Methods, design and development of clinical decision support systems based on the Arden syntax : with applications in the clinical laboratory

• Decision Support Systems (DSSs) are computer systems that permit explicit representation of knowledge and can execute a computer program that uses the knowledge to assist in the decision process. A clinical DSS uses both clinical data and medical knowledge to help health professionals make clinical decisions. Developing a DSS can be a costly and time-consuming task. Sharing of knowledge can decrease costs and development time and therefore increase the use of DSSs in clinical practice. One of the most important categories of DSSs in medicine is data driven systems where the knowledge base is linked to a clinical database. Methods that support knowledge base-database mapping are necessary when developing clinical DSSs.Arden Syntax is a standard specification for defining and sharing modular health knowledge bases among inflamation systems and institutions. The medical knowledge is written as independent Medical Logic Modules (MLMs) and the scope of the standard is to facilitate sharing of knowledge.In this thesis methods for implementing DSSs based on the Arden Syntax are described. Furthermore, DSS development tools and a system architecture for building clinical DSSs are presented, and design aspects such as database access, system validation, and platform independence are discussed. Clinical laboratory applications, where the DSS is integrated with an existing laboratory infotmation system for use in real time validation and interpretation of laboratory data, are implemented and evaluated. Since the DSS performs tasks that used to be done manually, such as retrieving the patients' previous test results, the system not only increases laboratory services but may also enhance diagnostic proficiency and save time. This experience has led to a search for other areas within the laboratory where a DSS can be both clinically valuable and cost-effective.

Keyword

MEDICINE

MEDICIN

Publication and Content Type

vet (subject category)

dok (subject category)