Medical Research and Education

Medical Research and Education, or MRE, is a platform for designing and developing information systems for research purposes, particularly with medical data.

The goal is long-term archiving and reusability, focusing on data protection and security.

The Medical Information Systems research group is part of Medical Informatics at the Department of Computer Science and Engineering, Faculty of Applied Sciences, University of West Bohemia, in the Czech Republic. The research group is part of the research centre NTIS - New Technologies for Information Society.

Overview

Data de-identification

Personal privacy is the most critical task. We develop and maintain our applications for making data without sensitive personal data.

  • AnonMed is proper for the de-identification of metadata content in several file formats.
  • PureImage identifies any text content in the pixel data and does its classification. It is also possible to redact parts of images with classified text content.

Both tools can de-identify any identity of a person, equipment, health care facility or staff.

Most importantly, any secondary use of medical data requires removing identity.

Metadata schema

  • Metadata has its schema in conformance with the Resource Description Framework (RDF) and ontologies, including Web Ontology Language (OWL2).
  • See the ontologies list for more information about the schema for data.

Metadata extraction

Platform provides MetaMed for metadata extraction based on ontologies.

DASTA code-lists to OWL/RDF produce RDF data for DASTA code-lists that are regularly published/updated four times yearly.

  • MetaMed application does metadata extraction, transformation, and loading into graph store.
  • Metadata extraction is fully automatic for all incoming/received data files. Metamed is helpful before and after de-identification.
  • New RDF data can produce a web-based HTML5 application, Web Form Generator.

Use metadata

  • SPARQL Endpoint(s)
  • Sparkle is a desktop application where you can build SPARQL queries.
  • Sparkle's Web View is part of Sparkle. It is a web-based JavaScript application that builds SPARQL queries visually in your web browser. You must use your SPARQL Endpoint instead of the default one if you are not a member.

By maximizing the use of ontology, the user interface can be as dynamic as possible. It can be quickly and easily changed/extended or adapted to the new domain of interest or purpose of use.

  • Web Form Generator provides a speedy way to use ontology (OWL) to produce structured and validated RDF data using modern responsive HTML5 web forms.
  • Query Builder web-based application can build SPARQL queries for temporal data.

Services

Only members can use available services directly or using VPN}.

Metadata is available through SPARQL Endpoints with restricted access.