Fully Automated Surveillance of Healthcare-Associated Infections with MONI-ICU: A Breakthrough in Clinical Infection Surveillance.

OBJECTIVE: Expert surveillance of healthcare-associated infections (HCAIs) is a key parameter for good clinical practice, especially in intensive care medicine. Assessment of clinical entities such as HCAIs is a time-consuming task for highly trained experts. Such are neither available nor affordable in sufficient numbers for continuous surveillance services. Intelligent information technology (IT) tools are in urgent demand. METHODS: MONI-ICU (monitoring of nosocomial infections in intensive care units (ICUs)) has been developed methodologically and practically in a stepwise manner and is a reliable surveillance IT tool for clinical experts. It uses information from the patient data management systems in the ICUs, the laboratory information system, and the administrative hospital information system of the Vienna General Hospital as well as medical expert knowledge on infection criteria applied in a multilevel approach which includes fuzzy logic rules. RESULTS: We describe the use of this system in clinical routine and compare the results generated automatically by MONI-ICU with those generated in parallel by trained surveillance staff using patient chart reviews and other available information ("gold standard"). A total of 99 ICU patient admissions representing 1007 patient days were analyzed. MONI-ICU identified correctly the presence of an HCAI condition in 28/31 cases (sensitivity, 90.3%) and their absence in 68/68 of the non-HCAI cases (specificity, 100%), the latter meaning that MONI-ICU produced no "false alarms". The 3 missed cases were due to correctable technical errors. The time taken for conventional surveillance at the 52 ward visits was 82.5 hours. MONI-ICU analysis of the same patient cases, including careful review of the generated results, required only 12.5 hours (15.2%). CONCLUSION: Provided structured and sufficient information on clinical findings is online available, MONI-ICU provides an almost real-time view of clinical indicators for HCAI - at the cost of almost no additional time on the part of surveillance staff or clinicians.

Knowledge-based interpretation of toxoplasmosis serology test results including fuzzy temporal concepts--the ToxoNet system.

Transplacental transmission of Toxoplasma gondii from an infected, pregnant woman to the unborn that occurs with a probability of about 60 percent [1] results in fetal damage to a degree depending on the gestational age. The computer system ToxoNet processes the results of serological antibody tests having been performed during pregnancy by means of a knowledge base containing medical knowledge on the interpretation of Toxoplasmosis serology tests. By applying this knowledge ToxoNet generates interpretive reports consisting of a diagnostic interpretation and recommendations for therapy and further testing. For that purpose it matches the results of all serological investigations of maternal blood with the content of the knowledge base returning complete textual interpretations for all given findings. The interpretation algorithm derives the stage of maternal infection from these that is used to infer the degree of fetal threat. To consider varying immune responses of particular patients, certain time intervals have to be kept between two subsequent tests in order to guarantee a correct interpretation of the test results. These time intervals are modelled as fuzzy sets, since they allow the formal description of the temporal uncertainties. ToxoNet comprises the knowledge base, an interpretation system, and a program for the creation and modification of the knowledge base. It is available from the World Wide Web by starting a standard browser like the Internet Explorer or the Netscape Navigator. Thus ToxoNet supports the physician in Toxoplasmosis diagnostics and in addition allows to adopt the way of making decisions to the characteristics of the particular laboratory by modifying the underlying knowledge base.

A prospective evaluation of the medical consultation system CADIAG-II/RHEUMA in a rheumatological outpatient clinic.

To evaluate the performance of CADIAG-II/RHEUMA as consultant in the primary evaluation of patients visiting a rheumatological outpatient clinic, a CADIAG-II/RHEUMA consultation was done for 54 patients and the list of generated diagnostic hypotheses was compared to each clinical discharge diagnosis. For 26 of a total of 126 rheumatological discharge diagnoses, no matching CADIAG-II/RHEUMA diagnosis was available. 94% of all other discharge diagnoses were found in the list of CADIAG-II/RHEUMA hypotheses, 82% among the first third of the list of hypotheses and 48% among the first five hypotheses. We identified the following factors limiting the ability of CADIAG-II/RHEUMA to generate a comprehensive and correctly ranked list of diagnostic hypotheses: (1) a large percentage of patients with early stages of not clearly identified rheumatological conditions; (2) the limited number of CADIAG-II/RHEUMA diagnoses compared to the large number of known rheumatological conditions; (3) the fact that rheumatological diseases are rarely characterized by a single pathognomonic feature but are usually diagnosed by combinations of rather unspecific findings.

[Knowledge-based diagnosis and therapeutic recommendations with fuzzy-set theory methods in patients with acute lung failure (ARDS)]. / Wissensbasierte Diagnostik und Therapieempfehlung mit Methoden der Fuzzy-Set-Theorie bei Patienten mit akutem Lungenversagen (ARDS).

OBJECTIVE: Since the treatment of patients with severe ARDS using the extracorporal lung assist (ECLA) methods remains a cost intensive and speculative procedure, a knowledge based computer system should be created and evaluated in order to support clinical decisions. METHODS: The model was based on the fuzzy set theory and therefore able to give decisions between yes and no, that means that a criterion could also be fulfilled to 35% or 80% for example. The development of this computer program consists of two steps: first, the entry criteria for the ECLA therapy were established within a framework of an international evaluation of clinical data from 3 centres (Berlin, Marburg, Vienna). Here, inherent vagueness, uncertainty of the occurrence and limited availability of medical data are to be considered to establish a useful tool. Secondly, this was done by grouping and weighting of parameters by the system and the status of each patient or patient group was assigned by the percentage of fulfillment of the criterion. RESULTS: By using a mixed sample of patients from these three centres, the fulfillment of entry criteria according either to definitions of Berlin or to definition of Marburg was different (68% versus 36%). Other differences (36% vs. 22% and 68% vs. 60%) were found between the fuzzy based score and the crisp score which represents the usually performed method. CONCLUSIONS: This now preevaluated minimal data set to describe severe ARDS patients based on the fuzzy set theory may be useful to evaluate patients for ECLA therapy or for another controlled ARDS-therapy.

[RHEUMexpert: a documentation and expert system for rheumatic diseases]. / RHEUMexpert: Ein Dokumentations- und Expertensystem für rheumatische Erkrankungen.

A computer assisted documentation of signs and findings in rheumatic diseases is described. This documentation was developed by the Austrian Society for Rheumatology and thought to be a minimal standard for the use by general practitioners. In addition, a knowledge-based basic differential diagnosis support was developed, which differentiates between major groups of rheumatic diseases as inflammatory spine diseases, mechanical or metabolic reasons for spine disorders, inflammatory joint diseases, degenerative or metabolic joint diseases, soft tissue diseases. This presentation describes the results of an evaluation of 75 typical case histories and a second study where 252 case histories were documented retrospectively in this new system. The results of the first showed a pretty good discrimination between the described groups of different diagnoses (sensitivity between 71 and 100 percent for all groups with the exception of metabolic joint diseases, specificity between 75 and 94 percent). The second--retrospective--documentation and diagnostic support showed much weaker results (sensitivity for major groups 74-76 percent). The reasons for the different outcomes are discussed: On the one hand, signs and symptoms from case reports could not be transferred completely in the new documentation, as some findings retrospectively could not be defined sharp enough. On the other hand the study showed, that the sensitivity of well defined disorders as inflammatory joint diseases (exp. rheumatoid arthritis) reaches almost 100 percent, whereas it is as low as 50 percent in some other diseases (e.g. gout) whose characteristic findings and symptoms are suppressed by treatment (drug medication) in many cases. The results show that computer based documentation of rheumatic diseases facilitates the systematized and standardised documentation of patient data. However, a few modifications of the knowledge base as well as the knowledge representation formalisms are necessary to achieve a better performance in differential diagnostic support.

[Toxoplasmosis diagnosis in pregnancy: computer assisted follow-up interpretation of serologic tests]. / Toxoplasmose-Diagnostik in der Schwangerschaft: Computerunterstützte Verlaufsinterpretation von serologischen Tests.

Primary infection with Toxoplasma gondii during pregnancy can result in fetal infection with serious sequelae for the unborn if not treated properly. Early diagnosis enables drug therapy and significantly reduces the risk of fetal disease. A systematic serological screening procedure was established in Austria in 1975 to detect primary toxoplasma infection as early as possible during pregnancy. Since the screening program is based solely on observation and interpretation of serological data, the question arises whether a knowledge-based system for automatic interpretation can achieve a sufficient interpretative accuracy for introduction to routine work. For this reason the system Toxopert-I was developed. The system is aimed at facilitating routine laboratory work, as well as assuring quality by setting standards for therapy. The required knowledge base was designed as a knowledge graph, each state representing a certain interpretation. One or more available serological test results cause the knowledge graph to change its current state. If all available test results are processed, the final state reached corresponds to the respective current interpretation for the patient. A retrospective analysis of 1000 pregnant women yielded a total diagnostic sensitivity and specificity of over 99% in comparison with the clinician's diagnosis which was used as the Gold Standard.

Toxopert-I: knowledge-based automatic interpretation of serological tests for toxoplasmosis.

Primary infection with Toxoplasma gondii, a parasite found in most regions of the world, is asymptomatic in more than 80% of cases. However, primary infection with Toxoplasma gondii in a pregnant woman might cause fetal infection and severe damage. Most cases do not require treatment. This applies to women without any infection (denoted as seronegative) and women who have acquired the infection before conception (denoted as latent). In contrast, women with postconceptual infection require immediate treatment to prevent or ameliorate fetal infection. We have developed an expert system, called Toxoport-I, designed for routine laboratory work, which automatically interprets serological test results of toxoplasma infection. By using the system the clinician can also examine questionable cases by interactively exploring possible results. We used a popular method of designing expert systems applied to medical interpretation and therapy advice, the rule-based one. In order to meet the requirements of automatic interpretation in toxoplasma serology the following characteristics were introduced: the interpretation of sequences of test results, the possibility of excluding inconsistent test results and the adaptability of the knowledge base. A decision graph that covers the different kinds of infections as well as therapy and recommendations for further tests was designed, implemented and was clinically tested by carrying out a retrospective study including 1000 pregnant women. A comparison of Toxoport-I and the clinician's interpretations yielded sensitivity and specificity rates of over 99% each.

Knowledge-based interpretation of serologic tests for hepatitis on the World Wide Web.

HEPAXPERT is a knowledge-based system that interprets the results of routine serologic tests for infection with hepatitis A and B viruses. The following tests are included: hepatitis A virus antibodies, IgM antibodies to the hepatitis A virus, hepatitis A virus in stool, hepatitis B surface antigen and antibodies, antibodies to hepatitis B core antigen, and hepatitis B envelope antigen and antibodies. HEPAXPERT/WWW, an implementation of HEPAXPERT-III for the World Wide Web, can be reached by URL http://www.med-expert.co.at/hepax. After selecting HEPAXPERT/WWW, serologic test results can be entered and will be transferred as an E-mail message for subsequent interpretation, which is done off-line with HEPAXPERT-III. The textual interpretation is sent back via E-mail. Each qualitative test for hepatitis A and B antibodies and antigens may produce one of four possible results: positive, negative, borderline, and not tested. To cover the resulting 64 (A) and 57,344 (B) combinations of findings, the knowledge base of HEPAXPERT/WWW contains 16 rules of hepatitis A and 131 rules for hepatitis B serology interpretation. This basic knowledge is structured such that all possible combinations of findings can be interpreted, and there is no overlap in the premises underlying the rules. The reports that the system automatically generates include the transferred results of the tests; a detailed analysis of the results, including virus exposure, immunity, stage of illness, prognosis, infectiousness, and vaccination recommendation; and, as an option, an identification, to distinguish the origin of the interpretation requests.

Feasibility analysis of a case-based reasoning system for automated detection of coronary heart disease from myocardial scintigrams.

Myocardial perfusion scintigraphy is a noninvasive diagnostic method for the evaluation of patients with suspected or proven coronary artery disease (CAD). We utilized case-based reasoning (CBR) methods to develop the computer-based image interpretation system SCINA which automatically derives from a scintigraphic image data set an assessment concerning the presence of CAD. We compiled a case library of 100 patients who underwent both perfusion scintigraphy and coronary angiography to document or exclude the presence of CAD. The angiographic diagnosis of the retrieved nearest neighbor match of a scintigraphic input case was selected as the CBR diagnosis. We examined the effects of input data granularity, case indexing, similarity metric, and adaptation on the diagnostic accuracy of the CBR application SCINA. For the final prototype, sensitivity and specificity for detection of coronary heart disease were 98% and 70% suggesting that CBR systems may achieve a diagnostic accuracy that appears feasible for clinical use.

The knowledge model of MedFrame/CADIAG-IV.

The medical consultation system MedFrame/CADIAG-IV is a successor of the prior CADIAG projects. It is the result of a complete redesign to account for today's demands on state-of-the-art software. Its knowledge representation and inference process are based on fuzzy set theory and fuzzy logic. Fuzzy sets are used for conversions from measured numeric values and observational data into symbolic ones. Medical relationships between findings, diseases, and therapies, the rules, are represented by fuzzy relations, that express positive or negative associations. Findings, diseases, and therapies are organised in hierarchies.

A WWW-accessible knowledge base for the interpretation of hepatitis serologic tests.

HEPAXPERT is a knowledge-based system that interprets the results of routine serologic tests for infection with hepatitis A and B viruses. The following tests are included: hepatitis A virus anti-bodies (anti-HAV), IgM antibodies to the hepatitis A virus (IgM anti-HAV), hepatitis A virus (HAV in stool, hepatitis B surface antigen (HBsAg) and antibodies (qualitative anti-HBs, quantitative anti-HBs titre), antibodies to hepatitis B core antigen (anti-HBc and IgM anti-HBc), and hepatitis B envelope antigen (HBeAg) and antibodies (anti-HBe). HEPAXPERT/WWW--an implementation of HEPAXPERT-III for WWW--can be reached by URL http://www.swun.com/hepax of the World Wide Web. After selecting HEPAXPERT/WWW, serologic test results can be entered and will be transferred as an e-mail message for subsequent interpretation which is done off-line with HEPAXPERT-III. The textual interpretation is sent back via e-mail. Each qualitative test for hepatitis A and B antibodies and antigens may produce one of four possible results: positive, negative, borderline, and not tested. To cover the resulting 64 (A) and 57344 (B) combinations of findings, the knowledge base of HEPAXPERT/WWW contains 16 rules for hepatitis A and 131 rules for hepatitis B serology interpretation. This basic knowledge is structured such that all possible combinations of findings can be interpreted and there is no overlap in the premises underlying the rules. The reports that the system automatically generates include: (a) the transferred results of the tests; (b) a detailed analysis of the results, including virus exposure, immunity, stage of illness, prognosis, infectiousness, and vaccination recommendation; and (c) optional: an ID to distinguish the origin of the interpretation requests.

Development and evaluation of fuzzy criteria for the diagnosis of rheumatoid arthritis.

In 1987, the American Rheumatism Association issued a set of criteria for the classification of rheumatoid arthritis (RA) to provide a uniform definition of RA patients. Fuzzy set theory and fuzzy logic were used to transform this set of criteria into a diagnostic tool that offers diagnoses at different levels of confidence: a definite level, which was consistent with the original criteria definition, as well as several possible and superdefinite levels. Two fuzzy models and a reference model which provided results at a definite level only were applied to 292 clinical cases from a hospital for rheumatic diseases. At the definite level, all models yielded a sensitivity rate of 72.6% and a specificity rate of 87.0%. Sensitivity and specificity rates at the possible levels ranged from 73.3% to 85.6% and from 83.6% to 87.0%. At the superdefinite levels, sensitivity rates ranged from 39.0% to 63.7% and specificity rates from 90.4% to 95.2%. Fuzzy techniques were helpful to add flexibility to preexisting diagnostic criteria in order to obtain diagnoses at the desired level of confidence.

[New approaches to computer-assisted diagnosis of rheumatologic diseases]. / Neue Ansätze zur computerassistierten Diagnose rheumatologischer Erkrankungen.

UNLABELLED: Since the 1960s, several knowledge-based systems for computer-assisted diagnosis in radiology have been developed. The great majority of these tools has been implemented as off-line systems. This requires interaction with the system solely for the purpose of consultation and therefore interrupts the radiologist's work flow. This and inadequate man-machine interfaces may have inhibited the routine clinical use of such systems. The goal of this paper is to describe the current research toward the development of the on-line expert system Cadiag-4/Rheuma-Radio. The underlying fundamentals of the system design, including client/server architecture, communication interfaces, and fuzzy set theory and fuzzy logic as methods for knowledge representation and interference, are presented. METHODS: In radiology today, computers are routinely used to acquire radiological images in hospital and radiology information systems (HIS/RIS) and picture archiving and communication systems (PACS). In our approach, we make use of pre-existent sources of information to build an expert system that minimizes the interaction between radiologists and the computer. To handle uncertainty and vagueness of medical knowledge, fuzzy set theory and fuzzy logic are used. Given data of a specific case, a deductive inference procedure combines the observed radiological signs, establishes confirmed and excluded diagnoses as well as diagnostic hypotheses, and provides explanations for these conclusions. Furthermore, proposals for confirmation or exclusion of diagnostic hypotheses are offered. RESULTS: For evaluation purposes, an early prototype of Cadiag-4/Rheuma-Radio was tested on radiological disorders of the hip joint related to rheumatological diseases. Twenty radiological cases were used as test cases, reaching a diagnostic accuracy of about 80%. CONCLUSION: The first results are acceptable and encourage further work to cover the whole area of rheumatologically relevant radiological signs and diagnoses. Furthermore, research into the development of user-oriented data acquisition tools will be carried out.

Development and retrospective evaluation of Hepaxpert-I: a routinely-used expert system for interpretive analysis of hepatitis A and B serologic findings.

Hepaxpert-I is an expert system that interprets the results of routine serologic tests for infection with hepatitis A or B virus. The tests measure antibody to the hepatitis A virus (anti-HAV), IgM antibody to the hepatitis A virus (IgM anti-HAV), hepatitis A virus (HAV) in the stool, hepatitis B surface antigen (HBsAg) and antibody (anti-HBs), antibody to hepatitis B core antigen (anti-HBc and IgM anti-HBc), and hepatitis B envelope antigen (HBeAg) and antibody (anti-HBe). The knowledge base of Hepaxpert-I contains 13 If-Then rules for hepatitis A and 106 If-Then rules for hepatitis B serology. Formally, knowledge acquisition was done by forming a partition of each of the two sets of possible serologic finding patterns that contain patterns of serologic test results, 64 for hepatitis A and 4096 for hepatitis B, respectively. After entering an input pattern of serologic test results in Hepaxpert-I, a rule pattern matching algorithm based on indexing is internally employed as efficient access method for providing the respective interpretive text. Since 1 September 1989, Hepaxpert-I has been routinely applied at the Hepatitis Serology Laboratory of the 2nd Department of Gastroenterology and Hepatology at the University of Vienna Medical School (Vienna General Hospital). Beforehand, a retrospective evaluation of the expert system based on 23,368 hepatitis A and 24,071 hepatitis B serology requests was carried out.

[Medical informatics in research, teaching and patient management]. / Medizinische Informatik in Forschung, Lehre und Patientenversorgung.

The field of medical informatics in its current understanding is defined and criteria distinguishing this field from similar areas are provided. Special consideration is given to its position at a School of Medicine - in particular to the University of Vienna Medical School with the Vienna General Hospital as its teaching hospital. Demands for medical informatics and electronic data processing (EDP) in this extended field of activity come from four different sources: (1) research in medical informatics, (2) teaching of medical informatics as well as EDP training, (3) EDP service for research and teaching, and (4) EDP hospital operations to assist patient care. (Purely administrative EDP demands are not considered here.) It is shown that the different demands can be fulfilled by the usually available institutions involved in medical informatics and EDP at a School of Medicine. At many places these institutions are as follows: (1) a department or division of medical informatics with a possibly attached computer center dedicated to provide assistance in the area of research and teaching, (2) the computer center of the respective university the School of Medicine belongs to, (3) the computer center of the hospital-owned institution responsible for all EDP activities connected to patient care, and (4) external software companies and EDP training centers. To succeed in the development of an exhaustive, school-wide system of medical informatics and EDP that considers the different demands in research, teaching, and EDP hospital operations equally, close and well-suited coordination between the institutions involved is necessary.

[Clinical aspects of arthrosis]. / Die Klinik der Arthrose.

Symptoms, physical, laboratory, and x-ray findings in osteoarthrosis are reviewed, special emphasis is drawn to the different clinical forms as compensated, activated, decompensated osteoarthrosis. A sample of 2,494 patients with degenerative joint disease or functional disorders as e.g. myalgia were studied with respect to symptoms and physical findings. As previously described in other studies it turned out, that osteoarthrosis of the knee specially in women is correlated with high body weight; furthermore it could be shown that foot deformities--hallux valgus and pes planus in gonarthrosis, hallux valgus and hammer-toe in polyarthrosis of the finger joints--are significantly more frequent in some degenerative joint diseases.

HEPAXPERT-III: knowledge-based interpretation of serologic tests for hepatitis A, B, C, and D.

1. The HEPAXPERT-III SYSTEM. HEPAXPERT-III--the successor of HEPAXPERT-I[1] and HEPAXPERT-II [2]--is a routinely-used, integrated medical database and knowledge-based system that stores and interprets the results of serologic tests for infection with hepatitis A, B, C, and D viruses. The following tests are included: Anti-HAV, IGM anti-HAV, and HAV in stool; HBsAg, anti-HBs, anti-HBc, IGM anti-HBc, HBeAg, anti-HBe, and anti-HBs titre; Anti-HCV, HCV-immunoblot, and HCV-PCR; Delta-Ag and anti-delta. HEPAXPERT-III provides the following functions: a) screen input of patient's personal data (patient ID, surname, first name, name at birth, date of birth, and sex), administrative data (department requiring the tests and date of specimen sample), and medical data (results of serologic tests); and/or b) automatic transfer of patient's personal, administrative, and medical data by connecting HEPAXPERT-III to a laboratory information system, a hospital information system, or an automated laboratory analyzer; and c) automatic generation of interpretive reports of the obtained serologic findings, including an analysis of possible virus exposition, immunity, disease stage, prognosis, and degree of infectiousness. HEPAXPERT-I and HEPAXPERT-II have been routinely used at the Vienna General Hospital, the teaching hospital of the University of Vienna Medical School. The interpretive reports are well-accepted and lead to several improvements in patient care [3]. HEPAXPERT-III will not only extend the scope of interpretation to hepatitis C and D serologic tests, but will also offer a state-of-the-art graphical user interface. 2. HARDWARE AND SOFTWARE. IBM-compatible personal computer (minimum 80386 SX processor), 8 MB RAM (OS/2) and 4 RM RAM (MS-Windows), resp., graphic adapter (minimum 640x480) and printer supported by OS/2 and MS-Windows 3.1, resp., IBM OS/2-2.1 or higher and MS-Windows 3.1 or higher, resp., and for the OS/2 version IBM Database 2 (DB2/2).

MONI: an intelligent database and monitoring system for surveillance of nosocomial infections.

Recording, recognition, and prevention of nosocomial infections are the primary responsibilities of the hospital infection control unit. To perform these tasks, this unit needs information from diverse sources--the patient's symptoms and signs, microbiological and virological test results, and information regarding antibiotics and treatment come from different levels of healthcare delivery. Because of the large amount of data (e.g., about 300 microbiological requests daily) a computer system is required to store this information and to provide a means for subsequent evaluation. MONI (Monitoring of nosocomial infections) is an intelligent database and monitoring system for surveillance and detection of nosocomial infections. Data can be entered into the system manually as well as transferred automatically from external information systems. The central feature of the system is the automatic detection of and calling attention to conditions that may be a detriment to patient recovery, such as possible hospital-acquired infections, risk factors, diseases to be reported, etc. By using this system, we seek to reduce the frequency of infection and the frequency of nosocomial deaths by improving the quality of patient treatment, shortening the length of stay in a hospital, and the use of fewer and/or cheaper antibiotics. MONI provides a means to access relevant medical data (names of infectious agents, antibiotics, department names, monitoring rules, etc.) from a library. This library can be updated or otherwise modified, even during use. An infection control team using this system can customize it to suit the demands of that particular unit. Automatic data transfer from external information systems is made possible by tables that translate between different code systems. The system also offers flexibility; the program can be configured to adapt it for use in other hospitals and institutions. The core element of MONI is the monitoring module, which is implemented as a layer between data input and the database. Upon data acquisition, the system checks the input against several monitoring tools and alerts the user to matches, which may indicate an infection risk. Processing of a rule may be deferred, depending on complexity of the rule and the actual and estimated workload of the system. Examples of the monitoring guidelines are: (1) suspicion of nosocomial infection; (2) infection at a normally sterile site; (3) infection due to bacteria with unusual antibiotic sensitivity patterns; (4) lab report indicates that patient is treated with ineffective antibiotic; (5) possible choice for less expensive antibiotic; (6) infection which is required to be reported to state and/or health authorities; (7) patients receiving prophylactic antibiotics longer than medically indicated; and (8) infections of two or more patients in different wards with the same bacteria (cf. Evans 85). The MONI system was developed at one of the largest hospitals in Europe, the Vienna General Hospital (2,200 beds). This facility serves as the teaching hospital of the University of Vienna Medical School. The size of the hospital and the large amount of data made it necessary to introduce such a system into clinical routine. MONI was programmed in C and C++ with a state-of-the-art graphical user interface (Presentation Manager, Workplace Shell) for OS/2. IBM Database 2 for OS/2 (dB 2/2) was used in constructing the database. The layer between the database and the monitoring application is driven by the multitasking and interprocess communication abilities of OS/2. A pen-based support system that assists in mobile data acquisition is currently under development.

[The knowledge base of the Hepaxpert I System: automatic interpretation of Hepatitis A and B serology]. / Die Wissensbasis des Hepaxpert-I-Systems: Automatische Interpretation der Hepatitis-A und -B-Serologie.

The knowledge base of Hepaxpert-I, a medical expert system for interpretive analysis of hepatitis A and B serologic findings, contains 13 rules for hepatitis A and 106 rules for hepatitis B serology. Formally, the construction of the knowledge base was done by forming a partition of the sets of possible serologic finding patterns--64 for hepatitis A and 4096 for hepatitis B serology--induced by an equivalence relation, divides the elements of the sets into disjoint subsets, the equivalence classes. Each equivalence class is represented as one If-Then rule that assigns to every member of the equivalence class one interpretive text. The partition of the possible finding patterns into equivalence classes and the disposal of one and only one interpretive text for each equivalence class made the creation of a very practical and efficient computer program for the precise interpretation of any finding pattern of serologic tests for hepatitis A and B possible. The complete set of the provided If-Then rules is represented in this paper.