CHIC-CDR: A repository for managing multi-modality clinical data and its application to in-silico oncology
Description
Access to high-quality clinical data is a prerequisite for medical data analysis and in-silico medicine. Data drives the development of research questions, feeds computational algorithms, and provides the evidence base for validating complex disease models which form the basis for personalized simulations in the future. However, provision of uniform access to, and secondary use of, clinical treatment or study data is hampered by the data's intrinsic characteristics: its confidential nature, and its heterogeneity in terms of sources, quality and information.
The clinical data repository (CDR) has been designed to address these critical issues by providing a unique access-point to clinical data in compliance with the European medico-legal framework. CDR has been initiated as an exchange platform for medical images, but was quickly extended to other health-related data, in particular clinical treatment & follow-up, histological and genetic information. Data access is granted based on a role-based policy within a single-sign-on security framework. CDR also supports the (pseudo-)anonymization process and provides facilities for semantic annotation of data during and after the upload process, enabling semantically mediated queries for improved data discovery.
CDR has been developed by SICAS (http://www.si-cas.com) and the University of Bern (http://www.istb.unibe.ch/) with support from the EU CHIC (http://chic-vph.eu/) and Co-Me (http://co-me.ch) projects.
This presentation outlines the principal requirements and main challenges related to the sharing of heterogeneous clinical data for research purposes, discusses properties of a generic solution, and details implementation and application in the CHIC context.
Files
CHIC-CDR.pdf
Files
(5.4 MB)
Name | Size | Download all |
---|---|---|
md5:11dc71aed4ff148c22b6a4d93fdfabc6
|
5.4 MB | Preview Download |