Transcript ppt

Intelligent Techniques for Data Integration
and Decision Support in the Medical Domain
Mirjana Ivanović, Hans-Dieter Burkhard
AGENDA
 General information
 Scientific Objectives
 Project Description and Structure
 Expected benefit of project results
 Conclusion
General information
 Bilateral project Germany – Serbia
 DAAD (G) - Ministry of Education and Science (S)
 (Gabriela Lindemann – v. Trzebiatowski) Hans-Dieter
Burkhard, Mirjana Ivanovic
 Duration: 2011-2012 (Application June 2010)
 Joint research, papers, mutual visits
AGENDA
 General information
 Scientific Objectives
 Project Description and Structure
 Expected benefit of project results
 Conclusion
Scientific Objectives
 Data commonly produced during medical practice falls
into the category of complex data:
- text documents, time-series (i.e., sequences of data
points measured at successive times)
- multimedia objects (images, sound, etc.)
 The main objective of this project: investigation and
application of different techniques for reasoning,
mining and retrieval to problems recognized in
medical domain
Scientific Objectives
The main focus of the project will be placed on:
 Investigating how to overcome difficulties in organizing (in
appropriate way) huge amount of different kinds of medical data
(numbers, text, images)
 Incorporating reasoning and mining techniques in implementation of
reliable general-purpose intelligent (DSS)
 Developments of specific-purpose DSS (developed from general one) in
several medical fields (Nephrology, Haematology, Pathology, and
Neurology).
 Usage of Ontology to improve results of more complex data analyses.
AGENDA
 General information
 Scientific Objectives
 Project Description and Structure
 Expected benefit of project results
 Conclusion
Project Description and Structure
 Nowadays, medical imaging, analysis of fluid and tissue
samples and storage and retrieval of patient records
require applications of sophisticated computer
technologies
 Realization of DSS for diagnosis, therapy planning or
integration into a large-scale centralized repository.
 Project will perform activities divided into 5 Working
packages
Project Description and Structure
 WP1. Analysis of the wide spectrum of heterogeneous data
appearing in medical domains in order to identify characteristic ones.
 WP2. Analysis of essential functionalities of different intelligent
techniques and ways of their integration in general-purpose DSS.
 WP3. Proposition of an architecture of a general-purpose DSS and
its implementation.
 WP4. Implementation of 2 special-purpose DSS for application in
German and Serbian Clinics.
 WP5. Development of a prototype with key functionalities for
therapy planning.
AGENDA
 General information
 Scientific Objectives
 Project Description and Structure
 Expected benefit of project results
 Conclusion
Expected benefit of project results
 Survey, containing:
1) Proposition of characteristic medical data extracted from
the wide spectrum of heterogeneous sources.
2) Ways of integration of key intelligent functionalities in a
decision support system.
 General-purpose DSS for medical domains.
 2 special-purpose DSS for particular medical fields.
 A prototype of a therapy planning system.
AGENDA
 General information
 Scientific Objectives
 Project Description and Structure
 Expected benefit of project results
 Conclusion
Conclusion
 Interesting: Selected projects, usually – environmental,
nano technologies, medicine and food production
 Low selection rate
 Significant to make joint research, inclusion of young
researchers
 General expectation: joint applications for bigger EU
projects