PubMed HealthSearch

Biomedical subjects

H U Prokosch

Publications and source records attributed to H U Prokosch.

6 recordsLinked to original sources

Knowledge based functions for routine use at a German university hospital setting: the issue of fine tuning.

In this paper we present the introduction of knowledge based functions into clinical routine at Giessen University Hospital. For this purpose a therapy planning module at the medical intensive care unit has been extensively redesigned in order to support the structured documentation of drug prescriptions. After introduction of this new HIS component in January 1996 research has been initiated to establish a basic drug therapy knowledge base. The main components of a knowledge based system have been fully incorporated into the hospital information system WING and are in routine use since December 1996. During a pre-production phase warnings of reminder functions were logged and reviewed by an interdisciplinary team in order to adapt the system to the actual clinical environment. The paper describes experiences during this fine tuning and adaptation process which was necessary to bring a small set of knowledge modules into clinical routine.

Artificial Intelligence

A semantic network model for the medical record of a rheumatology clinic.

For the development of a rheumatology information system, a medical data dictionary was developed that supports all phases of software development. In the design phase, the medical expert described his clinical environment and the rheumatology medical record in a semantic network structure. Causal relationships between different items of the medical record (e.g., a problem may be related to an adverse drug event caused by a particular drug) are also represented in the semantic network and transferred into referential integrity constraints of the patient database. Furthermore, by also integrating the domain management as a feature of the medical data dictionary, the elementary attributes of the medical record and the associated lists of valid attribute entries have also been defined within the semantic network. This structure allowed the automatic generation of data entry screens, thus making the clinical applications as independent from any hardcoded program module as possible.

Algorithms

From HIS to IAIMS: expanding the scope of information processing applications in a German university hospital.

Since the mid eighties the department of medical informatics at the University Hospital of Giessen (Germany) has been engaged in the development of a comprehensive hospital information system. The installation of a campus wide network has set the basis to provide not only clinical patient-oriented information, but also general information resources for research, medical education and administrative purposes, thus creating an environment which in the U.S. became known as an integrated academic information management system (IAIMS). The underlying concept of the whole approach is to provide one-stop information shopping capabilities at the clinicians and administrators desktop in order to meet the increasing information needs of health professionals with the emerging reality of the potential benefits of computer and communication technologies. This paper describes the various steps performed to realize this concept at Giessen University Hospital and the evaluation results derived from analysis of the acceptance of these new technologies among our hospital staff.

Computer Systems

WING--entering a new phase of electronic data processing at the Giessen University Hospital.

At the Giessen University Hospital electronic data processing systems have been in routine use since 1975. In the early years developments were focused on ADT functions (admission/discharge/transfer) and laboratory systems. In the next decade additional systems were introduced supporting various functional departments. In the mid-eighties the need to stop the ongoing trend towards more and more separated stand-alone systems was realized and it was decided to launch a strategic evaluation and planning process which sets the foundation for an integrated hospital information system (HIS). The evaluation of the HELP system for its portability into the German hospital environment was the first step in this process. Despite its recognized capabilities in integrating decision support and communication technologies, and its powerful HIS development tools, the large differences between American and German hospital organization, influencing all existing HELP applications, and the incompatibility of the HELP tools with modern software standards were two important factors forcing the investigation of alternative solutions. With the HELP experience in mind, a HIS concept for the Giessen University Hospital was developed. This new concept centers on the idea of a centralized relational patient database on a highly reliable database server, and clinical front-end applications which might be running on various other computer systems (mainframes, departmental UNIX satellites or PCs in a LAN) integrated into a comprehensive open HIS network. The first step towards this integrated approach was performed with the implementation of ADT and results reporting functions on care units.

Database Management Systems

Knowledge representation in pharmacology. A possible application area for the Arden Syntax?

In 1990 the Arden Syntax was proposed as a first version of a standardized syntax for the representation of medical knowledge. For the evaluation of the practicability of this first release we have analyzed the medical and pharmacological knowledge applied in the process of drug prescription. The separation of declarative (e.g. in a semantic network) and procedural knowledge is a basic issue of our research. We therefore propose to further extend the Arden syntax with declarative knowledge representation facilities. One way to do this may be the incorporation of a standardized medical data dictionary (e.g. the UMLS Metathesaurus) which promotes the representation of medical terms in a semantic network. Furthermore the problem of 'institution-specific knowledge', which is especially important for the issue of knowledge sharing between different institutions, is analyzed based on examples of knowledge modules for monitoring drug allergies and drug-drug-interactions.

Artificial Intelligence

Applying expert system techniques to human genetics.

Although expert systems have been developed in a variety of medical areas, there has been very little application of expert system techniques to the field of human genetics. The purpose of the research project described in this paper was (1) to experiment with different types of knowledge representation for data and knowledge structures in human genetics and (2) to explore the applicability of different inference mechanisms for various genetic problems. We present an object-oriented and a fact-based model for the representation of genealogical information and describe two prototype systems.

Algorithms