Upload
gin
View
27
Download
0
Tags:
Embed Size (px)
DESCRIPTION
Research Issues Related to the Construction and Use of Advanced Controlled Medical Terminologies. James J. Cimino, M.D. Department of Medical Informatics September 12, 2000. The Challenge. Build a central, multipurpose clinical data repository with coded data - PowerPoint PPT Presentation
Citation preview
Research Issues Related to the Construction and Use of
Advanced Controlled Medical Terminologies
James J. Cimino, M.D.
Department of Medical Informatics
September 12, 2000
The Challenge
Build a central, multipurpose clinical data repository with coded data
Contributing systems have different coding systems
These coding systems change over time
There are no satisfactory standards
Solution: a Central Terminology Repository
K#1 = 4.2K#1 = 3.3
K#2 = 3.2
K#1 = 3.0
K#3 = 2.6
Additional Challenge:Communication of Changes
K#1
K#2 K#3
K#1 = 4.2K#1 = 3.3
K#2 = 3.2
K#1 = 3.0
K#3 = 2.6
Solution: Hierarchical Integration
K#1
K#2 K#3
K
Seeking an Elegant Solution
The DXplain experience
The UMLS experience
The Theory:
"A knowledge-based approach to
vocabulary representation will improve
maintenance and utility."
The Medical Entities Dictionary (MED)
Multiple hierarchy Synonyms Translations Semantic links Attributes Frame-based 65,000 concepts
MED Structure
MedicalEntity
LaboratoryProcedure
CHEM-7PlasmaGlucose
LaboratorySpecimen
PlasmaSpecimen
Substance
Sampled
Part of
Has S
pecimen
Substance Measured
Event
LaboratoryTest
DiagnosticProcedure
Plasma
AnatomicSubstance
Substance
BioactiveSubstance
Glucose
Chemical
Carbo-hydrate
"A knowledge-based approach to
vocabulary representation will improve
maintenance and utility."
The Theory:
"A knowledge-based approach to
vocabulary representation will improve
maintenance and utility."
"A knowledge-based approach to
vocabulary representation will improve
maintenance and utility."
Maintenance Tasks
New Vocabularies (Laboratory)
Changing Vocabularies (Pharmacy)
New Vocabulary: Laboratory
Original lab: 2533 terms
New lab: 5291 terms
Vocabulary delivered: June 15, 1994
“Go live” date: July 24, 1994
Changing Vocabulary: Pharmacy
Started with 2091 drugs
In two years, added 1827 drugs
Classification by:– Ingredients– AHFS Class– Allergy– DEA– Form
Adding New Terms
Identify redundant terms
Put new terms into existing classes
Create new classes where appropriate
Put Terms into Existing Classes
Theory: The attributes of new terms can be used to identify classes
Practice: "Pushing" Terms
“Pushing” a TermMedicalEntity
LaboratoryTest
PlasmaGlucose
Test
BioactiveSubstance
Glucose
Carbo-hydrate
ChemistryTest
Chem-7Glucose
Test
Chem-20Glucose
Test
StatGlucose
Test
Chemical
“Pushing” a TermMedicalEntity
LaboratoryTest
PlasmaGlucose
Test
BioactiveSubstance
Glucose
Carbo-hydrate
ChemistryTest
Chem-7Glucose
Test
Chem-20Glucose
Test
StatGlucose
Test
StatGlucose
Test
Chemical
“Pushing” a TermMedicalEntity
LaboratoryTest
PlasmaGlucose
Test
BioactiveSubstance
Glucose
Carbo-hydrate
ChemistryTest
Chem-7Glucose
TestStat
GlucoseTest
Chem-20Glucose
Test
StatGlucose
Test
StatGlucose
Test
Chemical
Create New Classes
Theory: Attribute patterns can be detected which identify potential classes
Practice: Recursive partitioning of existing classes
Finding a New Class
MedicalEntity
LaboratoryTest
Hepatitis BCore Antigen
Chemical
ChemistryTest
CoreAntigenHBC
Antigen
Finding a New Class
MedicalEntity
LaboratoryTest
Hepatitis BCore Antigen
Chemical
ChemistryTest
CoreAntigen
Hepatitis BCore Antigen
Test
HBC
Antigen
MedicalEntity
LaboratoryTest
Hepatitis BCore Antigen
Chemical
ChemistryTest
CoreAntigenHBC
Antigen
Semi-Automated Maintenance
Read formulary file
Identify new drugs
Link new drug to ingredient(s)
Suggest classifying in “preparation” class
Add new drug as per human reviewer
Interactive Classification
Adding "LASIX 20MG TAB"Generic Ingredient "FUROSEMIDE"AHFS Class "DIURETICS"Add to "FUROSEMIDE PREPARATION"? y
Adding "ZAROXOLYN 5MG CAP"Generic Ingredient "METOLAZONE"AHFS Class "DIURETICS"Add to "DIURETICS"? nCreate METOLAZONE PREPARATION" Class? y
Automated ClassificationMedicalEntity
Drug
PharmacologicSubstance
Sulfameth-oxizole
Chemical
Antibiotic
Bactrim"S1", "65"
Trimethoprim/Sulfamethoxizole
PreparationsTrimeth-
oprim
Septra"S1"
Sulfa Allergy"S1"
AllergyClass
TrimethoprimAllergy
"65"
Formulary Correction Statistics
Among original 2091 drugs:– 334 unclassified drugs assigned classes– 289 drugs assigned multiple classes– 173 drugs discovered to be missing allergy codes
Among additional 1827 drugs added:– 25 unclassified drugs assigned classes– 121 drugs assigned multiple classes– 38 drugs discovered to be missing allergy codes
Impact of "Theory into Practice":Better management
Easier to merge new vocabularies
Easier to automate change management
Higher quality through better modeling
The Theory:
"A knowledge-based approach to
vocabulary representation will improve
maintenance and utility."
"A knowledge-based approach to
vocabulary representation will improve
maintenance and utility."
Advanced Uses of Coded Data
Primary use Other patient care reuse Financial Management Information transfer (messaging) Clinical research Expert systems Information retrieval Vocabulary discovery
Case Studies
Summary reporting
Case Study:Summary Reporting
Plasma Glucose Test
Serum Glucose TestFingerstick Glucose Test
Lab Test
Intravascular Glucose Test Chem20 Display
Lab Display
DOP Summary
WebCIS Summary
Case Studies
Summary reporting HCFA requirements
Case Study:HCFA Requirements
HCFA won’t pay for lab batteries Individual tests now treated as
orderable procedures Need to appear in database as
procedures and as tests
Lab Procedure
Chem 7
Lab Test
Intravascular Glucose Test
Plasma Glucose Test
Serum Glucose TestFingerstick Glucose Test
Case Study:HCFA Requirements
Lab Procedure
Chem 7
Lab Test
Intravascular Glucose Test
Plasma Glucose Test
Serum Glucose TestFingerstick Glucose Test
Orderable Test
Case Study:HCFA Requirements
Case Studies
Summary reporting HCFA requirements Clinical research
Clinical Research
Epidemiology - symptoms, incidence, history of disease
Outcomes - effectiveness of therapy, ideal length of stay
Recruitment - identifying eligible participants
Case Studies
Summary reporting HCFA requirements Clinical research Expert systems
Case Studies
Summary reporting HCFA requirements Clinical research Expert systems Automated decision support
Terminology and Automated Decision Support
Data monitor checks for triggering conditions
Medical Logic Modules decide if warning conditions are present
Message sent to appropriate channel
Example: Tuberculosis culture result
Decision Support Example: TB
Monitors for delayed culture results
Sends message if result not equal to the code “No growth”
One day, dozens of alerts about positive results but no organism was reported
What happened?
How the Lab Fooled the Alert
Alert looked for results = “No Growth”
Lab started reporting “No Growth to Date” “No Growth to Date” “No Growth”
Solution: Use the controlled terminology to map all No-Growth-like lab terms into a single class, and have the alert logic refer to the class.
How We Outsmarted the Lab(Before)
No Growth
Medical Logic Module
No Growth to Date
No Growth after ...
How We Outsmarted the Lab(After)
No Growth
No Growth after 48 Hours
No Growth after 72 Hours
“No Growth” Results
No Growth after 24 Hours
No Growth to Date
Medical Logic Module
Case Studies
Summary reporting HCFA requirements Clinical research Expert systems Automated decision support Linking to on-line information sources
Linking to On-line Resources with Terminology
Clinician reviewing reports will have information needs
On-line information sources can satisfy that need
Data from report can be used to automate the query
Translations with the MED
Gentamicin
EtiologyMeasures
Sensitivity
Substance Measured
InjectableGentamicin
Gentamicn Sensitivity
Test
SerumGentamicin
Level
is-a
IntravascularGentamicin
Tests
GentamicinToxicity
Has ingredient
SummaryReports
DecisionRule
ExpertSystem
DrugInformation
Linking Text Reports to On-line Information Sources
Natural Language Processing
Data representation to support reuse
Codification of information needs
Impact of Better Management:More Useful Vocabulary
MED is up-to-date for ancillary systems
Easier to find terms in the MED
Support for multiple conceptual levels
More accurate database queries
Support for reuse of clinical data
The MED Today
Concept-based (65,000) Multiple hierarchy (85,000) Synonyms (149,000) Translations (103,000) Semantic links (114,000) Attributes (136,000)
MED Editor/Server ArchitectureMED
MUMPS Globals
Datatree
MUMPS
MED Editor Functions
MED Editor Update Programs
medlog
medlog
MED Server
Query Functions
Command Line
Interface
MED Browser
Translation Tables
Unix
IBMUnix
UnixUnix
Unix Shared Memory Server
ShMMED
MS AccesMED db
MedLEENLP
MEDlib
MLMcompiler
Dxplainbutton
Medlinebutton
MLMcomposition
tool
Integratedresults
design tool
Web MEDbrowser
accessMED
rpc MEDlib
qrymed
MEDbrowser
MEDviewer
Lab uploadextract
Radiologybuploadextract
Bloodbanupload
Integratedresultsreview
ICU resultsdisplay
dop
Data engine
pse
WebCIS
Qrymed Functions-allslts: lists all the slots, with their names.-scd: returns the slotcode with string x as its name.-snm: returns the name of slotcode x.-srecip: returns the reciprocal of slotcode x.-stype: returns the type for slotcode x.-cd: returns the medcode with string x as its name (exact match).-find: lists medcodes that have string x in their names (pattern match).-nm: returns the name for medcode x.-pnm: returns the print name for medcode x.-par: lists the parents of medcode x.-child: lists the children of medcode x.-anc: lists the ancestors of medcode x.-desc: lists the descendants of medcode x.-ianc: lists the ancestors of medcode x (including x).-idesc: lists the descendants of medcode x (including x).-slts: lists the slots of medcode x.-sltsval: lists the slots, with their values, for medcode x.-isval: lists the medcodes which have value y in slotcode x.-val: returns the value(s) of slotcode x for medcode y.
MED in DB2 on Mainframe
1234
Entities
10 Name 20 UMLS 30 Part-of
40 Specimen
Slots
1 102 102 202 30
Entity-Slots
1 10 Entity2 10 C00012 40 12342 50 mg/dl
Entity/Slot/Values1 11 21 32 3
Ancestry
Future Directions
Knowledge management system
User interface
Automated maintenance
Expansion of breadth and depth