Clinical information system for advanced cerebral aneurysm management

1
Clinical information system for advanced cerebral
aneurysm management

• Jimison Iavindrasana,
• Adrien Depeursinge,
• Rodolphe Meyer,
• Stéphane Spahni,
• Antoine Geissbuhler
• and Henning Müller

2
Outline

• Introduction
• Clinical context
• The _at_neurIST project
• Research database architecture
• Objective
• Method
• Results
• Requirements
• Functionality
• Security
• Discussion conclusion

3
Clinical context

• Cerebral aneurysm
• More and more detection of unruptured aneurysm
• Low rupture rate
• Fragmented and rich knowledge
• From molecular, cellular, organ, organism
• Multi-format (image, structured and unstructured
  data)
• Multi-specialties

4
_at_neurIST project

• Provide an IT-infrastructure to integrate
  existing and new knowledge to improve research
  and management of the cerebral aneurysm
• Budget 17M (2006-2009)
• 177 persons
• 32 partners
• 5 clinical centers (patient recruiter, primary
  data provider, research results evaluation)

5
Research database setting

• Centralized vs Distributed architecture
• The clinical information system (CIS) may be used
  as data repository
• Storage of primary and secondary data
• Privacy and confidentiality issues

6
Objective

• Set up an infrastructure to manage the data for
  the _at_neurIST project
• CIS as primary and secondary data repository
• Use existing software components for data
  management and security facilities

7
Methodology

• Design based on
• legal and technical constraints
• Use cases scenarios proposed by the end-users

8
Architecture requirements

• 2 functional cycles
• Information capture and structuring
• Information and enrichment

9
Architecture requirements

• Data categorization (storage rules)
• Patient identification
• Not identical for primary and secondary data
• Reversible pseudonym generation (on-the-fly)
• No personal data for secondary use
• Data communication
• Data normalization on-the-fly
• Data and hospital infrastructure security
• Compatibility with GRID and SOA technology

10
Architecture functionalities
11
Security

• Pseudonymization
• depersonalization pseudonym generation
• Access control
• Local access issued by each institution
• Security token service between institutions
• Incoming query and results verification
• Logging and monitoring

12
Discussion and conclusion

• Main advantages of the infrastructure
• Re-usability
• Scalability
• Low maintenance cost
• Main requirement open CIS

13
Questions?!

• In French or English please!