A Prot g g Ontology as The Core Ontology as The Core A Prot - - PowerPoint PPT Presentation

▶
a prot g g ontology as the core ontology as the core a
SMART_READER_LITE
LIVE PREVIEW

A Prot g g Ontology as The Core Ontology as The Core A Prot - - PowerPoint PPT Presentation

Mar 30, 2024 295 likes •781 views

A Prot g g Ontology as The Core Ontology as The Core A Prot Component of a BioSense Component of a BioSense Message Analysis Framework Message Analysis Framework Cecil Lynch 1,2 , Craig Cunningham 1 , Eric Cecil Lynch 1,2 , Craig

slide-1
SLIDE 1

1 1

A Prot A Proté ég gé é Ontology as The Core Ontology as The Core Component of a BioSense Component of a BioSense Message Analysis Framework Message Analysis Framework

Cecil Lynch Cecil Lynch1,2

1,2, Craig Cunningham

, Craig Cunningham1

1, Eric

, Eric Schripsema Schripsema1

1, Tim Morris

, Tim Morris3

3, Barry Rhodes

, Barry Rhodes3

3

1 OntoReason,LLC, 2 UC Davis, 3 US Centers for Disease Control and Prevention

slide-2
SLIDE 2

2 2

Outline Outline

  • BioSense description

BioSense description

  • Describe the current environment

Describe the current environment

  • Describe the ontology

Describe the ontology

  • Describe the ontology framework

Describe the ontology framework

  • Describe the analysis workbench

Describe the analysis workbench

  • Future directions

Future directions

  • Questions

Questions

slide-3
SLIDE 3

3 3

BioSense Description BioSense Description

slide-4
SLIDE 4

4 4

What is BioSense? What is BioSense?

  • Real

Real-

  • time and near real

time and near real-

  • time national public health

time national public health message analysis framework message analysis framework

  • Consists of

Consists of

  • Message acquisition and translation interfaces

Message acquisition and translation interfaces

  • Secure message transmission network

Secure message transmission network

  • Message classification components

Message classification components

  • Data storage and query components

Data storage and query components

  • Data analysis component

Data analysis component

  • CDC Monitors

CDC Monitors

  • Local data visualization and distribution

Local data visualization and distribution

slide-5
SLIDE 5

5 5

BioSense Functions BioSense Functions

Confirm or refute existence of an event Confirm or refute existence of an event

  • Environmental signal

Environmental signal

  • Suspect illness

Suspect illness

  • Intelligence warning

Intelligence warning

  • Known outbreak/public health event

Known outbreak/public health event Monitor ongoing event and effectiveness of response Monitor ongoing event and effectiveness of response

  • Ascertain size of event

Ascertain size of event

  • Ascertain rate of spread

Ascertain rate of spread

  • Track efficacy of response efforts

Track efficacy of response efforts

  • Monitor for adverse events

Monitor for adverse events

  • Know when an event has passed

Know when an event has passed

CDC Slide

slide-6
SLIDE 6

6 6

Data Sources Data Sources

Data Source Data Source 2006 2006 Rationale Rationale

Orders & results from 3 Orders & results from 3 major commercial major commercial clinical laboratories clinical laboratories

Represent 20% of all US lab testing; 60% of Represent 20% of all US lab testing; 60% of independent testing; critical to many PH independent testing; critical to many PH efforts efforts

Real Real-

  • time data from VA

time data from VA

150 hospitals and ~1000 ambulatory care 150 hospitals and ~1000 ambulatory care clinics; share data with many state and local clinics; share data with many state and local PH communities PH communities

Real Real-

  • time data from

time data from DoD DoD

45 US hospitals and ~800 ambulatory; 45 US hospitals and ~800 ambulatory; share data share data

Poison Control Centers Poison Control Centers call data call data All 62 poison control centers; display and All 62 poison control centers; display and compare with other community health compare with other community health data data Private Hospitals Private Hospitals 500 Clinical care Hospitals provide 500 Clinical care Hospitals provide national view and local data national view and local data

CDC Slide

slide-7
SLIDE 7

7 7

Target Data Types Target Data Types

  • Foundational*

Foundational*: : demographics, chief complaint, discharge demographics, chief complaint, discharge diagnoses, disposition, hospital utilization diagnoses, disposition, hospital utilization

  • Clinical*

Clinical*: : vitals, triage notes, working diagnosis, discharge vitals, triage notes, working diagnosis, discharge summary summary

  • Laboratory

Laboratory: : orders, microbiology results

  • rders, microbiology results
  • Pharmacy

Pharmacy: : medication orders medication orders

  • Radiology

Radiology: : orders, interpretation results

  • rders, interpretation results

All structured in HL7 2.5 BioSense messages All structured in HL7 2.5 BioSense messages

CDC Slide

slide-8
SLIDE 8

8 8

Current Classification Current Classification

  • Data mapped to 11 syndrome categories

Data mapped to 11 syndrome categories

  • Botulism

Botulism-

  • like

like

  • Fever

Fever

  • Gastrointestinal

Gastrointestinal

  • Hemorrhagic illness

Hemorrhagic illness

  • Localized cutaneous lesion

Localized cutaneous lesion

  • Lymphadenitis

Lymphadenitis

  • Neurological

Neurological

  • Rash

Rash

  • Respiratory

Respiratory

  • Severe illness/death

Severe illness/death

  • Specific infection

Specific infection

  • 79 sub

79 sub-

  • syndrome categories

syndrome categories

CDC Slide

slide-9
SLIDE 9

9 9

Watch what you ask for! Watch what you ask for!

  • BioSense message volume capacity today

BioSense message volume capacity today

  • 837 messages a second

837 messages a second

  • >72 million messages a day

>72 million messages a day

  • How does an epidemiologist review that

How does an epidemiologist review that volume of data? volume of data?

  • How do you link messages to an individual

How do you link messages to an individual

  • ver time to refine the diagnostic info?
  • ver time to refine the diagnostic info?
slide-10
SLIDE 10

10 10

Current BioSense Framework Current BioSense Framework

slide-11
SLIDE 11

11 11

Message Processing

slide-12
SLIDE 12

12 12

Load Balancing

slide-13
SLIDE 13

13 13

Message Type Filter

slide-14
SLIDE 14

14 14

ETL Processing

slide-15
SLIDE 15

15 15

AV and OTP

slide-16
SLIDE 16

16 16

End User Views

slide-17
SLIDE 17

17 17

The OntoReason PH Ontology The OntoReason PH Ontology

slide-18
SLIDE 18

18 18

Rule Engine Rule Engine Rule Engine Rule Engine

Ontology Model Profile Applications

slide-19
SLIDE 19

19 19

slide-20
SLIDE 20

20 20

slide-21
SLIDE 21

21 21

Information Model Information Model

slide-22
SLIDE 22

22 22

Concept In HL7 V3 Concept In HL7 V3 DataType DataType

Code Term Children Parent Other code Systems and synonyms BioSense Terms

slide-23
SLIDE 23

23 23

Conceptual and Syntactical Conceptual and Syntactical Level Level

slide-24
SLIDE 24

24 24

HL V3 Class Object HL V3 Class Object

References for each object Frequency for each

  • bject
slide-25
SLIDE 25

25 25

Clinical Domain Object Clinical Domain Object

Nested MetaClass

slide-26
SLIDE 26

26 26

slide-27
SLIDE 27

27 27

Laboratory Observation HL7 Laboratory Observation HL7 V3 mapped to V2 V3 mapped to V2

OBX-3 OBX-8 OBX-7 OBX-17 OBX-5 SPM-4

slide-28
SLIDE 28

28 28

Map HL7 Message segments to Map HL7 Message segments to Ontology Slots Ontology Slots

slide-29
SLIDE 29

29 29

Ontology Services Platform Ontology Services Platform

slide-30
SLIDE 30

30 30

Technical Foundations Technical Foundations

Platform Models

  • Enterprise PHIN SOA
  • Web Services
  • Application Libraries
  • LexPHIN Database

Application Models

  • Individual Reasoners Patterns - Languages
  • Intelligence & Analytics Workbench - Tools
  • CTS & LexPHIN Services - Standards

Domain Models Message Structure

  • PH Reference Ontology
  • PHIN VS
  • BioSense Msg HL7 V2.x
slide-31
SLIDE 31

Ontology Extraction Ontology Extraction

  • Creating an application ontology from the reference ontology

Creating an application ontology from the reference ontology

  • Identify the core ontology classes

Identify the core ontology classes

  • Create an object representation that maintains the ontology data

Create an object representation that maintains the ontology data

  • Generate cross reference indexes for core relationships

Generate cross reference indexes for core relationships

  • Lab tests to case investigations

Lab tests to case investigations

  • Organism/Agent to case investigations

Organism/Agent to case investigations

  • Other significant relationships

Other significant relationships

  • Identify

Identify “ “Used Used” ” vocabulary vocabulary

  • Create vocabulary subsets that identify specific vocabularies

Create vocabulary subsets that identify specific vocabularies concepts that are used within the ontology concepts that are used within the ontology

  • Create code to code mapping indexes

Create code to code mapping indexes

  • This produces a general purpose extraction that is suitable for

This produces a general purpose extraction that is suitable for various various purposes purposes

slide-32
SLIDE 32

Ontology Extraction Ontology Extraction

  • Additional activities performed for specific problem solutions

Additional activities performed for specific problem solutions

  • Inclusion of additional vocabulary value sets

Inclusion of additional vocabulary value sets

  • Generation of additional vocabulary indexes to maintain certain

Generation of additional vocabulary indexes to maintain certain parent/child relationships parent/child relationships

  • Incorporation of certain additional term mappings

Incorporation of certain additional term mappings

  • Alternate spellings

Alternate spellings

  • Concept mappings to syndrome/sub

Concept mappings to syndrome/sub-

  • syndrome

syndrome

  • Generation of text search algorithms

Generation of text search algorithms

  • Loadable data married with functional API

Loadable data married with functional API

  • Java object serialized for easy loading

Java object serialized for easy loading

  • Java API providing lookup/query functionality

Java API providing lookup/query functionality

slide-33
SLIDE 33

Ontology Representation Ontology Representation

  • Jess rule engine representation

Jess rule engine representation

  • The Jess rule engine utilizes an enhanced RETE algorithm to

The Jess rule engine utilizes an enhanced RETE algorithm to provide an execution platform for declarative rule base provide an execution platform for declarative rule base

  • Data in Jess is represented as a set of declared facts

Data in Jess is represented as a set of declared facts

  • Facts can be either structured on unstructured

Facts can be either structured on unstructured

  • Ontology data is represented as a set of instance data

Ontology data is represented as a set of instance data represented as structured facts represented as structured facts

  • The ontology can either be expressed as a script or loaded direc

The ontology can either be expressed as a script or loaded directly tly into the rule engine at runtime into the rule engine at runtime

  • Rule definition

Rule definition

  • Rules which describe core case definitions are constructed

Rules which describe core case definitions are constructed

  • The ontology facts are merged with the core set of rules to prov

The ontology facts are merged with the core set of rules to provide ide the base representation for the entire ontology the base representation for the entire ontology

slide-34
SLIDE 34

34 34

Controller Components

Framework Controller Knowledge Registry Knowledge Controller HL/7 Message Factory Data Factory

Configuration Loader / Controller

HL/7 Message Source Operational Data Source

Data Components Reasoning Components

Jess Reasoner Wrapper Reasoner

General Business rules Jess Fact Renderer

Rule Editing

Editor Pattern Template Pattern Configuration Generated Rules

Dashboard Components

Dashboard Event API Standard Dashboard Dashboard Visualization Framework Standard Dashboard Visualization

Analytics Components

Dashboard Visualization Framework Query / Loader Visualization UI Frame External Data Sources Visualization Pane Visualization Pane Visualization Pane

slide-35
SLIDE 35

35 35

Message Analytics Workbench Message Analytics Workbench

slide-36
SLIDE 36

36 36

Model Driven Expert System Model Driven Expert System

  • Public health domain model
  • Highly constrained standardized

vocabulary

  • Clinical reference material
  • Expert knowledge representation
  • Statistical information
  • Empirical evidence

Calculated Knowledge Institutional Knowledge Public Health Reference Ontology Web-Service based application components

  • Reasoning Patterns
  • Platform Descriptions
slide-37
SLIDE 37

37 37

BioSense Message Data Source BioSense Message Data Source

  • HL7 Version 2.5

HL7 Version 2.5

  • XML

XML representation representation

  • Laboratory (ORU)

Laboratory (ORU) message message

  • Spinal fluid protein

Spinal fluid protein

slide-38
SLIDE 38

38 38

Demonstration of Basic Demonstration of Basic Platform Platform

Messages Message Processing Message Classification Patient Msg. Correlation/ Classification Cross Patient Correlation

Knowledge Bus

Cross Patient Correlation Dynamic Syndrome Definition User Profile Data Entry Intelligence & Analytics Dashboard

slide-39
SLIDE 39

39 39

Message Analytical Workbench Message Analytical Workbench

slide-40
SLIDE 40

40 40

slide-41
SLIDE 41

41 41

Reasoner Reasoner results results

slide-42
SLIDE 42

42 42

slide-43
SLIDE 43

43 43

How the Rules Work How the Rules Work

  • JESS template is like a class

JESS template is like a class in Java in Java

  • Template can but does not have to declare

Template can but does not have to declare attribute type, default values, and if an attribute type, default values, and if an attribute is a single value or a list attribute is a single value or a list

  • Facts asserted into the expert system directly

Facts asserted into the expert system directly from the ontology from the ontology -

  • based upon the template

based upon the template

slide-44
SLIDE 44

44 44

Interaction of Ontology And Interaction of Ontology And Rules Rules

  • Information from the ontology and generated

Information from the ontology and generated template facts imported into the expert template facts imported into the expert system and operated on by a variety of system and operated on by a variety of reasoners reasoners

  • This way the

This way the reasoner reasoner knowledgebase can knowledgebase can have a relatively small footprint have a relatively small footprint vs vs the the reference ontology reference ontology

slide-45
SLIDE 45

45 45

JESS Rule From Template JESS Rule From Template

(defrule ClinicalFindingConditionMatchesInitial (Classifier-State DevelopFindingsAndEndorsements) (observation (obsId ?obsId)(msgId ?mId)(patientId ?patientId)(chiefComplaint ?chief) (code ?observation) (obsType ClinicalFinding) (dataQualityFactor ?quality) (dqfReason ?qualityReason) (originationDate ?oDate) (originationDateType ?oType)) (nnd-finding (cond-code ?condCode) (finding-code ?observation)(high ?prob) (ratio ?ratio)) (nnd-condition (cond-code ?condCode) (description ?desc)) (not (BSFinding (msgId ?mId)(finding ?condCode))) => (assert (BSFinding (msgId ?mId)(patientId ?patientId) (findingId ?*ClassId*) (finding ?condCode)(findingType Condition)(findingDesc ?desc) (originationDate ?oDate)(originationDateType ?oType))) (assert (Endorsement (msgId ?mId)(findingId ?*ClassId*)(findingCorrelation ?ratio) (findingType Condition)(finding ?condCode)(findingProb ?prob) (endorsementId (+ ?*ClassId* 1))(endorsement ?*Supportive*)(endSymbol *Support*) (endorsementType *ClinicalFinding*)(rule *ConditionMatch*)(endorsementContext ?context ) (obsId ?obsId)(obsCode ?observation)(obsQuality ?quality)(explaination ?qualityReason ))) (bind ?*ClassId* (+ 2 ?*ClassId*)))

slide-46
SLIDE 46

46 46

A couple of things to A couple of things to remember remember

  • This is an Intelligence and Analytics toolkit

This is an Intelligence and Analytics toolkit

  • Used to exploit the expert knowledge of the organization to prov

Used to exploit the expert knowledge of the organization to provide ide simple to configure application components simple to configure application components

  • Real

Real-

  • time processing

time processing

  • Historical data for analysis, knowledge discovery and re

Historical data for analysis, knowledge discovery and re-

  • classification

classification

  • Findings can be reused to tune and validate real

Findings can be reused to tune and validate real-

  • time processing

time processing

  • Classification tools are based upon a very quick assessment gene

Classification tools are based upon a very quick assessment generalized ralized across all conditions across all conditions

  • The classification weights can be greatly improved based upon em

The classification weights can be greatly improved based upon empirical pirical data analysis data analysis

  • Algorithms are simple to tune and extend (including geo

Algorithms are simple to tune and extend (including geo-

  • spatial and

spatial and temporal services) temporal services)

  • The use

The use-

  • cases were made from some limited set of assumptions

cases were made from some limited set of assumptions

  • We used a condition centric analysis

We used a condition centric analysis

slide-47
SLIDE 47

47 47

Next Steps Next Steps

  • Add additional domain centric rules for better agent

Add additional domain centric rules for better agent classification classification

  • Overcome limitations of ontology size and

Overcome limitations of ontology size and maintenance issues by subdividing into smaller maintenance issues by subdividing into smaller

  • ntologies
  • ntologies
  • Apply a novel technique to use the best aspects of

Apply a novel technique to use the best aspects of Frames and OWL structures Frames and OWL structures

  • (see the demo)

(see the demo)

  • Develop simple domain expert editing tools for rules

Develop simple domain expert editing tools for rules and knowledge and knowledge

slide-48
SLIDE 48

48 48

Questions and Answers Questions and Answers