Transcript Managing Vocabulary for a Centralized Clinical System:

Managing Vocabulary for a
Centralized Clinical System
James J. Cimino, MD; Stephen B. Johnson, PhD;
George Hripcsak, MD; Claire L Hill, MA;
Paul D. Clayton, PhD
Department of Medical Informatics - Columbia University
New York, New York, USA
Managing Vocabulary for a
Centralized Clinical System:
Automated Knowledge-Based Techniques
Putting Theory into Practice
Distributed Systems
Central Clinical Repository
(Fantasy)
Central Clinical Repository
(Reality)
Central Controlled Vocabulary
Vocabulary Construction Issues
• Understanding
• Modeling
• Creation
• Maintenance
The Theory:
"A knowledge-based approach to
vocabulary representation will improve
maintenance and utility."
Medical Entities Dictionary
(MED)
• Semantic network of concepts
• Multiple hierarchies
• Frame-based concept representation
• 46,000 concepts
MED Structure
Medical
Entity
Substance
Chemical
Laboratory
Specimen
Anatomic
Substance
Plasma
Carbohydrate
Bioactive
Substance
Glucose
Plasma
Specimen
Event
Diagnostic
Procedure
Laboratory
Test
Plasma
Glucose
Laboratory
Procedure
CHEM-7
Part of
Retrieving Results Individually
K+1 = 4.2
K+1 = 3.3
K+2 = 3.2
K+1 = 3.0
K+3 = 2.6
K+1
K+2
K+3
Retrieving Results by Class
K+1 = 4.2
K+1 = 3.3
K+2 = 3.2
K+1 = 3.0
K+3 = 2.6
K
K#1
K#2 K#3
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 2327 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 Term
Medical
Entity
Laboratory
Test
Stat
Glucose
Test
Stat
Glucose
Test
Stat
Glucose
Test
Chemistry
Test
Plasma
Glucose
Test
Chem-7
Glucose
Test
Chem-20
Glucose
Test
Chemical
Carbohydrate
Bioactive
Substance
Glucose
Create New Classes
• Theory: Attribute patterns can be
detected which identify potential classes
• Practice: Recursive partitioning of
existing classes
Finding a New Class
Medical
Entity
Medical
Entity
Laboratory
Test
Chemical
Laboratory
Test
Chemical
Chemistry
Test
Antigen
Chemistry
Test
Antigen
Hepatitis B
Core Antigen
Hepatitis B
Core Antigen
Test
Hepatitis B
Core Antigen
HBC
Core
Antigen
HBC
Core
Antigen
screen shot: MED Editor
screen shot: MED Editor
proposing a new class
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"?
Create METOLAZONE PREPARATION" Class?
n
y
Automated Classification
Medical
Entity
Chemical
Drug
Antibiotic
Pharmacologic
Substance
Sulfameth- Trimethoxizole
oprim
Allergy
Class
Sulfa Allergy
"S1"
Trimethoprim/
Sulfamethoxizole
Preparations
Septra
"S1"
Bactrim
"S1", "65"
Trimethoprim
Allergy
"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 2327 drugs added:
– 28 unclassified drugs assigned classes
– 141 drugs assigned multiple classes
– 57 drugs discovered to be missing allergy codes
Future Directions:
Web-Browser
• Platform independence
• Available everywhere
screen shot: Web
Future Directions:
X-Based Browser/Editor
• Runs directly off vocabulary server
• Multi-user environment
• Ready for use in real world
screen shot: Accessmed
Future Directions:
K-Rep
• IBM product
• Knowledge-based approch built in
• Automated term subsumption
• Moving from research to real world
screen shot: K-Rep
Impact of "Theory into Practice":
Better management
• Easier to merge new vocabularies
• Easier to automate change management
• Higher quality through better modeling
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
See for Yourself
http://www.cpmc.columbia.edu/homepages/ciminoj