Duc Ha Doan Post-graduate Student, Department of Computational and Applied Mathematics, Ryazan State Radio Engineering University
S.Yu. Zhuleva Senior Lecturer, Department of Computational and Applied Mathematics, Ryazan State Radio Engineering University
A.V. Kroshilin Dr.Sc. (Eng.), Associate Professor, Professor of Department of Computational and Applied Mathematics, Ryazan State Radio Engineering University
S.V. Kroshilina Ph.D. (Eng.), Associate Professor, Department of Computational and Applied Mathematics, Ryazan State Radio Engineering University
V.V. Tishkina Post-graduate Student, Department «Computational and applied mathematics», Ryazan state radio engineering university
The article describes the methodology of implementation of the modified algorithm of fuzzy data clustering the medical-technological process in systems for medical purposes to support decision-making in conditions of uncertainty in the selection of the variant of disease in assessment of the patient's health status based on various sources of information. Given: the General scheme of the method of fuzzy clustering, the main group of health indicators used for analysis and fuzzy cluster of stilizacii. In the medical subject area cluster analysis is useful when you need to categorize large amounts of information, for example, uses clustering of disease, course of illness, patterns of treatment of diseases, symptoms of diseases, taxonomy of the drugs, patient populations, etc.
The proposed method of fuzzy clustering based on fuzzy equivalence relations arising from properties of the investigated data the medical-technological process and without the use of additional information about clusters that do not depend on the cluster shape and have centers different from traditional methods. The algorithm of fuzzy clustering allows data productively in the study to identify clusters of various shapes, the relations between the elements while solving the problem of fuzzy clustering solves the problem of choosing the variant of disease and will make its accuracy by 8-19 %.
A software package of decision support based on fuzzy logic "Expert 4. Statistical data processing of medical and technological processes by fuzzy clustering" was successfully out-dren and is used in State health care institution of Tver region "Tver regional clinical TB dispensary", which allows you to: implement efficient data processing of medical and technological process for intelligent decision support, to implement the construction of various subject areas of medical records, to carry out simulation of medical technological process, the development of effective management of medicines-cal material resources; formation of various types of reports and notifications, effective analysis of statistical information. Testing and expert evaluation show that the proposed system recommendations are accurate from a medical point of view, and are of practical importance for solving the problem of choosing the variant of disease.
- Dolzhenko E.N., Zhuleva S.Ju., Kroshilin A.V., Py'l'kin A.N. Podderzhka prinyatiya reshenij na osnove nechetkoj logiki v sistemax mediczinskogo naznacheniya // Biomediczinskaya radioe'lektronika. 2015. № 5. S. 62–68.
- Zhuleva S.Ju., Kroshilin A.V., Kroshilina S.V. Predstavlenie modeli predmetnoj oblasti na osnove semanticheskoj seti v sistemax mediczinskogo naznacheniya // Dinamika slozhny'x sistem. 2015. T. 9. № 4. S. 29–33.
- Kroshilin A.V., Cy'bikova E'.B., Dolzhenko E.N., Vinogradova L.I., Sabgajda T.P. Oczenka faktorov riska, vliyayushhix na rezul'taty' lecheniya vpervy'e vy'yavlenny'x bol'ny'x tuberkulezom legkix // Nauchno-prakticheskij zhurnal «Tuberkulez i bolezni legkix». 2014. № 12. S. 40–46.
- Doan D.H., Tishkina V.V., Kroshilina S.V., Kroshilin A.V., Pylkin A.N. Support of decision-making in the conditions of uncertainty of different types (02006) / Published online: 25 March 2016 / DOI: http://dx.doi.org/10.1051/itmconf/20160602006 / 6th Seminar on Industrial Control Systems: Analysis, Modeling and Computation, ITM Web of Conferences. 2016. V. 6. Moscow, Russia, February 25–26, E.V. Nikulchev and E.I. Veremey (Eds.)
- Svidetel'stvo o gosudarstvennoj registraczii programm dlya bazy' danny'x № 2016618420, Programmny'j kompleks podderzhki prinyatiya reshenij na osnove nechetkoj logiki «E'kspert 4. Obrabotka statisticheskix danny'x mediko-texnologicheskix proczessov metodom nechetkoj klasterizaczii» / S.V. Kroshilina, A.V. Kroshilin, A.N. Py'l'kin, D.X. Doan. Ver. 4.04, zaregistrirovana v Reestre programm dlya E'VM 28.07.2016 g.