THE ANALYSIS OF MEDICAL EXPERT SYSTEMS’ USE RISKS

In the following paper the characteristics of medical information computer systems are presented. There were separated out medical expert systems as having significant features. Actual medical expert systems are considered as critical ones, especially sensitive to the occurrence of risks and risk situations. In general, there are common approaches to risk analysis in the development and operation of software. In this work, the classification of risks in accordance with selected methodologies associated with information support, or automation of the diagnostic process is presented. The abovementioned systems are complex software and computer combinations. When creating such complexes, a large number of risks arise, which in turn can cause abnormal situations that, first of all, can lead to serious consequences. Therefore, the identification and qualification of risks in advance, according to existing risk classifications, can prevent such extraordinary situations. To create such risk-qualifying systems, certain databases have been developed that are founded on studies conducted by a large number of people. In turn, we need to use the work of a large number of specialists to create the necessary knowledge base. Among the chosen risks classification systems, the MSF systemizing was selected for the following study. In order to realize the comparative characteristics of the known MES, the method of inconstant (variant) networks was used based on the selected classification. According to this method, each system was evaluated. Based on the analysis of several methodologies, it has been determined that the use of the MSF methodology is most appropriate when it comes to medical expert systems. The most significant risks that arise during the operation of such systems are identified. When analyzing each risk, the probability of its occurrence is determined and the consequences of its implementation are taken into account. In the abovementioned MES we have taken into account the risks associated with the operation, interface and user qualification. The selected classification has a large number of risk categories that are relevant to the various stages of the product life cycle, and it makes it possible to identify, classify and deal with the risks arising from the operation of the MES.

Abstract licence: CC BY-NC

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.