caTissue CoreERNE integration

1
caTissue Core/ERNE integration

• Martin Ferguson Ian Fore
• 1 Nov 2005
• EDRN Informatics meeting

2
Sources

• Acknowledgements
• Dan Crichton
• Sean Kelly
• Mark Thornquist
• Rakesh Nagarajan

3
Rationale

• UCSF needs
• A biospecimen management system
• To become a node on ERNE
• Proposed Solution
• UCSF adopt caTissue Core
• Expose caTissue as an ERNE node
• Fall out
• Availability of caTissue to other ERNE sites as a
  premapped system

4
ERNE elements
Query Client
Web Client
Clients
Query Server
Central Server
Product Server
Product Server
Product Server
Sites
Query Handler
Query Handler
Query Handler
Relational DB
MS Access
Flat Files
5
caTissue Architecture
6
caTissue - ERNE architecture
Query Client
Web Client
Clients
Query Server
Central Server
Product Server
Product Server
Product Server
Sites
Query Handler
Query Handler
Query Handler
caTissue API
MS Access
Flat Files
Relational DB
7
Query handlers

• Provide the site specific adaptation to in house
  systems
• Hard coded mappings of CDEs
• Mostly written in Java
• Executes query on local database, and receives
  results

8
Query Handlers using an API

• Existing EDRN Query Handlers
• Talk directly to the database
• Typically JDBC or ODBC
• Also flatfiles.
• No examples of query via API to the middle tier
  of an n-tier architecture
• JPL has built such Query Handlers on other
  projects
• SOAP, XML-RPC, RMI, REST, CORBA

9
ERNE Common Data Elements

• 131 CDEs
• 50 are clinical annotation concepts
• 10 are protocol, participant, or demographic
• 70 / 130 are specimen specific

10
caTissue coverage of EDRN CDEs
11
Non direct mappings - examples

• EDRN CDE_ID 552 URINE_TIME_PERIOD_CODE
• 1 First morning, 2 8 hour, 3 24 hour,
  7 other
• Will probably require mapping from numeric time
  value in protocol
• Specific vs General CDEs
• Many current caTissue classes are general
• EDRN CDEs often more specific
• caTissue likely amenable to adding specific CDEs

12
Specimen specific CDEs

• Could be added to caTissue Core
• Specimen Event Parameters
• Specimen Review Parameters

13
Non integration issues

• Necessary IRB approvals
• HIPAA compliance
• Review of informed consent for specimens and
  data, and permissible use.
• IP, licensing rights, etc.
• These are not specific to a caTissue
  implementation
• Address these locally at UCSF

14
Summary of work

• Write Query Handler
• JPL developed
• Use caTissue Core API
• Level of effort?
• Map CDEs
• 1 FTE week, not including approvals, reviews,
  etc.
• Skip some of the more difficult ones for
  proof-of-concept
• Other components require further investigation

15
Further investigation

• Clinical Annotation Elements
• Investigate caTissue Clinical Annotation adoption
• Specimen specific CDEs
• Investigate caTissue Core Extensibility