Auditorium Presentation

CardioOP-Data-Clas: A Meta-Thesaurus for Content Management and Meta-Authoring in Cardiac Surgery

Reinhard Friedl

Dept. Heart Surgery, University of Ulm, Germany

ABSTRACT:

Introduction: Web-based training (WBT) is often enthusiastically thought to play a future key-role in surgical education. However, the development of such applications requires tremendous financial and time resources. Existing WBT-applications are mostly designed for specific target user groups in predetermined instructional contexts. We describe an integrated and economic approach to WBT by providing multiple reuse of multimedia-data in different instructional scenarios and flexible composition of content to different target user groups on the basis of anon-line repository and a meta-information system.

Methodology: An on-line architecture for presentation-neutral storage,retrieval, authoring and flexible presentation of instructional content has been established. Multimedia content in heart surgery has been prototypically authored for different user groups (students, physicians,patients), on different instructional levels and for different teaching applications. This includes digital video of operative sequences and visualisations of surgical procedures with interactive three-dimensional models. Existing major meta-thesauri and classification systems have been reviewed in order to annotate our multimedia data and to make them accessible and computable in terms of knowledge management and multi-user tele-teaching.

Results: The evaluated thesauri did lack the appropriate granularity. Therefore we developed the CardioOP-Data-Class meta-thesuarus (CDC). It represents a mono-axial hierarchical system and is build upon classes and concepts and currently contains 13000 terms. Each concept is differentiated into the correct medical term, synonyms, colloquial terms and German and English translations. It consists of a detailed and fine-grained vocabulary of surgical procedures, instruments, materials,anatomy, physio- and pathophysiology, etiology, diseases, drugs and terms related to medical professions. Further properties are an unambiguous relationship between concepts and numeric codes and the possibility for online using, maintaining and editing the CDC from different institutions. We annotated educational units like video, images and text with the CDC in order to describe its content in a precise and computable form and with standardized terms according to our approach. Discussion/ Conclusion: The possibilities of CBT-applications are often overestimated. So far such programs do not provide individualised information and learning styles to different user groups on identical topics in heart surgery as well as in other domains. We therefore prototypically implemented an on-line architecture which technically supports principles of knowledge management. That requires a precise description of content with meta-information. The developed CDC meta-thesaurus serves as a controlled, descriptive and evaluative meta-information system for cardiac surgery. It is designed for educational purposes but may be applicable and extensible to other requirements as e.g. quality management in clinical settings.

BENEFIT TO PARTICIPANTS ATTENDING SESSION:

This work describes the development, use and maintenance of a domain-specific, extensible meta-thesaurus, the CardioOP-DataClas (CDC) to support multimedia content management within the CardioOP educational system. We also introduce the application of the Cardio-Op computer system that allows for administration and maintenance of the CDC together with storage, content-based retrieval, annotation, and meta-authoring of multimedia data in heart surgery

Reinhard Friedl
University of Ulm
Dept. Cardiac Surgey
89070 Ulm
Germany
Fax: 0731-502-6698
e-mail: reinhard.friedl@medizin.uni-ulm.de
Phone: 0731-5002-7321
Fax: 0731-5002-6698
Email: reinhard.friedl@medizin.uni-ulm.de

CO-AUTHORS:

Melitta Preisack, University of Ulm, Dept. Cardiac Surgery, Germany;
Wolfgang Klas, Dept. of Computer Science, Univ. Vienna, Austria;
Thomas Rose, Research Institute for Applied Knowledge Processing, Ulm, Germany;
Oliver Gˆdje, University of Ulm, Dept. Cardiac Surgery, Germany;
Andreas Hannekum, University of Ulm, Dept. Cardiac Surgery, Germany;