An modular approach to fMRI metadata in a Virtual Laboratory - generic tools for specific problems

1
An modular approach to fMRI metadata in a Virtual
Laboratory-generic tools for specific problems

• M. Scott Marshall, Kasper van den Berg, Kamel
  Boulebiar, Piter de Boer, Marco Roos, Tristan
• Glatard, Silvia Olabarriaga
• Virtual Laboratory for e-science (VL-e)
• University of Amsterdam

2
Outline

• Vision an e-science virtual laboratory
• Everything is a Resource - Explicit Metadata
  Support
• Components AIDA web services
• Platforms Taverna, Web, Vbrowser
• What we did to manage fMRI data

3
Vision Concept-based interfaces

• The scientist should be able to work in terms of
  commonly used concepts.
• The scientist should be able to work in terms of
  personal concepts and hypotheses.
• - Not be forced to map concepts to the terms
  that have been chosen for a given application.

4
What is metadata (in this talk)?

• Metadata data about data
• Metadata can be syntactic such as a data type,
  e.g. Integer.
• Metadata can be semantic such as chromosome
  number.
• Note not always ontology, but metadata can be
  stored in the Web Ontology Language (OWL)

5
Common approaches to metadata

• Code it into the GUI or application (in
  datastructures, object types, etc.)
• Create special tables or fields for it in a
  relational database
• Map it into substrings of filenames
• Mix it in with data in proprietary file formats
• Let the user figure it out
• Conclusion There is a need for semantic
  disclosure.

6
The Semantic Gap
Resources
Middleware
Application
User
7
The Model in the middle
My Model
Model
Model
Resources
Middleware
Application
User
8
RDF a web format for knowledge

• RDF is a W3C language to express
• statements.
• RDF Triple
• Subject Predicate Object
• Graph of Knowledge
• Node Edge Node

9
Adaptive Information Disclosure
(AID)participating in the VL-e project
10
The AIDA toolbox for knowledge extraction and
knowledge management in a Virtual Laboratory for
e-Science
11
Example scenario of Taverna application
12
(No Transcript)
13
Components of the AIDA toolbox used for Life
Science knowledge extraction
14
BioAID Disease Discovery workflow
AIDA
AIDA
Taverna shim
AIDA
OMIM service (Japan)
Taverna shim
15
BioAID Disease Discovery results
16
Enriched ontology (snapshot)
17
Example scenario on Web platform
Looking at custom terminologies, ontologies for
search in personalized index http//aida.science.u
va.nl9999/search/
18
(No Transcript)
19
VBrowser AIDA

• VBrowser provides locators, viewers, access to
  grid storage and transport, a resource-oriented
  interface
• AIDA provides services for search, annotation,
  storage, and metadata extraction

20
VBrowser Resource Overview
21
(No Transcript)
22
MRI more than structural information
anatomical
perfusion MRI
functional MRI
23
Functional MRI (fMRI) What do we do?

• Goal observe brain function during cognitive or
  physical activity.
• Combination of stimulation and imaging.
• Based on the increase in blood flow to the local
  vasculature that accompanies neural activity in
  the brain.

24
fMRI
25
fMRI Paradigms in clinical fMRI

• Motor area
• Language regions (Broca, Wernicke)
• Visual cortex

26
fMRI in ClinicalPreparation of Neurosurgery
27
Neurosurgery Planning
28
Functional MRI Analysis
Stimulus System
Brain activation maps
29
fMRI use case

• Feature Extraction parameter sweeps are performed
  on the fMRI data on the grid.
• The desire is to study the results due to
  different combinations of parameters.
• Each parameter set serves as metadata associated
  with a particular result set location.

30
Metadata for fMRI data search
31
A quick peek at the VBrowser

• A look at fMRI parameters (browsing RDF), RDF
  queries, SRB access
• http//staff.science.uva.nl/ptdeboer/vlet/

32
Acknowledgements

• AIDA team Marco Roos, Sophia Katrenko, Edgar
  Meij, Willem van Hage, Kasper van den Berg
• Vbrowser Piter de Boer
• VL-e Medical Imaging Silvia Olabarriaga, Kamel
  Boulebiar,Tristan Glatard
• Guus Schreiber, Maarten de Rijke, Pieter Adriaans
• Food Informatics partners Wageningen University,
  TNO, Unilever,
• Martijn Schuemie, Erasmus University Rotterdam
• myGrid team, especially Katy Wolstencroft, Stian
  Soiland, Stuart Owen, Andrew Gibson, Alan Rector,
  Robert Stevens, Carole Goble
• Science Commons Alan Ruttenberg
• W3C Semantic Web Health Care and Life Sciences
  Interest Group
• http//adaptivedisclosure.org
• Work supported by VL-e and BioRange projects
  (BSIK grants)