Medical Information Platform

Motivation

In a distributed research network it is very important to maintain a central work platform in order to integrate partners as well as necessary applications and databases. By constructing a modern information and communication portal, central services and information will be made available to diverse target groups (including network partners, external experts and practical specialists). Such a platform has been implemented, for example, for malignant lymphoma, where the up-to-dateness and quality assurance of the contents play a central role.

Objective

The main objective is the to present the research network on the internet in a professional and sustainable manner. Since researchers in a medical network collaborate across fields and centers, the information platform must offer intuitive operability for users, possess secure channels of communication, support standardized data formats and allow for the integration of software applications into a unified platform.

Methods and Implementation

While printed (professional) literature generally meets certain quality standards, this is not always true of internet documents. In order to be accepted as a reference for scientific research work, internet documents must strive for the same level of quality. This requires the development and implementation of binding criteria for the quality of information available through the internet and the definition of clear work practices and responsibilities. The quality of internet documents can be determined on the basis of the following criteria:

• contents: up-to-dateness, quality (assurance) and appropriate quantity (obligatory contents)
• providers: transparency of information providers and sources
• security: protection of personal data, system security
• target group: personalization, user-friendliness and ease of use
• technical aspects: technical quality criteria (validity, code conformity, usability, consistency)

The basis of the information platform is a Content Management System (CMS) for the technical administration of the data. Building on this, detailed concepts have been prepared and implemented for each of the above points, e.g. instructions for daily operations, training materials and quality manuals for editors, forms and scripts for the automatization of work procedures as well as cooperative ventures with national standardization bodies.

Evaluation

Within the scope of a multi-network quality offensive, users are currently being surveyed online with respect to the acceptability of the contents and technical aspects of the information portal. Parallel to this the anonymous and personalized access modes are being evaluated statistically based on protocol files. The resulting insights will influence the structural presentation of the portal.

Further Developments

The introduction of a CMS represents a solid foundation for the presentation of the static contents of an information platform. The full potential of the World Wide Web, however, can be utilized only through the introduction of expanded description languages. Today's contents cannot be processed automatically, as the further utilization of contents is limited by a lack of metadata and access interfaces. The relational correlations between web objects are not defined and semantically well-founded terminologies are only just being developed. Our work aims at achieving a distributed, web-based information management system by using technologies such as XML, RDF/S, OWL, SOAP, WebDAV and Web Services, in particular:

• structuring and distributed annotation of web documents for health information systems with regard to quality, long-term utilization and interchangeability
• semantic search techniques and methods, topic-based and personalized searches, integration in existing portals
• web services from autonomous agents in a semantic network using modern communications protocols
• investigating the possibilities of Content Syndication (integration of external contents / export of internal contents) through the example of a central page for press releases, deadlines, newsletters etc. of all medical competence networks
• recommendations for the use of semantic metadata on documents (Dublin Code), classification of content (PICS), the seriousness and credibility of contents (HONcode)
• automated consistency checks in order to avoid invalid references with respect to temporal and contextual properties
• agent-based, adaptive evaluation of information service use and the creation of profiles for the optimization of topological structure

Project team: Sebastian Dietzold Matthias Löbe Roland Mücke Sebastian Stäubert Project management: Prof. Markus Löffler

Contact address:   IMISE, Universität Leipzig, Liebigstr. 27, 04103 Leipzig Phone: 0341 97 16104, Fax: 0341 97 16130 E-mail: E-Mail-Adresse