular Properties

#1 the patient who is treated #1 member_of C1 since t2

#2 #1’s treatment #2 instance_of C3 #2 has_participant #1 since t2

#2 has_agent #3 since t2

#3 the physician responsible for #2 #3 member_of C4 since t2

#4 #1’s arthrosis #4 member_of C5 since t1

#5 #1’s anti-inflammatory treatment #5 part_of #2 #5 member_of C2 since t3

#6 #1’s physiotherapy #6 part_of #2

#7 #1’s stomach #7 member_of C6 since t2

#8 #7’s structure integrity #8 instance_of C8 since t0 #8 inheres_in #7 since t0

#9 #1’s stomach ulcer #9 part_of #7 since t3

#10 coming into existence of #9 #10 has_participant #9 at t3

#11 change brought about by #9 #11 has_agent #9 since t3 #11 has_participant #8

since t3

#11 instance_of C10 (harm) at t3

#12 noticing the presence of #9 #12 has_participant #9 at t3+x #12 has_agent #3 at t3+x

#13 cognitive representation in #3 about #9 #13 is_about #9 since t3+x

134


R T U

Time line and dependencies (1)

• At t0, the patient is born, and since that time, his stomach is part of him and a structure integrity (C8) inheres in it:– #1 instance-of person since t0

– #7 part-of #1 since t0

– #8 instance_of C8 since t0

– #8 inheres_in #7 since t0

#7’s structure integrity#8

#1’s stomach#7

the patient (#1) who is treated#1

t0 t1 t2 t3

structure integrity C8

135


R T U



#1’s stomach#7

#1’s arthrosis#4

the patient who is treated#1

t0 t1 t2 t3


• At t1, the patient acquires arthrosis:– #4 member_of C5 since t1

– #4 inheres_in #1 since t1

underlying disease C5

136


R T U



#1’s stomach#7

#1’s arthrosis#4

the physician responsible for #2#3

#1’s physiotherapy#6

#1’s anti-inflammatory treatment #5

#1’s treatment#2


t0 t1 t2 t3

subject of care C1

involved structure C6

care giver C4

act of care C3



• At t2, the patient consults #3 who starts treatment. It is then that the patient becomes a member of the class subject of care (C1) and his stomach a member of the class involved structure (C6)

137


R T U

Time line and dependencies …

act under scrutiny C2

cognitive representation in #3 about #9#13

noticing #9#12

#1’s stomach ulcer#9


#1’s stomach#7

#1’s arthrosis#4

the physician responsible for #2#3

#1’s physiotherapy#6

#1’s anti-inflammatory treatment #5

#1’s treatment#2


t0 t1 t2 t3

subject of care C1

involved structure C6

care giver C4

act of care C3



change brought about by #9#11 harmC10

138


R T U

Domotics and RFID systems

• Avoiding adverse events in a hospital because of insufficient day/night illumination:– Light sensors and motion detectors in rooms and corridors

• and representations thereof in an Adverse Event Management System (AEMS)

– What are ‘sufficient’ illumination levels for specific sites is expressed in defined classes,

– Each change in a detector is registered in real time in the AEMS,

– Action-logic implemented in a rule-base system, f.i. to generate alerts.


R T U

RT-based representation (1): IUI assignmentReality level 1 #1: that

corridor

#3: that motion detector#4: that light detector

#2: that lamp

#6: that patient withRFID #7

#5: that RFID reader

#8: that RFID reader#9: this elevator#10: 2nd floor of clinic B


R T U

RT-based representation (2): relationships• (Semi-)stable relationships:

– #1 instance-of ReM:Corridor since t1– #2 instance-of ReM:Lamp since t2– #2 contained-in #1 since t3– #6 member-of ReM:Patient since t4– #6 adjacent-to #7 since t4– #18 instance-of ReM:Illumination since t1– #18 inheres-in #1 since t1– …

• Semi-stable because of: – lamps may be replaced– persons are not patients all the time– …

keeping track of these changes provides a history for each tracked entity


R T U

RT-based representation (3): rule base *• Setting illumination requirements for lamp #2:

– #18 member-of ReM:Insufficient illumination during ty

• if – tx part-of ReM:Daytime– #y1 instance-of ReM:Motion-detection– #y1 has-agent #3 at ty

– ty part-of tx

– #y2 instance-of ReM:Illumination measurement– #y2 has-agent #4 at ty

– #y2 has-participant #18 at ty

– #y2 has-result imrz at ty

– imrz less-than 30 lumen• else

– tx part-of ReM:Night time– …

• endif* Exact format to be discussed with ReMINE partners


R T U

RT-based representation of events• Imagine #6 (with RFID #7) walking through #1

– #2345 instance-of ReM:Motion-detection– #2345 has-agent #3 at t4– #2346 instance-of ReM:RFID-detection– #2346 has-agent #5 at t4– #2346 has-participant #7 at t4– …

• Here, the happening of #2345 fires the rule explained on the previous slide.

• If imrz turns out to be too low, that might invoke another rule which sends an alert to the ward that lamp #2 might be broken.

• #2346 might trigger yet another rule, namely an alert for imminent danger for AE with respect to patient #6

• …


R T U

Making existing EHR systems RT compatible


R T U

Tracking versions of representations


R T UWays representational units do or do not refer

Reality Under-

standing Encoding

OE ORV BE BRV Int. Ref.

G E

P+1 Y Y Y Y Y R+ G1 0

A+1 N - N - - - G2 0

A+2 Y N Y N - - G3 0

P-1 N - Y Y Y R - 3

P-2 N - Y Y N R G4 4

P-3 N - Y Y N R- G5 5

P-4 Y Y Y Y N R G4 1

P-5 Y Y Y Y N R- G5 2

P-6 Y N Y Y Y R+ G1 1

P-7 Y N Y Y N R G4 2

P-8 Y N Y Y N R- G5 3

A-1 Y Y Y N - - G3 1

A-2 Y Y N - - - G2 1

A-3 N - Y N - - G3 1

A-4 Y N N - - - G2 1

OE: objective existence; ORV: objective relevance; BE: belief in existence; BRV: belief in relevance; Int.: intended encoding; Ref.: manner in which the expression refers; G: typology which results when the factor of external reality is ignored. E: number of errors when measured against the benchmark of reality. P/A: presence/absence of term.


R T U

Revisioning beliefs


R T U

Comparing terminologies with reality as benchmark

mn

en

ii

45

)5(1


R T U

Comparing ontology versions

Ceusters W. Applying Evolutionary Terminology Auditing to the Gene Ontology. Journal of Biomedical Informatics 2009;42:518–529.


R T U

Quality evolution of the Gene Ontology


R T U

Quality forecasting


R T U

Referent Tracking enabled Websites


R T U


R T U

Architecture

Web Browser

Web Server

Oracle DB

Zend Framework

RTS

MiddleWare

RtuWebStat

IndexController

Web Pages stored by version

Index.phtml

main.phtml

Index.php


R T U

Some central ideas

1. Informative websites are about portions of reality. If the latter change, so should the former.

2. Synchronization should be auditable.

3. Enforce responsibility of information providers and consumers, yet protect their integrity.

4. Cross-fertilization with Information Artifact Ontology.


R T U

Some key insights

• Static versus dynamic pages;• Web pages usually keep their name (URL), yet undergo

changes;– ‘page’ versus ‘file’

– Server file never ‘changes’: always replaced by a new file with the same name

• Changes to a file do not always involve changes to the propositional content;

• Requests to view a page do not lead the file on the server to be transmitted, but a new copy of it in each single case;


R T U

Entities to assign IUIs to

• The content file of each page

• The content of each content file

• The propositional content of each content

• Each browser page

• Each checksum

• Each ontology and terminology used in RT-tuples

• Each RT-tuple (except D-tuples)

• The middleware component


R T U

Use of the CEN Time Standard for HIT


R T U

Tuple generations when adding a page


R T U

Tuple generations when updating a page


R T U

Tuple insertions: generating a browser pageA-tuples

n IUIp IUIa tap Key

1 #24 #2 (EVENT("#24 assignment") has-occ AT TP(time-18)) #25



D-tuples

n IUId IUIA td E C S Key

2 #2 #25 (EVENT("#25 inserted") has-occ AT TP(time-19)) I CE #26






PtoP-tuples

n IUIa ta r IUIo P tr Key

5 #2 (EVENT("#30 is asserted") has-occ AT TP(time-22)) MainContentCopyOf #022 #27, #12 (EPISODE("#30 is true") has-occ SINCE TI(time-20)) #30

7 #2 (EVENT("#32 is asserted") has-occ AT TP(time-24)) InstigatorOf #022 #24, #27 (EVENT ("#32 is true") has-occ AT TP(time-18)) #32

11 #2 (EVENT("#37 is asserted") has-occ AT TP(time-28)) ChecksumOf #022 #34, #27 (EPISODE("#37 is true") has-occ SINCE TI(time-26)) #37

ICBO Tutorial Introduction to Referent Tracking July 22, 2009 112 Norton Hall, UB North Campus

Documents

Transcript of ICBO Tutorial Introduction to Referent Tracking July 22, 2009 112 Norton Hall, UB North Campus