Previous studies have shown the beneficial effect of mapping tools on medical students’ ability to create diagnostic hypotheses. This holds true for the number and for the correctness of hypotheses built by the learners. A graphical drag and drop-mapping tool was integrated into the case-based CASUS learning system in 1996. This tool supported the differential diagnostic reasoning process successfully but was only available for the MacOS.

Our work focuses on the tricks and pitfalls in the development of this highly interactive visualization tool for the Web. Furthermore, we will report on qualitative and quantitative evaluation data from its use in the context of clinical courses in internal medicine and pediatrics in German and US medical schools.

In the CASUS-player application each educational unit is synthesized by creating and updating of the diagnostic hypotheses within the mapping tool. The learners are asked to graphically connect the findings of a case step-wise with the respective hypotheses. Students can evaluate the completeness and coherence of their own diagnostic reasoning and correct themselves by comparison with the respective expert solution. In the course of the transition towards a web-based version of the CASUS-system, the mapping tool was implemented as a Java-Applet. This turned out to be problematic in the University network settings due to a heterogeneous IT-infrastructure.

As a consequence, we decided to implement the currently used tool in HTML and JavaScript on the client side. On the server side the requests are handled by a Java servlet connected to a relational database. To implement the mapping tool in this technical framework, the user interface had to be redesigned completely to allow for the full range of interaction. Hypotheses and findings are organized in a table-like fashion as rows and columns. The weighed connections between findings and hypotheses by the users are generated through pop-up lists. This new version is a trade-off between graphical clearness and technical feasibility. After analysis of further formative evaluation studies we improve the tool with focus on the user interface. This will be achieved by the use of sophisticated dynamic HTML features as provided in the latest releases of standard web browsers.

In order to formatively evaluate the new version of the mapping tool and compare it with the old graphically oriented one, we conducted studies with fourth year medical students at the University of Munich and with third year medical students at three US medical schools under the leadership of the Dartmouth Medical School Department of Pediatrics. The results of these ongoing studies on the technical and pedagogical acceptance of the tool, and the quantitative and qualitative data from teachers and students are reported and critically interpreted.

BENEFIT TO PARTICIPANTS ATTENDING SESSION:

The development process and general principles of mapping tools for supporting the visualization of diagnostic reasoning is presented. Technical and pedagogical experiences will be described in a general fashion in this field. The benefit of an international collaboration between Germany and the US will be highlighted.

Inga Hege
University of Munich
Medizinische Klinik - Innenstadt
Ziemssenstr. 1
80336 Munich
GERMANY
Phone: *49-89-5160-2289
Fax: *49-89-5160-2366
Email: inga.hege@stud.uni-muenchen.de

CO-AUTHORS:
Norman Berman and Leslie Fall
Dartmouth Hitchcock Medical Center
1 Medical Center Drive
Lebanon, NH 03756

Martin Adler, Markus Kernt and Martin R. Fischer
University of Munich, Germany