Format

Send to

Choose Destination
AMIA Annu Symp Proc. 2015 Nov 5;2015:880-5. eCollection 2015.

Towards data integration automation for the French rare disease registry.

Author information

1
Banque Nationale de Données Maladies Rares, Hôpital Necker Enfants Malades, Assistance Publique des Hôpitaux de Paris, Paris, France; INSERM, U1142, LIMICS, Paris, France.
2
Banque Nationale de Données Maladies Rares, Hôpital Necker Enfants Malades, Assistance Publique des Hôpitaux de Paris, Paris, France; Montpellier University, EA2415 & BESPIM, University Hospital Nîmes, France.
3
INSERM, U1142, LIMICS, Paris, France.

Abstract

Building a medical registry upon an existing infrastructure and rooted practices is not an easy task. It is the case for the BNDMR project, the French rare disease registry, that aims to collect administrative and medical data of rare disease patients seen in different hospitals. To avoid duplicating data entry for health professionals, the project plans to deploy connectors with the existing systems to automatically retrieve data. Given the data heterogeneity and the large number of source systems, the automation of connectors creation is required. In this context, we propose a methodology that optimizes the use of existing alignment approaches in the data integration processes. The generated mappings are formalized in exploitable mapping expressions. Following this methodology, a process has been experimented on specific data types of a source system: Boolean and predefined lists. As a result, effectiveness of the used alignment approach has been enhanced and more good mappings have been detected. Nonetheless, further improvements could be done to deal with the semantic issue and process other data types.

PMID:
26958224
PMCID:
PMC4765585
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for PubMed Central
Loading ...
Support Center