Holistic data and knowledge mining for health domains

In recent years there has been an explosion in the rate of acquisition of biomedical data including genotype-related data that induced an increasing demand for affective data mining and knowledge extraction. Although many data mining methods have already been developed for biomedical datasets there is no public repository of classification methods specialized for microarray data. To improve the quality of classification performance and beneficial knowledge extraction it is necessary to establish an extensive publicly available repository of microarray datasets. Therefore our goal is to establish an unified platform that would include novel data mining techniques and dataset repository that would provide health care experts with:
(i) data analysis, which will offer an assistance in prediction of diseases and/or early diagnosis
(ii) extraction and comprehensive interpretation of extracted knowledge
(iii) publication of datasets and novel data mining techniques
(iv) possibility of collaboration among healthcare and IT experts
(v) assess the quality of mined knowledge

Our innovative software tool will be able to integrate heterogeneous multimedia information from distributed databases and will be used for coupling scientific research data with clinical and large empirical databases focusing on improving diagnosing, treatment and nursing care.

Scientific and technological objectives:

– integration of different machine learning techniques for data analysis and knowledge extraction
– construct publicly available repository of microarray datasets and other datasets from health domains
– benchmarking of data mining technologies on datasets
– comprehensive interpretation and visualization of extracted knowledge
– encouraging worldwide communication and collaboration among IT and health care experts

Main results by the end of the project:
– Web portal including web service for benchmarking of data mining technologies and datasets
– publicly available repository of microarray datasets and other datasets from health domains
– publicly available repository of data mining techniques for mining datasets from health domains
– novel data mining technologies
– component based platform for modular combination of various data mining technologies
– dissemination of results

data analysis
data mining
knowledge mining
healthcare
biomedical databases
distributed databases

Experts in biomedicine
Experts in microarray experiments
Experts in stardardization of experimental protocols
Access to databases from health domains and healthcare experts for evaluation of extracted knowledge

PROFILE 1: Develop a real case study to test the framework
PROFILE 2: Organization that can perform microarray experiments
PROFILE 3: Offer an open access to databases from healthcare domains
PROFILE 4: Healthcare experts to evaluate extracted knowledge

Hospital
Industrial
Research

Yes