Method for Classification of the Functional State of Living Systems Based on Recurrent Voigt Models

Purpose of research is development of a recurrent algorithm for optimizing Voight models of biomaterial, which allows obtaining sets of descriptors for classifiers of the functional state of living systems in intelligent decision support systems for predicting and diagnosing socially significant diseases.

Methods. The essence of the proposed method lies in the use of a recurrent procedure for comparing the biomaterial impedance model obtained on the basis of series-connected Voigt links and the results of experimental studies. In the process of optimizing the Voigt model of the biomaterial, according to the proposed method, frequency pools are formed, at which, according to the results of the recurrent procedure, the optimal Voigt models are found by the number of links in the model. Then, according to the results of the integral error, the optimal number of model links is selected, corresponding to this number, the optimal pool of frequencies. The parameters of the Voight model make it possible to form descriptors for multimodal classifiers of the functional state of living systems and parameters of model links.

Results. ^s an example for testing the operation of the classifier built on the basis of the Voight model optimization algorithm, a group of patients with pneumonia with a clear diagnosis was taken. To obtain raw bioimpedance analysis data, an electrode belt was put on the chest of patients and impedance diagrams were determined corresponding to a certain combination of electrodes. The quality indicators of various classifier models reached 0,78% and did not fall below 0,62%

Conclusion. It is shown that the capabilities of multi-frequency probing and neural network models of multimodal classifiers make it possible to obtain new decision rules for diagnosing pathological conditions of the body (cardiovascular, infectious and oncological diseases).

1. Filist S. A., Shutkin A. N., Shkatova E. S., Degtyarev S. V., Savinov D. Yu. Model' formirovaniya funktsional'nyh sistem s uchetom menedzhmenta adaptatsionnogo potentsiala [Model of the formation of functional systems taking into account the management of adaptive potential]. Biotekhnosfera = Biotechnosphere, 2018, no. 1(55), pp. 32-37.

2. Kiselev A. V., Petrova T. V., Degtyarev S. V., Rybochkin A. F., Filist S. A., Shatalova O. V., Mishustin V. N. Neyrosetevye moduli s virtual'nymi potokami dlya klassifikatsii i prognozirovaniya funktsional'nogo sostoyaniya slozhnyh sistem [Neural network modules with virtual flows for classifying and predicting the functional state of complex systems]. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta = Proceedings of Southwest State University, 2018, no. 4(79), pp. 123-134.

3. Shatalova O. V. Intellektual'nye sistemy monitoringa meditsinskih riskov s uchetom bioimpedansnyh issledovaniy [Intelligent monitoring systems for medical risks taking into account bioimpedance studies]. Kursk, Southwest State University Publ., 2020. 356 p.

4. Popechitelev E. P., Filist S. A. Sposoby i modeli identifikatsii biomaterialov na osnove analiza mnogochastotnogo impedansa [Methods and models for identification of biomaterials based on multifrequency impedance analysis]. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel'naya tekhnika. Meditsinskoe priborostroenie = Proceedings of Southwest State University. Series: Control, Computer Engineering, Information Science. Medical Instruments Engineering, 2011, no. 1, pp. 74-80.

5. Shatalova O. V. Iteratsionnaya mnogoparametricheskaya model' bioimpedansa v eksperimentah in vivo [Iterative multiparameter bioimpedance model in in vivo experiments]. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel'naya tekhnika. Meditsinskoe priborostroenie = Proceedings of Southwest State university. Series Control, Computer Engineering, Information Science. Medical Instruments Engineering, 2019, no. 9(1), pp. 26-38.

6. Kassim K. D. A., Klyuchikov I. A., Shatalova O. V., Yaa Z. D. Parametricheskie modeli bioimpedansa dlya identifikatsii funktsional'nogo sostoyaniya zhivoy sistemy [Parametric bioimpedance models for identifying the functional state of a living system]. Biomeditsinskaya radioelektronika = Biomedical Radioelectronics, 2012, no. 4, pp. 50-56.

7. Filist S. A., Kuzmin A. A., Kuzmina M. N. Biotekhnicheskaya sistema dlya kontrolya impedansa biomaterialov v eksperimentah in vivo [Biotechnical system for controlling the impedance of biomaterials in in vivo experiments]. Biomeditsinskaya radioelektronika = Biomedical Radioelectronics. 2014, no. 9, pp. 38-42.

8. Miroshnikov A. V., Stadnichenko N. S., Shatalova O. V., Filist S. A. Modeli biomateriala dlya formirovaniya deskriptorov v intellektual'nyh sistemah diagnostiki infektsionnyh zabolevaniy [Biomaterial models for the formation of descriptors in intelligent systems for diagnosing infectious diseases]. Modelirovanie, optimizatsiya i informatsionnye tekhnologii. Nauchnyy zhurnal = Modeling, Optimization and Information Technology. Science Magazine, 2020, vol. 7, no. 1, pp. 1-14. https://doi.org/10.26102/2310-6018/2020.31.4.018

9. Miroshnikov A. V., O Shatalova. V., Zhilin V. V. Biomaterial impedance model for medical risk classifiers in in vivo experiments. Journal of Physics: Conference Series. International Scientific Conference Artificial Intelligence and Digital Technologies in Technical Systems, 2020. Р. 012045.

10. Miroshnikov A. V., Kiselev A. V., Shatalova O. V., Krupchatnikov R. A. Iterative models of bioimpedance in intelligent systems for early diagnosis of infectious diseases. CEUR Workshop Proceedings. ITIDMS 2021 - Proceedings of the International Scientific and Practical Conference "Information Technologies and Intelligent Decision Making Systems", 2021, p. 25.

11. Miroshnikov A. V., Shatalova O. V., Stadnichenko N. S., Shulga L. V. Klassifikatsii biologicheskih ob"ektov na osnove mnogomernogo bioimpedansnogo analiza [Classification of biological objects based on multidimensional bioimpedance analysis]. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya Upravlenie, vychislitel'naya tekhnika. Meditsinskoe priborostroenie = Proceedings of Southwest State University. Series: Control, Computer Engineering, Information Science. Medical Instruments Engineering, 2020, vol. 10, no. 3/4, pp. 29-49.

12. Buyanova E. S., Emel'yanova Y. V. Impedansnaya spektroskopiya elektroliticheskih materialov [Impedance spectroscopy of electrolytic materials]. Ekaterinburg, Ural State University Publ., 2008. 70 p.

13. Filist S. A., Shatalova O. V., Protasova Z. U., Stadnichenko N. S. Sposob klassifikatsii biologicheskih ob"ektov na osnove mnogomernogo bioimpedansnogo analiza i ustroystvo dlya ego realizatsii [Method for solving biological objects based on multidimensional bioimpedance analysis and devices for implementation]. Patent RF, no. 2752594, 2021.

14. Shatalova O. V., Filist S. A., Protasova Z. U., Korenevskiy N. A., Al-Kasasbeh R. T., eds. Application of Fuzzy Neural Model and Current-Voltage Analysis of Biologically Active Points for Prediction Post-Surgery Risks. Computer Method in Biomedical Engineering, 2021, vol. 24, pp. 1504-1516. https://doi.org/10.1080/10255842.2021.1895128.

15. Simchera V. M. Metody mnogomernogo analiza statisticheskih dannyh [Methods for Multivariate Analysis of Statistical Data: A Tutorial]. Moscow, Finance and statistics Publ., 2008. 400 p.

16. Filist S. A., Aleksenko V. A., Kassim Kabus. Gibridnye informatsionnye tekhnologii po ekspress-diagnostike infektsionnyh zabolevaniy na osnove mnogochastotnogo analiza passivnyh svoystv biotkaney [Hybrid information technologies for express diagnostics of infectious diseases based on multifrequency analysis of passive properties of biological tissues]. Izvestiya Yuzhnogo Federal'nogo universiteta. Tekhnicheskie nauki. Tematicheskiy vypusk. Meditsinskie informatsionnye sistemy = Proceedings Southern Federal University. Technical Science. Thematic Issue. Medical Information Systems, 2010, no. 8(109), pp. 12-17.

17. Surzhikova S. E., Filist S. A., Zhilin V. V., Kurochkin A. G. Ispol'zovanie gibridnyh neyrosetevyh modeley dlya mnogoagentnyh sistem klassifikatsii v geterogennom prostranstve informativnyh priznakov [Using hybrid neural network models for multi-agent classification systems in a heterogeneous space of informative features]. Prikaspijskij zhurnal: upravlenie i vysokie tekhnologii = Caspian Journal: Management and High Technologies, 2015, no. 3, pp. 85-95.

18. Filist S. A., Shatalova O. V., Efremov M. A. Gibridnaya neyronnaya set' s makrosloyami dlya meditsinskih prilozheniy [Hybrid neural network with macro layers for medical applications]. Neyrokomp'yutery. Razrabotka i primenenie = Neurocomputers. Development and Application, 2014, no. 6, pp. 35-39.

19. Khatatneh K., Filist S., Al-Kasasbeh R. T., Aikeyeva A. A., Namazov M., Shatalova O., Shaqadan A., Miroshnikov A. Hybrid neural networks with virtual flows in in medical risk classifiers. Journal of Intelligent & Fuzzy Systems, 2022, no. 43, pp. 1621-1632. https://doi.org/10.3233/JIFS-212617

20. Mednikov D. A., Protasova Z. U., Shatalova O. V., Serebrovskiy A. V. [Decision support systems in the cardiology field. Review]. Mediko-ekologicheskie informatsionnye tekhnologii - 2021. Sbornik nauchnyh statey po materialam XXIV Mezhdunarodnoy nauchno-tekhnicheskoy konferentsii (20-21 maya 2021 g.) [Medical and environmental information technologies - 2021. Collection of scientific articles based on the materials of the XXIV International Scientific and Technical Conference (May 20-21, 2021)]. Kursk, Southwest State University Publ., 2021, pp. 178-188. (In Russ).