The UMLS Semantic Network for Natural Language Processing
Download
1 / 53

- PowerPoint PPT Presentation


  • 199 Views
  • Uploaded on

The UMLS Semantic Network for Natural Language Processing Thomas C. Rindflesch, Ph.D. Lister Hill National Center for Biomedical Communications. Workshop on the Future of the UMLS Semantic Network. Goal. Sophisticated access to online information Supplement document retrieval with:

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about '' - taini


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
Slide1 l.jpg

The UMLS Semantic Network for Natural Language Processing Thomas C. Rindflesch, Ph.D.Lister Hill National Center for Biomedical Communications

Workshop on the Future of the UMLS Semantic Network


Slide2 l.jpg
Goal

  • Sophisticated access to online information

  • Supplement document retrieval with:

    • Information extraction

    • Automatic summarization

    • Question answering

    • Literature-based discovery

  • Central concern of informatics research


Challenge language complexity l.jpg
Challenge: Language Complexity

The average age of participants (approximately 63 years), the predominance of women, and the high prevalence of comorbid conditions (for example, hypertension and cardiovascular disease) reflect typical characteristics of patients with osteoarthritis.


Challenge language complexity4 l.jpg
Challenge: Language Complexity

The average age of participants (approximately 63 years), the predominance of women, and the high prevalence of comorbid conditions (for example, hypertension and cardiovascular disease) reflect typical characteristics of patients with osteoarthritis.

  • Language encodes a lot of information


Natural language processing l.jpg
Natural Language Processing

  • Various approaches

  • Correspond to levels of linguistic expression

    • Words

    • Phrases

    • Relations


Words l.jpg
Words

The average age of participants (approximately 63 years), the predominance of women, and the high prevalence of comorbid conditions (for example, hypertension and cardiovascular disease) reflect typical characteristics of patients with osteoarthritis.


Words7 l.jpg

ageapproximatelyaveragecardiovascularcharacteristicscomorbidconditionsdiseaseexamplehigh

hypertensionosteoarthritisparticipantspatientspredominanceprevalencereflecttypicalwomenyears

Words


Phrases l.jpg
Phrases

The average age of participants (approximately 63 years), the predominance of women, and the high prevalence of comorbid conditions (for example, hypertension and cardiovascular disease) reflect typical characteristics of patients with osteoarthritis.


Phrases9 l.jpg

average ageparticipants approximately 63 years predominancewomenhigh prevalencecomorbid conditions

examplehypertension cardiovascular diseasetypical characteristics patients osteoarthritis

Phrases


Semantic predications l.jpg
Semantic Predications

The average age of participants (approximately 63 years), the predominance of women, and the high prevalence of comorbid conditions (for example, hypertension and cardiovascular disease) reflect typical characteristics of patients with osteoarthritis.


Semantic predications11 l.jpg
Semantic Predications

The average age of participants (approximately 63 years), the predominance of women, and the high prevalence of comorbid conditions (for example, hypertension and cardiovascular disease) reflect typical characteristics of patients with osteoarthritis.


Semantic predications12 l.jpg
Semantic Predications

Cardiovascular DiseasesCO-OCCURS_WITHDegenerative polyarthritis

HypertensionCO-OCCURS_WITHDegenerative polyarthritis


Semantic interpretation l.jpg
Semantic Interpretation

  • Map syntactic structures to structured domain knowledge

    • Concepts

    • Relations

  • Output is semantic predication

    • Arguments and a predicate in a relationship

  • Supports enhanced access to online information


Related research in biomedicine l.jpg
Related Research in Biomedicine

[Friedman, et al.]

  • MedLEE, GENIES

    • Semantic grammar

  • AQUA

    • Definite clause grammar

  • MPLUS

    • Chart parser

  • MEDSYNDIKATE

    • Dependency grammar

[Johnson, Campbell]

[Haug, et al.]

[Hahn, et al.]


Semrep l.jpg
SemRep

  • Interpret semantic predications in Medline

  • Exploit the UMLS

    • Concepts: Metathesaurus

    • Relations: Semantic Network

    • Syntax: SPECIALIST Lexicon

  • Use other resources at NLM

    • MetaMap

    • UMLS Knowledge Source Server


Minimal commitment approach l.jpg
Minimal Commitment Approach

  • Focused processing

    • Syntax

    • Semantics

  • Incremental development

  • Useful results


Slide17 l.jpg

SemRep:System Overview

MedPost

Tagger

Lexical

Look-up

Resolve

Ambiguity

SPECIALIST

Lexicon

Metathesaurus

Parser

MetaMap

Construct

Relation

Semantic

Network

Semantic

Predication

MedicalText


Input l.jpg
Input

The aim of this study was the characterization of the specific effects of alprazolam versus imipramine in the treatment of panic disorder with agoraphobia and the delineation of dose-response and possible plasma level-response relationships.


Slide19 l.jpg

Syntactic Processing

Resolve

Ambiguity

SPECIALIST

Lexicon

MedPost

Tagger

Text

Lexical

Look-up

Parser


Syntactic processing l.jpg
Syntactic Processing

The aim of this study was the characterization of the specific effects NP[ofalprazolam][versus]NP[imipramine]NP[in the treatment]NominalizationNP[of panic disorder]NP[with Agoraphobia]and the delineation of dose-response and possible plasma level-response relationships.


Slide21 l.jpg

MetaMap: Metathesaurus Concepts

MedPost

Tagger

Text

Lexical

Look-up

Resolve

Ambiguity

SPECIALIST

Lexicon

Metathesaurus

Parser

MetaMap


Metamap metathesaurus concepts l.jpg
MetaMap: Metathesaurus Concepts

The aim of this study was the characterization of the specific effects NP[ofAlprazolam][versus]NP[Imipramine]NP[in treatment]NominalizationNP[ofPanic Disorder]NP[with Agoraphobia]and the delineation of dose-response and possible plasma level-response relationships.


Semantic types l.jpg
Semantic Types

The aim of this study was the characterization of the specific effects NP[of phsu][versus]NP[phsu]NP[in treatment]NominalizationNP[of dsyn]NP[with dsyn] and the delineation of dose-response and possible plasma level response relationships.

Pharmacologic Substance

Disease or Syndrome


Slide24 l.jpg

Construct Relation

MedPost

Tagger

MedicalText

Lexical

Look-up

Resolve

Ambiguity

SPECIALIST

Lexicon

Metathesaurus

Parser

MetaMap

Construct

Relation

Semantic

Network

Semantic

Predication


Semantic interpretation25 l.jpg
Semantic Interpretation

  • Indicator rules

    • Establish a link between words and predicates in the Semantic Network

  • Argument identification rules

    • Syntactic constraints

  • Validation of semantic predications

    • Semantic Network


Semantic network predicates l.jpg
Semantic Network Predicates

associated_with

physically

spatially

temporally

conceptually

functionally_related_to

occurs_in

affects

brings_about


Core semrep predicates l.jpg
Core SemRep Predicates

associated_with

physically

spatially

temporally

conceptually

LOCATION_OF

functionally_related_to

CO-OCCURS_WITH

OCCURS_IN

affects

brings_about

TREATS

PREVENTS

CAUSES


Semantic network predication l.jpg
Semantic Network Predication

Occupational Activity

Biologic Function

Health Care Activity

Pathologic Function

Therapeutic or Preventive Procedure

Disease or Syndrome

associated_with

physically

spatially

temporally

conceptually

functionally_related_to

occurs_in

affects

brings_about

treats


Indicator rules overview l.jpg
Indicator Rules: Overview

Item

Semantic Network

Structure

nominalization

TREATS

treatment

Drugs for the treatment of schizophrenia

preposition

in

TREATS

Hemofiltration in digoxin overdose

preposition

in

HAS_LOCATION

Severe infections in both feet

Establish a correspondence between a syntactic item and

a Semantic Network predicate


Semantic types30 l.jpg
Semantic Types

The aim of this study was the characterization of the specific effects NP[of phsu][versus]NP[phsu]NP[in treatment]NominalizationNP[of dsyn]NP[with dsyn] and the delineation of dose-response and possible plasma level response relationships.

Pharmacologic Substance

Disease or Syndrome


Apply indicator rule l.jpg
Apply Indicator Rule

The aim of this study was the characterization of the specific effects NP[of phsu][versus]NP[phsu]NP[in treatment]NominalizationNP[of dsyn]NP[with dsyn] and the delineation of dose-response and possible plasma level response relationships.

TREATS


Argument constraints l.jpg
Argument Constraints

The aim of this study was the characterization of the specific effects NP[of phsu] [versus]NP[phsu]NP[in treatment]NominalizationNP[of dsyn]NP[with dsyn]and the delineation of dose-response and possible plasma level response relationships.

TREATS


Semantic network predication33 l.jpg
Semantic Network Predication

phsu-TREATS-dsyn

medd-TREATS-dsyn

topp-TREATS-dsyn

topp-TREATS-inpo

The aim of this study was the characterization of the specific effects NP[of phsu] [versus]NP[phsu]NP[in treatment]NominalizationNP[of dsyn]NP[with dsyn]and the delineation of dose-response and possible plasma level response relationships.


Match semantic types l.jpg
Match Semantic Types

phsu-TREATS-dsyn

medd-TREATS-dsyn

topp-TREATS-dsyn

topp-TREATS-inpo

The aim of this study was the characterization of the specific effects NP[of phsu] [versus]NP[phsu]NP[in treatment]NominalizationNP[of dsyn]NP[with dsyn] and the delineation of dose-response and possible plasma level response relationships.


Substitute concepts l.jpg
Substitute Concepts

The aim of this study was the characterization of the specific effects NP[of phsu] [versus]NP[Alprazolam]NP[in treatment]NominalizationNP[ofPanic Disorder]NP[with dsyn]and the delineation of dose-response and possible plasma level response relationships.

Alprazolam-TREATS-PanicDisorder


Evaluation l.jpg
Evaluation

  • Developing a test collection

    • 2,000 sentences from MEDLINE

    • Mainly drug therapies

    • TREATS, OCCURS_IN, LOCATION_OF, ISA

  • Preliminary results

    • TREATS: 49% recall, 78% precision

    • ISA: 83% precision


Applications l.jpg
Applications

  • Automatic summarization

    • Marcelo Fiszman

  • Machine translation

    • Halil Kilicoglu

  • Discovery

  • Information extraction in genomics

    • Bisharah Libbus

  • Question answering

    • Dina Demner-Fushman


Semantic medline l.jpg
Semantic Medline

Enhanced Information Management

UMLS

Medline

Semantic Processing

PubMed


Automatic summarization l.jpg
Automatic Summarization

  • PubMed search with query “migraine”

  • Retain 500 most recent citations

  • Process with SemRep

  • Summarize SemRep output

    • Condense list of predications

  • Visualize results (Halil Kilicoglu)

  • Translate summarized results (using MeSH)


Summarization for discovery l.jpg
Summarization for Discovery

  • Investigate “unexpected” connections

  • PubMed search with

    • sleep AND gastrointestinal

  • Run SemRep and summarize on

    • Gastroesophageal reflux disease


Summarization for discovery44 l.jpg
Summarization for Discovery

  • Investigate “unexpected” connections

  • PubMed search with

    • sleep AND gastrointestinal

  • Run SemRep and summarize on

    • Gastroesophageal reflux disease

  • New PubMed search on “cpap AND gerd”



Slide47 l.jpg

Marked improvement in nocturnal gastroesophageal reflux in a large cohort of patients with obstructive sleep apnea treated with continuous positive airway pressure


Semantic network for nlp l.jpg
Semantic Network for NLP large cohort of patients with obstructive sleep apnea treated with continuous positive airway pressure

  • Very useful as is

  • Issues noted while developing SemRep

    • Missing relations

    • Infelicitous relations

    • Semantic type hierarchy

  • Recommendations for development

    • Theory and practice

    • Incremental development

    • Maintenance


Missing relations l.jpg
Missing Relations large cohort of patients with obstructive sleep apnea treated with continuous positive airway pressure

  • Genomics

    • “Genes” CAUSE “Disease”

    • “Genes” INTERACT_WITH “Genes”

  • Treatment

    • “Intervention” TREAT “Patients” / “Organism”

      Donepezil for patients with Alzheimer’s


Semantic type hierarchy l.jpg
Semantic Type Hierarchy large cohort of patients with obstructive sleep apnea treated with continuous positive airway pressure

  • Groups and group members (organisms)

    • dsyn OCCURS_IN podg

      • Adults with acidosis

    • dsyn PROCESS_OF mamm

      • Dogs with acidosis


Infelicitous relations l.jpg
Infelicitous Relations large cohort of patients with obstructive sleep apnea treated with continuous positive airway pressure

  • OCCURS_IN

    • Dementia OCCURS_IN Adults

    • Dementia OCCURS_IN Alzheimer’s disease


Recommendations l.jpg
Recommendations large cohort of patients with obstructive sleep apnea treated with continuous positive airway pressure

  • Create an “ontologically-oriented” Semantic Network

  • Developed incrementally

  • Loosely tied to the normal UMLS production cycle


Conclusion l.jpg
Conclusion large cohort of patients with obstructive sleep apnea treated with continuous positive airway pressure

  • The Semantic Network (with the Metathesaurus and Lexicon)

  • Supports semantic interpretation of medical text

  • Enables innovative applications in cooperation with PubMed

  • Provides enhanced access to information in Medline


ad