The use of mivar expert systems for diagnosis of bacterial antibiotic resistance

The study is dedicated to the use of mivar expert systems for identifying bacterial resistance to existing antibiotics. A modular architecture of the system was presented, which allows easy addition and updating of individual components. A knowledge base consisting of 56 rules for working with the expert system was created. It is proposed to implement the system using the KESMI software, which allowed for logical conclusions to be drawn. The system was tested on three different cases. The first case involved the presence of a mutation in the mecA gene, the second involved methylated ribosomes, and the third involved Gram-positive bacteria. Testing of the Mivar expert system showed that the bacteria's resistance results matched the established knowledge base. The impact of using Mivar expert systems on the process of detecting antibiotic resistance has been studied. A description of the methodologies used to evaluate the system's effectiveness was proposed. It was justified why the use of expert systems can significantly improve the diagnosis and treatment of infectious diseases.