Fuzzy Models of Differential Diagnosis of Pyelonephritis in Pregnant Women

Purpose of the research is to improve the quality indicators of differential diagnosis of pyelonephritis in pregnant women while reducing the decision-making time by reducing the dimension of the feature space.

Methods. Differential diagnosis of such forms of pathology of pregnant women as serous pyelonephritis, purulent pyelonephritis and its transitional form belongs to the class of poorly formalized tasks. Exploratory analysis of the data structure presented by the urological department of the Kursk BSMP showed that according to the criteria of decision -making time-diagnostic quality, preference should be given to the method of synthesis of hybrid fuzzy models, combining the logic of L. Zadeh and E. Shortlif. At the same time, for a reliable diagnosis, it is enough to use data on the immune status of pregnant women.

Results. Using the chosen synthesis method, a system of fuzzy decision rules is obtained that combines E. Shortlif's iterative rules with the functions of belonging to the studied forms of pyelonephritis. Mathematical modeling and statistical tests of the obtained decision rules showed that in terms of the immune status, the diagnostic sensitivity, specificity and effectiveness in the classification of serous, purulent and intermediate forms of pyelonephritis exceed the level of 0.9.

Conclusion. In the course of research, it was found that for the differential diagnosis of forms of pyelonephritis in pregnant women by indicators of immune status, hybrid fuzzy models should be used with the aggregation of iterative models for assessing confidence in decisions made with membership functions. The obtained hybrid models have diagnostic sensitivity, specificity, and effectiveness higher than 0.9, which allows improving the quality of medical services provided to pregnant women with acute forms of pyelonephritis.

1. Verbruggen H. B., Babuska R. Fuzzy logic control advances in applications, World Scientific Series in Robotics and Intelligent Systems. Technology and Engineering, 1999, vol. 23, p. 340.

2. Seregin S. Р., Dolzhenkov S. D., Korenevskaya S. N., Sapitonova T. N. Sistema pod-derzhki prinyatiya reshenij pri differencial'noj diagnostike form pielonefrita u beremennyh zhenshchin na osnove immunnogo i oksidantnogo statusa [Decision support system for differential diagnosis of pyelonephritis forms in pregnant women based on immune and oxidant status], Medicinskaya tekhnika = Medical Equipment, 2019, vol. 53, no. 5, pp. 45-48.

3. Holimenko I. M., Konoply A. I., Bratchikov O. I., Bystrova N. A., Mavrin M. Yu., Shatokhin M. N. Oksidantnyj ctress pri ostrom seroznom i gnojnom pielonefrite [Oxidant stress in acute serous and purulent pyelonephritis], Nefrologiya = Nephrology, 2017, no. 21 (1), pp. 87-94.

4. Shatohin M. N., Kholimenko I. M., Konoply A. I., Bratchikov O. I., Gavrilyuk V. P., Krasnov A. V., Mavrin M. Yu. Matematicheskaya ocenka diagnosticheskoj informativnosti laboratornyh pokazatelej pri seroznom i gnojnom pielonefrite [Mathematical assessment of diagnostic informativeness of laboratory parameters in serous and purulent pyelonephritis], Eksperimental'naya i klinicheskaya urologiya = Experimental and Clinical Urology, 2016, no. 4, pp. 104-110.

5. Bojcov A. V. A, Lazurina L. P., Korenevskaya S. N., Shutkin A. N. Primenenie teorii izmereniya latentnyh peremennyh dlya formirovaniya prostranstva informativnyh priznakov v zadachah ocenki funkcional'nogo sostoyaniya cheloveka [Application of the theory of measurement of latent variables for the formation of the space of informative features in the tasks of assessing the functional state of a person], Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel'naya tekhnika, informatika. Medicinskoe priboro-stroenie = Proceedings of the Southwest State University. Series: Management, Computer Engineering, Computer Science. Medical Instrumentation, 2014, no. 6 (57), pp. 52-58.

6. Korenevskij N. A. Ispol'zovanie nechetkoj logiki prinyatiya reshenij dlya medicinskih ekspertnyh system [Using fuzzy decision logic for medical expert systems], Medicinskaya tekhnika = Medical Equipment, 2015, no. 1 (289), pp. 33-35.

7. Korenevskij N. A. Proektirovanie sistem prinyatiya reshenij na nechetkih setevyh modelyah v zadachah medicinskoj diagnostiki i prognozirovaniya [Designing decision-making systems based on fuzzy network models for medical diagnostics and forecasting], Telekommunikacii = Telecommunications, 2006, no. 6, pp. 25-31.

8. Korenevskij N. A. Metod sinteza geterogennyh nechetkih pravil dlya analiza i uprav-leniya sostoyaniem biotekhnicheskih system [Method of synthesis of heterogeneous fuzzy rules for the analysis and management of the state of biotechnical systems], Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel'naya tekhnika, informatika. Medicinskoe priborostroenie = Proceedings of the Southwest State University. Series: Management, Computer Engineering, Computer Science. Medical Instrumentation, 2013, no. 2, pp. 99-103.

9. Reis B. A., e. a. Sposob diagnostiki razlichnyh form ostrogo pielonefrita [Method of diagnosis of various forms of acute pyelonephritis]. Patent RF, no. 2478968, 2013.

10. Korenevskij N. A., Razumova К. V. Sintez kollektivov gibridnyh nechetkih modelej ocenki sostoyaniya slozhnyh system [Synthesis of collectives of hybrid fuzzy models for assessing the state of complex systems], Naukoemkie tekhnologii = High-tech Technologies, 2014, vol. 15, no. 12, pp. 31-40.

11. Korenevskij N. A., Rodionova S. N., Khripina I. I. Metodologiya sinteza gibridnyh nechyotkih reshayushchih pravil dlya medicinskih intellektual'nyh sistem podderzhki prinya-tiya reshenij [Methodology of synthesis of hybrid fuzzy decision rules for medical intelligent decision support systems], Stary Oskol, TNT Publ., 2019. 472 p.

12. Korenevskij N. A., Shutkin A. N., Boitsova E. A. Ocenka i upravlenie sostoyaniem zdorov'ya na osnove modelej G. Rasha [Assessment and management of the state of health based on the models of G. Rush], Medicinskaya tekhnika = Medical Equipment, 2015, no. 6, pp. 37-40.

13. Korenevskij N. A., Shutkin A. N., Gorbatenko S. A., Serebrovsky V. I. Ocenka i upravlenie sostoyaniya zdorov'ya obuchayushchihsya na osnove gibridnyh intellektual'nyh tekhnologij [Assessment and management of the state of health of students on the basis of hybrid intellectual technologies], Stary Oskol, TNT Publ., 2016. 472 p.

14. Seregin S. P. Sovremennye informacionnye tekhnologii v urologii [Modern information technologies in urology], Kursk, Kursk Publ., 2009. 364 p.

15. Korenevskiy N. A. Application of fuzzy logic for decision-making in medical expert systems. Biomedical Engineering, 2015, vol. 49, is. 1, pp. 46-49.

16. Korenevskiy N. A., Degtyarev S. V., Seregin S. P., Novikov A. V. Use of an Interactive Method for Classification in problems of Medical Diagnosis. Biomedical Engineering, 2013, vol. 47, is. 4, pp. 169-172.

17. Bruce G. Buchanan, Edward H. Sportlife. Rule-Based Expert Systems The MYCIN Experiments of the Stanford Heuristic Programming Projext. Reading, Addison-Wesley Publ., 1984. 235 p.

18. Seregin S. P., Dolzhenkov S. D., Korenevskaya S. N., Sapitonova T. N. Sintez kom-binirovannyh nechetkih reshayushchih pravil dlya prognozirovaniya posleoperacionnyh oslozhnenij v urologii [Synthesis of combined fuzzy decision rules for predicting postoperative complications in urology], Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Beriya: Upravlenie, vychislitel'naya tekhnika, informatika. Medicinskoe priborostroenie = Proceedings of the Southwest State University. Series: Management, Computer Engineering, Computer Science. Medical Instrumentation, 2012, no. 2-3, pp. 293-297.

19. Korenevsky N. A., Shutkin A. N., Boytsova E. A., Dmitrieva V. V. Evaluation and management of health status based on G. Rush’s models. Medical Equipment, 2015, no. 6, pp. 37-40.

20. Zadeh L. A. Advances in Fuzzy Mathematics and Engineering Fuzzy Sets and Fuzzy Information-Granulation Theory. Beijing, Beijing Normal University-Press, 2005, pp. 71-78.