58
A semantic application for Healthcare Peter Scholten

Semantic Application for Healthcare

Embed Size (px)

DESCRIPTION

Semantic application: how to integrate information

Citation preview

Page 1: Semantic Application for Healthcare

A semantic application

for Healthcare

Peter Scholten

Page 2: Semantic Application for Healthcare

How to build a semantic application

• What is the goal of a semantic application.

• Not only focused on known requirements, but also anticipate on unknown…’future’ settings.

Page 3: Semantic Application for Healthcare

Goal of semantic application

• Social medium (twitter, hyves, facebook etc Communication

• Discussion platform (Linkedin..)Business oriented

• Information mediumQuestions like….

Page 4: Semantic Application for Healthcare

Semantic web for Healthcare

What where to find

Page 5: Semantic Application for Healthcare

Benefits of the semantic web

• Finding resources more quickly and easily

• Storing corporate knowledge

• To generate new knowledge• Improve the Clinic’s ability to use patient data for

generating new knowledge to improve future patient care through outcomes-based and longitudinal clinical research.

• Cross sectional data analysis

Page 6: Semantic Application for Healthcare
Page 7: Semantic Application for Healthcare

Problems on internet

• Format• Language

– Homograph: group of words that share the same spelling but have different meanings

– Homonym: group of words that share the same spelling or pronunciation (or both) but have different meanings

– Synonym: different words with identical or at least similar meanings

– Polysemy: the capacity for a word to have multiple meanings

Page 8: Semantic Application for Healthcare

Need for new semantic “functions” for information and knowledge processing

Page 9: Semantic Application for Healthcare

Example Internet is collection documents with data mostly

represented in tabular form with different formats and dimension.

How to integrate information

Page 10: Semantic Application for Healthcare

Health Care Civilians

How to define

Relation care takers and care need

Relation care takers and care need depending living place

Relation care takers and care need of older people depending living place

Age

Living place

Age

Page 11: Semantic Application for Healthcare

Geographic distribution for care need

Page 12: Semantic Application for Healthcare

Geographic distribution for care need older then 65 years

Page 13: Semantic Application for Healthcare

Relation cardiologist and care takers older then 65 years

Page 14: Semantic Application for Healthcare

Relation family doctor and care takers region Brabant

Page 15: Semantic Application for Healthcare

• Find models on the webRFD/XML files

• Direct access to selected documentsSpecial Google search

• Built a model from scratchSQL versus relational databaseUse of inferencing

Page 16: Semantic Application for Healthcare

• Find models on the webRFD/XML files

• Direct access to selected documentsSpecial Google search

• Built a model from scratchSQL versus relational databaseUse of inferencing

Page 17: Semantic Application for Healthcare

Selected search internet: Demency

Page 18: Semantic Application for Healthcare

• Find models on the webRFD/XML files

• Direct access to selected documentsSpecial Google search

• Built a model from scratchSQL versus relational databaseUse of inferencing

Page 19: Semantic Application for Healthcare

Inferencing

Semantic web !

Page 20: Semantic Application for Healthcare

Example inferencing

x

zy

Page 21: Semantic Application for Healthcare
Page 22: Semantic Application for Healthcare

An Ontology• Defines

– a common vocabulary – a shared understanding– re-use of domain knowledge.

• Is an explicit description of a domain:– Concepts (classes, subclasses and superclasses)– properties and attributes of concepts – constraints on properties and attributes– Individuals (often, but not always)

Page 23: Semantic Application for Healthcare

joints

drugs

Page 24: Semantic Application for Healthcare

Health care informationmodel

Page 25: Semantic Application for Healthcare

Health care ontology

Metadata (individuals)

Metadata (individuals)

Metadata (individuals)

Metadata (individuals)

Metadata (individuals)

Page 26: Semantic Application for Healthcare

Define Classes and the Class Hierarchy

Page 27: Semantic Application for Healthcare

Description of domain by RDF

RDF: Resource Description Framework is a data model for representing metadata

(information about Resources = URI) in the World Wide Web.

Page 28: Semantic Application for Healthcare

Protégé: an ontology editor

• RDF• RDFS• OWL• SPARQL

Page 29: Semantic Application for Healthcare
Page 30: Semantic Application for Healthcare

A typical relational database table for books

Page 31: Semantic Application for Healthcare

The rows represent the things you are storing information about

Page 32: Semantic Application for Healthcare

The columns represent the properties or attributes of those things

Page 33: Semantic Application for Healthcare

the book has a title with value "Javascript"

Page 34: Semantic Application for Healthcare

the book has a title with value "Javascript"

subject has a property with object "value" (s,p,o)

This is the essence of RDF: the (s,p,o) triple

Any expression in RDF is a collection of triples

Page 35: Semantic Application for Healthcare

Relations Between Entities

Page 36: Semantic Application for Healthcare

RDF names things with URLs

Create different URLs to name different things

Page 37: Semantic Application for Healthcare

Any RDF can be merged with any other RDF

Page 38: Semantic Application for Healthcare

Storage of RDF’s in an XML document with the tag rdf:RDFThe content of an XML document is a number of descriptions, which use rdf:Description tags.

<rdf:RDFxmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"xmlns:mydomain="http://www.mydomain.org/my-rdf-ns">

<rdf:Description rdf:about="http://www.cit.gu.edu.au/~db"> <mydomain:site-owner

rdf:resource=“#David Billington“/> </rdf:Description></rdf:RDF>

Page 39: Semantic Application for Healthcare

rdfsRDFS is a vocabulary description language, using

– Classes and Properties– Class Hierarchies and Inheritance– Property Hierarchies

OWL/OWL2: A richer ontology language, disjointness, cardinality, characteristics of properties (SymmetricProperty, TransitiveProperty, and inverseOf, FunctionalProperty, InverseFunctional-Property, sameAs.)

Page 40: Semantic Application for Healthcare
Page 41: Semantic Application for Healthcare

Some RDFS inference rules

• (X R Y), (R subPropertyOf Q) (X Q Y)• (X R Y), (R domain C) (X type C)• (X type C), (C subClassOf D) (X type D)

Page 42: Semantic Application for Healthcare

(X type C), (C subClassOf D) (X type D)

Doctor

Surgeon Anaesthesist

rdfs: subClassOfrdfs: subClassOf

Rdf:type

If ?p rdf:type ?SurgeonIf ?Surgeon rdfs: subClassOf ? DoctorThen ?p rdf:type ?Doctor

Page 43: Semantic Application for Healthcare

(X R Y), (R subPropertyOf Q) (X Q Y)

worksFor

freeLancesTo isEmployedBy

rdfs: subPropertyOfrdfs: subPropertyOf

?p

If ?p freeLancesTo ?HospitalIf freeLancesTo rdfs: subPropertyOf worksForThen ?p worksFor ?Hospital

Page 44: Semantic Application for Healthcare

domain range

If P(PROPERTY) rdfs: domain D and x P Y then x rdf: type D

If P(PROPERTY) rdfs: range R and x P Y then y rdf: type R

?Hospital hasSpecialism ?Physician

?Physician hasCompetences ?Competences

Page 45: Semantic Application for Healthcare

Terminology transfer

Physician Specialismequivalent

? Physician rdfs: subClassOf ? Specialism

Page 46: Semantic Application for Healthcare
Page 47: Semantic Application for Healthcare

SPARQL

Page 48: Semantic Application for Healthcare

SPARQL (Query Language for RDF)

SELECT ?hospital ?PhysicianWHERE { ?hospital rdf:value ?distance. ?physician category ?cardiologist.

FILTER (?distance<=40). }

Page 49: Semantic Application for Healthcare
Page 50: Semantic Application for Healthcare

Searching internet

Input: symptomsOutput: Url’s for description symptoms

Page 51: Semantic Application for Healthcare

Searching internet

Input: symptomsOutput: Url’s for description symptoms

Page 52: Semantic Application for Healthcare

Searching internet

Input: diseases or medicineOutput: Url’s for description medicine and diseases

Page 53: Semantic Application for Healthcare

Searching internet

Input: diseases or medicineOutput: Url’s for description medicine and diseases

Page 54: Semantic Application for Healthcare

Searching internet

Input: professional or instituteOutput: address

Page 55: Semantic Application for Healthcare

Searching internet

Input: professional or instituteOutput: address

Page 56: Semantic Application for Healthcare

Searching internet Input: assistive device disabled persons Output: description and Url’s of assistive devices

Page 57: Semantic Application for Healthcare

Searching internet Input: assistive device disabled persons Output: description and Url’s of assistive devices

WhatWhere to find

description

detailed

Page 58: Semantic Application for Healthcare

Searching internet

Input: assistive need for older or disabled persons

• Aids for low-vision or blind persons• Aids for motor disabilities• Persons hard of hearing• Demency• COPD• Chronic diseases• Home care• Emergency service

Output: description and Url’s of assistive advice