(Projektbeschreibung nur in Englisch)
C. Chizzali-Bonfadin, K.-P. Adlassnig
Department of Medical Computer Sciences, Section on Medical Expert and Knowledge-Based
Systems, University of Vienna Medical School, Spitalgasse 23, A-1090 Vienna, Austria
e-mail: klaus-peter.adlassnig@meduniwien.ac.at
Background
Knowledge-based decision support systems will become more and more important in
health care. Based on Internet technology such systems perfectly match the requirements
(which are access from every computer, ease of use, ...) to work in clinical infrastructures
for the next century.
Objective
Since summer 1998 a knowledge-base server is available in the World Wide Web at URL
http://medexpert.msi.meduniwien.ac.at
(Fig. 6). Users of this server have the possibility to
apply several of knowledge-based systems for decision support in medicine.
The following systems developed at our section are in operation at present:
A brief description of HEPAXPERT-III/WWW will follow below.
Material and Methods
HEPAXPERT-III/WWW an implementation of HEPAXPERT-III [1] for WWW is an ex-
pert system that interprets the results of qualitative and quantitative routine serologic tests
for infection with hepatitis A and B. The fully interactive system based on Cold Fusion
Application Server is the successor of an off-line HEPAXPERT/WWW [2] that returned
the results by e-mail up to 24 hours after request. The system automatically provides and
interprets the result of the laboratory tests measuring antigens and antibodies, without the
use of additional biochemical or clinical data, and thus helps physicians to understand
complex serologic findings. To deal with all possible combinations of findings, the knowledge
base of HEPAXPERT-III/WWW contains 16 rules for hepatitis A and 131 rules for
hepatitis B interpretation. In analyzing serologic test results, the program compares the
constellation of serologic findings with constellations that may occur in the course of
hepatitis A or B infection. Possible active or passive immunization is taken into account
as well. However, there are several possible sources of misinterpretation, such as deviations
in the course of the disease from the assumed courses, as well as falsely positive
and falsely negative findings. Therefore, in each case the program's conclusions have to
be correlated with the patient's overall clinical picture.
HEPAXPERT-III/WWW can be reached by URL http://medexpert.msi.meduniwien.ac.at/hepax. Serologic test results can be entered (Fig. 7) and will be interpreted immediately (Fig. 8).
Results
The reports that HEPAXPERT-III/WWW automatically generates include (a) the entered
results of the tests, and (b) a detailed analysis of the results, including virus exposure,
immunity, stage of illness, prognosis, infectiousness, and vaccination recommendation. It
is possible to interpret incomplete and uncertain results as well as prototypical results.
Furthermore, as a check on the laboratory results, any inconsistent combinations of findings
are identified.
Earlier versions of HEPAXPERT have been in routine use at the hepatitis laboratory of
the University of Vienna Medical School since September 1989. The structure of the
knowledge base has resulted in an exceptionally practice-oriented and efficient interpretation
program, among whose merits are convenient data input and fast interpretation of
findings. Its acceptance among physicians is high. HEPAXPERT-III/WWW is available
since April 1998.
Technical Specification
MedExpert/WWW is based on Microsoft Internet Information Server. The operating system of
the Dual-Pentium-II personal computer is Microsoft Windows NT. The web server
is extended by Allaire Cold Fusion Application Server. MedExpert/WWW can be accessed from
any state-of-the-art WWW browser. Some applications are developed using
Java and/or JavaScript. Therefore these features have to be enabled in the configuration
of the browser.
Conclusion
Additional systems are under development and will be offered soon. Because of the rapidly
growing importance of the Internet and of medical knowledge, interactive medical
knowledge-based systems which are based on Internet and World Wide Web technology
will become most important in the future.
References