James J. Cimino

1
The Challenge of Reuse of Information

• James J. Cimino
• Columbia University
• MIE 02
• Budapest, Hungary
• August 27, 2002

2
Overview

• Data types
• Information reuse
• Information mismatch
• Terminology solutions
• Experience
• Conclusions

3
Overview

• Data types
• Information reuse
• Information mismatch
• Terminology solutions
• Experience
• Conclusions

4
Data Types
5
Overview

• Data types
• Information reuse
• Information mismatch
• Terminology solutions
• Experience
• Conclusions

6
Information Reuse
7
Overview

• Data types
• Information reuse
• Information mismatch
• Terminology solutions
• Experience
• Conclusions

8
Information Mismatch

• Form
• Meaning
• Language
• Granularity
• Semantics
• Version

9
Information Mismatch Form
10
Information Mismatch Form
11
Information Mismatch Meaning
of the bone
12
Information Mismatch Language
13
Information Mismatch Granularity
14
Information Mismatch Semantics
15
Information Mismatch Version
2001 ICD - Smallpox - Cowpox - Virus, NEC
Virus, NEC
16
Overview

• Data types
• Information reuse
• Information mismatch
• Terminology solutions
• Experience
• Conclusions

17
Terminology Solutions

• Standards
• Distribution
• Semantic Representation

18
Terminology Solutions Standards

• Advantages
• Less duplication of work
• Plug and play compatibility
• Disadvantages
• Cost of adoption
• Unresponsive to change
• Developers ? Users

19
Terminology Solutions Distribution

• Media
• 9-track tape
• Floppy disks
• CD-ROM
• Web
• Models
• ICD annual
• UMLS change files
• HL7 server

20
Terminology Solutions Semantic Representation

• Concept oriented
• Concept permanence
• True is-a hierarchies
• Multiple hierarchies (heterarchy)
• Semantic relationships
• Inheritance

21
Terminology Solutions Semantic Representation
Goodpastures Syndrome
22
Semantic Representation Galen

• Structured Meta Knowledge from PenPad
• Common Reference Terminology
• Requires terminology server
• Automated classification
• Open source terminology

23
Semantic Representation Galen

• Fracture which lt
• hasLocation Bone
• hasCause Conditiongt
• Fracture which lt
• hasLocation (AnatomicalNeck which isDivisionOf
  Femur)
• hasCause (Osteoporosis which hasCause
  PostMenopausalChange)gt
• Can be classified as
• Fracture
• Fracture which hasLocation LongBone.
• Fracture which hasLocation (AnatomicalNeck which
  isDivisionOf LongBone).
• Fracture which hasLocation Thigh.
• Fracture which hasLocation Hip.
• Lesion which isCausedBy Osteoporosis.
• Lesion which isCausedBy PostmenopausalChange.

24
Semantic RepresentationSNOMED-CT

• Merger of SNOMED and Read Clinical Terms
• Reference terminology
• Many domains
• Heterarchy
• Semantic relations (roles)
• Postcoordination
• gt300,000 concepts

25
Semantic RepresentationSNOMED-CT
Pulmonary Tularemia
26
Semantic Representation LOINC

• Logical Observations, Identifiers, Names and
  Codes
• Codes for observations in HL7 messages
• Fully-specified names
• Codes for orderable observations
• Codes for results

27
Semantic Representation LOINC
24356-8 URINALYSIS PANEL
28
Semantic Representation Drugs

• Food and Drug Administration
• Veterans Administration
• National Library of Medicine
• Drug knowledge base vendors
• Common model for Clinical Drug
• RxNorm

29
Semantic Representation Drugs
Clinical Drug
30
Semantic Representation MED

• Medical Entities Dictionary
• Data dictionary and controlled terminology
• Columbia-Presbyterian Medical Center
• Heterarchy
• Semantic network
• Multiple domains
• gt70,000 concepts

31
Semantic Representation MED
Plasma Glucose Test
32
Overview

• Data types
• Information reuse
• Information mismatch
• Terminology solutions
• Experience
• Conclusions

33
Matching Granularity and Semantics
Injectable Gentamicin
Serum Gentamicin Level
Gentamicin
Gentamicin Sensitivity Test
Gentamicin Toxicity
34
Example of ReuseSummary Reporting

• Spreadsheets for trends in lab data
• Defined as concepts in the MED
• Linked to test classes

35
Example of ReuseSummary Reporting
36
Example of ReuseSummary Reporting
37
Example of ReuseSummary Reporting
38
Example of Reuse Merging Data

• Merger between Presbyterian Hospital and New York
  Hospital
• Separate departmental systems
• Common repository
• Merger of terms in MED allows cross-institution
  data aggregation

39
Example of Reuse Merging Data
45748 - Diazepam 5 mg Tablet
40
Example of Reuse Merging Data
2478 - Plasma Glucose Measurement
41
Example of ReuseAutomated Decision Support

• Data stored in repository reviewed in real time
• Arden Syntax rules triggered by data
• Generation of alerts and reminders
• High-level concepts in rules map to low-level
  concepts in database

42
Automated Decision Support Tuberculosis

• 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?

43
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.

44
Automated Decision Support Tuberculosis
Medical Logic Module
No Growth to Date
No Growth
45
How We Outsmarted the Lab
Medical Logic Module
No Growth to Date
No Growth
46
Example of ReuseInformation Retrieval
47
(No Transcript)
48
(No Transcript)
49
(No Transcript)
50
Example of Reuse Expert Systems

• Expert system has high-level concepts
• Database has quantitative results
• Semantic mismatch
• Translation through semantic net traversal

51
Example of Reuse Expert Systems
52
Expert System DXplain
53
Expert System DXplain
54
Expert System DXplain
55
Expert System Lipid Guideline
56
Example of ReuseProblem-Oriented Views
Heart
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)
61
Experience with Information Reuse

• Summary reporting
• Merging data
• Automated decision support
• Information retrieval
• Expert systems
• Problem-oriented views

62
Overview

• Data types
• Information reuse
• Information mismatch
• Terminology solutions
• Experience
• Conclusions

63
Information, Then and Now
Uncoded, Unstructured
Then (1993)
- Cimino JJ, Int J Biomed Comput. 1994
34185-194
64
Information, Then and Now
Locally Coded
Uncoded, Unstructured
Universally Coded
Uncoded, Structured
Radiology Reports
Discharge Summaries
Discharge Diagnoses
Physical Exams
Patient Histories
Medication Lists
Near Future
- Cimino JJ, Int J Biomed Comput. 1994
34185-194
65
Information, Then and Now
Locally Coded
Uncoded, Unstructured
Universally Coded
Uncoded, Structured
Radiology Reports
Discharge Summaries
Discharge Diagnoses
Physical Exams
Patient Histories
Medication Lists
Far Future
- Cimino JJ, Int J Biomed Comput. 1994
34185-194
66
Current Status
Locally Coded
Uncoded, Unstructured
Universally Coded
Uncoded, Structured
Radiology Reports
Discharge Summaries
Discharge Diagnoses
Physical Exams
Patient Histories
Medication Lists
67
Current Status
Standard Semantic Terminology
Standard Code Set
Uncoded, Unstructured
Uncoded, Structured
Radiology Reports
Discharge Summaries
Discharge Diagnoses
Physical Exams
Patient Histories
Medication Lists
68
Current Status
Standard Semantic Terminology
Uncoded, Unstructured
Standard Code Set
Uncoded, Structured
Laboratory Reports
Text Reports
Discharge Diagnoses
Problem Lists
Text Reports
Medication Lists
69
Conclusions

• Advanced health care means information reuse
• Semantic-based terminologies support reuse
• Terminologies are moving in the right direction