Am 18. April fand eine Vortragsreihe unter dem Titel
HealthGRID / SHARE: Grids for Health – International
Interoperability statt, organisiert von der Universität Göttingen
und den Organisationen HealthGrid, (EU) und MediGRID. Sitzungsleiter Yannick
Legre (Pont-de-Chevau) und Ulrich Sax (Göttingen). Alle Vorträge waren auf
Englisch.
|
EU-HealthGrid Initiative –
Current International State of the Art
Yannick Legre, Pont-du-Chateau
Download:
Presentation 510 kB
|
|
Scope of Technology
Otto Rienhoff, Göttingen
|
|
Legal and Ethical Results
from the EU-Share Project
Celine van Doosslaere; Karl Stroetmann, Bonn; Petra Wilson,
Bruxelles
Download:
Presentation 420 kB
Addressing the Ethical, Legal, and
Socio-Economic (ELSE) aspects proves to be critical for HealthGrids
to enter day-to-day working practices in the health sector, thus
bridging the gap between, and improving effectiveness of both
medical research and practice. The storage and processing of
personal data spread on a grid network raises issues such as
privacy, autonomy, control, data protection, and liability that
become significant barriers; just as the lack of (knowledge about)
private incentives and the need for a cultural change towards
collaborative working, with the associated change management and
training requirements. The roadmap for a European Healthgrid, in
development by the SHARE project, will include milestones for
addressing the ELSE aspects.
|
|
Usability of
Solutions from caBIGProject in MediGRID
Th. Steinke; Ulrich Sax, Berlin
Download:
Presentation 3,1 MB
|
|
Interoperability of
Grid-Middlewares
A. Weisbecker, Stuttgart
Download:
Presentation 1,2 MB
|
|
Integrating Portals
for Colaboratories and Grid-Services
Y. Mohammed; S. Rey, Göttingen
Download:
Presentation 189 kB
The
sustainability of the infrastructure depends on how it
becomes a normal element of the scientific process –
including its financing as part of a scientist’s workbench,
which can be easily configured to the needs of the
scientist. The results should be also distributed through a
suitable platform to other research networks. This objective
could be achieved by the integration of several portals of
different research networks and the interfacing between
these portals to maximize the usability, minimize the cost,
and save implementation time and efforts. Our analysis shows
that a three-step procedure should be performed:
-
Firstly: proving and interfacing the security policies
and technologies between the different project/research
networks.
-
Secondly: harmonizing the different portals-access
methods in order to have one GUI.
-
Thirdly: integrating all needed applications and
services as portlets in the suitable portal, where the
needed resources can be accessed.
By building
such an interfacing the different medical research groups
could work together easier and could use the advantages of
the available IT-infrastructure more efficiently.
|
|
Ontological
Application at mediGRID
M. Hartung, E. Rahm Leipzig
Download:
Presentation 2,0 MB
Many advanced grid
applications, especially in the biomedical domain (e.g. the
MediGRID project), need access to complex ontologies
representing knowledge about a certain domain. To deal with
the high heterogeneity of available ontologies, we present a
general service-oriented middleware for making ontologies
accessible to grid applications. Biomedical applications in
the MediGRID portal use the proposed system to interlink
their data with knowledge information of available
ontologies like GeneOntology, NCI Thesaurus,
SequenceOntology and several OBO ontologies.
|
|
Importance of
terminology and communication standards for HealthGrids and
care
Sebastian C. Semler, Berlin
Download:
Presentation 666 kB
For Data Grid applications
in patient care and medical research usage of international
standardized ontologies, terminologies and nomenclatures is
essential. The problem of establish data repositories using
standardized terminologies could not be solved by the Grid
projects. A lot of work and investment to establish the use
of standardized terminologies throughout the life science
field has to be done on different levels. Existing standards
and initiatives must be accounted for.
|
|
Cost Analysis of the
mediGrid Gridinfrastructure
St. Scholz, O. Rienhoff
Download:
Presentation 397 kB
Among other aspects,
sustainable grid infrastructures are based on a
comprehensive understanding of its future cost. Therefore, a
transparent and plausible costing of all participants is
necessary. In addition, hence cost rises while grid-usage
increases, a feasible estimate of the grids capacity
utilization is strongly needed.
|
|
Business Model’s
Analysis
A. Dobrev, Bonn
Download:
Presentation 1,4 MB
The economic case, which
includes non-financial aspects and all stakeholders, for
inter-organisational HealthGrids is demonstrated, yet the
business case, which focuses on financial impact to selected
stakeholders, is still to be developed and proven.
Currently, running HealthGrid initiatives are mainly based
on R&D project funding and focus on computing grids. Meeting
the technical and organisational requirements towards the
next steps - routine operation data and knowledge grids -
can be helped by presenting the economic case to policy
makers, yet a real driver will be only a sound business
case.
|
|
Commonalities of
Engineering Sciences and Medical Applications
M. Assel, S. Roller, Stuttgart
Download:
Presentation 653 kB
Engineering and
Healthcare: Two different research fields with lots of
commonalities, particularly on strong and sophisticated
security mechanisms, regarding Grid-Computing.
|
Speakers Round Table:
Expected Time Frame to Real World Applications in
Research
Download:
Presentation 20 kB (Notes by Fred Viezens)
|
|
Research Projects in
Biomedicine
Th. Lingner, Göttingen
Download:
Presentation 2,7 MB
AUGUSTUS is a gene
prediction software and a prototypical bioinformatics
application. It is easily parallelizable and profits from a
grid environment in many ways. AUGUSTUS is integrated in the
MediGRID port as a pilot application, representing similar
projects from biomedical computer science.
|
Research Projects in Image
Analysis
D. Krefting, Berlin
Download:
Presentation 368 kB
Medical image processing
benefit highly from grid computing. Complex workflows for
medical research can be successfully implemented. Due to
high security demands, further developments and extensions
of existing middleware are required for clinical
applications.
|
|
Research Projects in
Clinical Medicine
Th. Penzel, Berlin; F. Viezens, Göttingen
Download:
Presentation 1,1 MB
For clinical studies, few
applications are suitable for GRID computing. Beside medical
image processing, biological signal analysis of extensive
time series (long-term recordings, more than 8 hours) are
the main target. Such recordings can consist of different
signals such as brain waves, eye movements, muscle tone,
respiration, electrocardiogram, blood oxygen saturation and
others. Pattern recognition, correlation analysis and
derivation of characteristics for disorders and for patients
(also genetic components) need extensive computing. Also
data are distributed at many different university hospitals.
The scenario taken is sleep recordings with 320 sleep
centers in Germany. Data Protection and Data Security are
special issues in clinical research projects and must be
considered. Implementation of Enhanced Security is
compellingly necessarily. Patients use PHR also in the
background of used Grid technology, which is safe, through
access control steered in document level.
|
|
Relevance for Pharma
Research
Martin Hofmann, Bonn
Download:
Presentation 2,7 MB
The pharmaceutical
industry if a data and knowledge - driven industry. The
number of high performance computing applications in pharma
is rather limited (when compared to engineering simulation
applications in e.g. automotive industry). Thus, Grid -
enable solutions for the pharmaceutical industry do not
necessarily have to deal with number crunching, although the
most prominent application of grid technology in pharma is
still the structure-based virtual screening application as
published by Novartis or as done in the course of the WISDOM
project. However, complex information extraction processing
workflows ("text mining") in scientific literature including
patents seem to open a new field for grid-enabled problem
solving in pharma. The PharmaGRID project aims at making the
unstructured information management architecture (UIMA)
framework, a service-oriented architecture for text analysis
services, available as a complex workflow execution
environment for text analysis under UNICORE.
|
zurück
|