Biomedical Informatics in Periodontal Disease Research
Objectives: Common diseases, such as cardiovascular disease, cancer and chronic inflammatory illnesses impose a major drain on society. Like other chronic inflammatory diseases periodontitis has a complex pathogenesis and etiology, involving interactions between multiple genes, microbial and environmental factors. Within the European network INFOBIOMED, periodontitis is a model for investigations into integration of bioinformatics and medical informatics to the new field of biomedical informatics (BMI).
Methods: Data was collected from different sources and a Periodontitis Data Warehouse (PDW) was built taking care of ontodatacleaning, anonymisation, connection with public web-databases and data import of digitally analyzed dental images. Included in the PDW are from periodontitis cases and non-periodontitis controls: intermediate phenotype, genomic, microbial, environmental and phenomic data. Data mining tools Decision Tree Learning (C4.5) and Association Rule Mining (HealthObs) were applied.
Results: Periodontitis has served as a model for BMI. The PDW was constructed and is fully functional. It is located on a dedicated server at the VUmc and is accessible from different research centres with clearance and anonymisation measures. The PDW contains cross-sectional data over 800 patients and controls. The data mining results show that genomic and microbial data together provide the best accuracy in distinguishing patients from controls. Caucasian patients and controls could be classified on basis of a genetic susceptibility index, and sum of the proportions of the 7 microbial species detected in cultivable subgingival microflora. The subjects were denoted as periodontitis patients or healthy when sum of the proportions of the 7 species was >64% or <3.6%, respectively.
Conclusions: The Periodontitis Data Warehouse has been constructed and allows an integrative research approach. Available tools from BMI will contribute to new insights in peridontitis. Similarly the infrastructure and concepts are generally applicable to other chronic inflammatory and infectious diseases. Supported by the INFOBIOMED, 6thR&D Framework, EC (IST2002 507585).
Continental European and Israeli Divisions Meeting
2007 Continental European and Israeli Divisions Meeting (Thessaloniki, Greece) Thessaloniki, Greece
2007 37 Scientific Groups
Laine, Marja Leena
( ACTA, Vrije Universiteit and Universiteit van Amsterdam, Amsterdam, N/A, Netherlands
)
Potamias, G.
( FORTH Hellas, Institute of Computer Science, Heraklion, N/A, Greece
)
Loos, B.g.
( ACTA, Vrije Universiteit and Universiteit van Amsterdam, Amsterdam, N/A, Netherlands
)
Moustakis, V.
( FORTH Hellas, Institute of Computer Science and Technical University of Crete, Heraklion and Chania, N/A, Greece
)
Van Der Palen, Carol
( Vrije Universiteit Medical Centre, Amsterdam, N/A, Netherlands
)
Lessmann, Francis
( Vrije Universiteit Medical Centre, Amsterdam, N/A, Netherlands
)
Van Winkelhoff, Arie Jan
( ACTA, Vrije Universiteit and Universiteit van Amsterdam, Amsterdam, N/A, Netherlands
)
Van Der Velden, U.
( ACTA, Vrije Universiteit and Universiteit van Amsterdam, Amsterdam, N/A, Netherlands
)
Crusius, J.b.a.
( Vrije Universiteit Medical Centre, Amsterdam, N/A, Netherlands
)
Morre, S.a.m.
( Vrije Universiteit Medical Centre, Amsterdam, N/A, Netherlands
)
Pena, A.s.
( Vrije Universiteit Medical Centre, Amsterdam, N/A, Netherlands
)
