Classification models of the adaptive potential of a living system

As a result of the research, a method for classifying the adaptive potential of the human body was developed. The method is based on the use of data obtained by conducting a functional test associated with the Heaviside function, through which a model of the transition process in a living system is obtained. Representing a living system as quasi-linear, based on its impedance model, the spectral characteristics of the living system are obtained, on the basis of which descriptors are formed for the machine learning model. To obtain an impedance model of a living system, a three-phase experiment technique is proposed. The three-phase experiment technique consists of modeling the Heaviside function in the process of performing a bicycle ergometer functional test at three levels of the functional state of the human body. This allows us to calculate descriptors for the three “branches” of the adaptive potential classifier. The adaptive potential classifier includes a driver for constructing a linear impedance model of a living system, a descriptor generator, and a decision-making module. As a linear impedance model of a living system, the amplitude-phase-frequency characteristic of a four-terminal network is used, constructed from the transient characteristic of a model of a living system, and the descriptors are calculated using the Voight impedance model, which is adequate to the experimentally obtained amplitude-phase-frequency characteristic of a model of a living system. The quality indicators of the dichotomous classifier of adaptive potential were assessed on an experimental group of undergraduate and graduate students, divided into two classes using an indicator of the activity of regulatory systems. They showed that the level of true positive and true negative results when classifying unknown examples satisfactorily corresponds to expert estimates. This allows us to recommend it for use in practical medicine, for example, in biotechnical rehabilitation systems, sports medicine, as well as for monitoring the dynamics of the patient’s functional state during treatment.

1. Anokhin P.K. Izbrannye trudy: Kibernetika funktsional'nykh sistem. Moscow: Meditsina; 1998. 400 p. (In Russ.).

2. Apanasenko G.L., Popova L.A., Maglevanyi A.V. Sanologiya. Osnovy upravleniya zdorov'em. Saarbrucken: LAP Lambert Academic Publishing; 2012. 404 p. (In Russ.).

3. Miroshnikov A.V., Shatalova O.V., Efremov M.A., Stadnichenko N.S., Novoselov A.Yu., Pavlenko A.V. Method for classification of the functional state of living systems based on recurrent Voigt models. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel'naya tekhnika, informatika. Meditsinskoe priborostroenie = Proceedings of the Southwest State University. Series: Control, Computer Engineering, Information Science. Medical Equipment Engineering. 2022;12(2):59–75. (In Russ.).

4. Miroshnikov A.V., Stadnichenko N.S., Shatalova O.V., Filist S.A. Biomaterial impedance models for the formation of descriptors in intelligent systems for the diagnosis of infectious diseases. Modelirovanie, optimizatsiya i informatsionnye tekhnologii = Modeling, Optimization and Information Technology. 2020;8(4). (In Russ.). https://doi.org/10.26102/2310-6018/2020.31.4.018.

5. Shatalova O.V. Intellektual'nye sistemy monitoringa meditsinskikh riskov s uchetom bioimpedansnykh issledovanii. Kursk: Southwest State University; 2020. 356 p. (In Russ.).

6. Filist S.A., Shatalova O.V., Bogdanov A.S. Bioimpedance models at the nonlinear volt-ampere characteristic and reversible breakdown of the dielectric component of the biomaterial. Byulleten' sibirskoi meditsiny = Bulletin of Siberian Medicine. 2014;13(4):129–135. (In Russ.). https://doi.org/10.20538/1682-0363-2014-4-129-135.

7. Shutkin A.N. Informative-analytical model of decision-making on professional suitability and the assessment of the health of workers in hazardous industries. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel'naya tekhnika, informatika. Meditsinskoe priborostroenie = Proceedings of the Southwest State University. Series: Control, Computer Engineering, Information Science. Medical Equipment Engineering. 2017;7(2):56–61. (In Russ.).

8. Shutkin A.N. Theoretical and experimental forecasting model of professional suitability of employees of extreme professions. Prikaspiiskii zhurnal: upravlenie i vysokie tekhnologii = The Caspian Journal: Management and High Technologies. 2017;(3):36–45. (In Russ.).

9. Kudryavtsev P.S., Shutkin A.N., Protasova V.V., Filist S.A. Structural functional model for monitoring the influence of control actions on self-organizing systems state. Prikaspiiskii zhurnal: upravlenie i vysokie tekhnologii = The Caspian Journal: Management and High Technologies. 2015;(2):105–118. (In Russ.).

10. Ermakov S.A., Bolgov A.A. Risk assessment using a neuro-fuzzy system. Informatsiya i bezopasnost'. 2022;25(4):583–592. (In Russ.). https://doi.org/10.36622/VSTU.2022.25.4.012.

11. Korenevsky N.A., Filist S.A., Kraskovsky A.B., Afanasyev V.I. The theory of designing of indistinct network expert systems for management of mediko-ecological safety. Nauchnye vedomosti Belgorodskogo gosudarstvennogo universiteta. Seriya: Ekonomika. Informatika. = Belgorod State University. Scientific Bulletin. Series: Economics. Information technologies. 2009;(9):146–151. (In Russ.).

12. Filist S.A., Tomakova R.A., Degtyarev S.V., Rybochkin A.F. Gibridnye intellektual'nye modeli dlya segmentatsii izobrazhenii rentgenogramm grudnoi kletki. Meditsinskaya tekhnika = Biomedical Engineering. 2017;(5):41–45. (In Russ.).

13. Filist S.A., Derkhim A. Kassim K., Rutskoi R.V., Hybrid decisive systems for forecasting of postoperative sequelae at patients with benign gyperplazia of the prostate gland. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta = Proceedings of the Southwest State University. 2013;(5):40–49. (In Russ.).

14. Tomakova R.A., Filist S.A., Naser A.A. Indistinct neural-net technologies for separation of segments with pathologic formations and morphological structures in medical images. Biomeditsinskaya radioelektronika = Biomedicine Radioengineering. 2012;(4):43–50. (In Russ.).

15. Emel'yanov S.G., Rybochkin A.F., Filist S.A., Khalaed A.R. Neirosetevoi reshayushchii modul' dlya issledovaniya zhivykh sistem. Izvestiya Kurskogo gosudarstvennogo tekhnicheskogo universiteta. 2008;(2):77–82. (In Russ.).

16. Zhilin V.V., Filist S.A., Rakhim Khaled Abdul, Shatalova O.V. A method for creating fuzzy neural-network models using the MATLAB package for biomedical applications. Meditsinskaya tekhnika = Biomedical Engineering. 2008;42(2):64–66. https://doi.org/10.1007/s10527-008-9019-y.

17. Pikhlap S.V., Tomakova R.A., Filist S.A. Indistinct neural network structures for segmentation of images of an eyeground. Vestnik Voronezhskogo gosudarstvennogo tekhnicheskogo universiteta = Bulletin of Voronezh State Technical University. 2009;5(4):42–45. (In Russ.).

18. Petrova T.V., Kuzmin A.A., Savinov D.Yu., Serebrovskiy V.V. Distributed autonomous intellectual agents for monitoring and meta-analysis of living systems management effectiveness. Prikaspiiskii zhurnal: upravlenie i vysokie tekhnologii = The Caspian Journal: Management and High Technologies. 2017;(4):61–73. (In Russ.).

19. Zhilin V.V., Filist S.A., Al-Muaalemi Vail Abdulkarim. Hybrid Method for Classification of Biosignals Based on Fuzzy Logic Decisions and Neural Networks Technology. Biomeditsinskaya radioelektronika = Biomedicine Radioengineering. 2009;(5):77–82. (In Russ.).

20. Kurochkin A.G., Zhilin V.V., Surzhikova S.Ye., Filist S.A. Use of hybrid neural network models for mnogoagentny systems of classification in heterogeneous space of informative signs. Prikaspiiskii zhurnal: upravlenie i vysokie tekhnologii = The Caspian Journal: Management and High Technologies. 2015;(3):85–95. (In Russ.).

21. Kurochkin A.G., Protasova V.V., Filist S.A., Shutkin A.N. Neural network model for meta-analysis of medical and ecological data. Neirokomp'yutery: razrabotka, primenenie = Neurocomputers. 2015;(6):48–53. (In Russ.).

22. Filist S.A., Shatalova O.V., Efremov M.A. Hybrid neural network with macrolayers for medical application. Neirokomp'yutery: razrabotka, primenenie = Neurocomputers. 2014;(6):35–69. (In Russ.).

23. Trifonova T.A., Mishchenko N.V., Klimov I.A. Otsenka adaptatsionnogo sostoyaniya studentov. Vladimir: Arkaim; 2016. 94 p. (In Russ.).