A.P. Bryukhovetsky1, A.A. Budaev2

1,2 MPEI Research University (Moscow, Russia)

1 evap15r@rambler.ru, 2 a.budaevkdb@mail.ru

The activity of the heart muscle as a transient process, represented by an electrocardiogram, is evaluated by the first (cardiographic) channel and undergoes a wavelet transformation. Transformation coefficients are visualized and can be considered as elements of the dictionary of the characteristic space defining classes (diseases). In the second (optical) measuring channel, it is proposed to carry out targeted laser diagnostics, the obtained components of Raman spectra, in case of correlation, can be used for a more subtle consideration of the state of the cardiovascular system.

Electrocardiogram is the most important diagnostic tool that allows you to analyze the current state of the human cardiovascular system. Electrocardiogram analysis and interpretation is associated with visual assessment of signal components such as P, QRS, T waves, time segments, and ratios between components. Despite the history of use and widespread use of classical methods for analyzing cardiograms, the interpretation of the results is mainly related to the subjective assessment of the doctor and has a number of limitations on accuracy. There is a need for research aimed at developing methods for improving the accuracy of the diagnosis and, if possible, organizing objective, automated diagnostic tools.

The goal is to develop a method for improving the accuracy of assessing cardiac activity using a two-channel approach to analysis: in the first channel, a cardiogram is obtained, its processing by the method of continuous wavelet transformation to obtain time characteristic parameters, approximating and detailing coefficients. The second measuring channel records the response of the molecules of the laser exposure region to the skin - a Raman spectrum used to study the correlation with the amplitudes of the approximating coefficients.

Programs for wavelet transformation of cardiograms using the Dobeshi db4 wavelet function have been developed. Cardiograms from PhysioBank Databases were taken to test the programs. Based on the study, it was established:

when performing wavelet transformation of cardiograms, the approximating coefficients are most sensitive to changes in the functional capabilities of the cardiovascular system, which makes it possible to build algorithms for recognizing changes; correlation relationships of distribution of approximating coefficients and Raman spectra can be constructed.

The use of the developed technique will significantly expand information about the state of the cardiovascular system and will allow to identify earlier algorithms for detecting deviations from the norm in the activity of the heart.

Bryukhovetsky A.P., Budaev A.A. Electrocardiogram recognition problems // Biomedicine Radioengineering. 2026. V. 29. № 4. P. 81–98. DOI: https:// doi.org/10.18127/ j15604136-202604-08

1. Patent RF №2820398. Sposob issledovaniya raboty serdca pri ispol'zovanii vejvlet-preobrazovaniya elektrokardiogramm i spektroskopii kombinacionnogo rasseyaniya. A. P. Bryuhoveckij, V.A. Ivanov, E.I. Bugaev, R.I. Zaripov, O.V. Chernoyarov. 2023 (In Russian).
2. Goldberger A.L. Goldberger's Clinical Electrocardiography [Electronic resource]: A Simplified Approach. 9th ed. Philadelphia: Elsevier. 2017.
3. Sudakov K.V. Normal'naya fiziologiya. M.: OOO «Medicinskoe informacionnoe agentstvo». 2006. 920 s. (In Russian).
4. Orlov V.N. Rukovodstvo po elektrokardiografii. M.: MIA. 2012 (In Russian).
5. Bozhokin S.V. Matematicheskoe opisanie serdechnogo ritma: Ucheb. posobie. SPb.: Izd-vo SPbGPU. 2005. 80 s. (In Russian).
6. Dobeshi I. Desyat' lekcij po vejvletam: Per. s angl. E.V. Mishchenko. M.: Izhevsk. 2001. 463 s. (In Russian).
7. Svid. o gos. registracii № 2013661588. Metod raspoznavaniya elektrokardiogramm metodom vejvlet-analiza . M.R. Bogdanov, A.V. Zaharov, Yu.Z. Gabidullin. 2013 (In Russian).
8. PhysioBank Databases [Elektronnyj resurs]. URL: http://www.physionet.org, svobodnyj.
9. Yakovlev A.N. Vvedenie v vejvlet preobrazovaniya. Novosibirsk: NGTU. 2003. 104 s. (In Russian).
10. Astaf'eva N.V. Vejvlet-analiz: osnovy teorii i primery primeneniya . Uspekhi fizicheskih nauk. 2012. T. 166. № 111. S. 1145–1170 (In Russian).
11. Ladyaev D.A. Modelirovanie EKG-signalov v srede MATLAB. MATLAB.Exponenta. 2004. S. 1744–1751 (In Russian).
12. Gorelik A.L., Skripnik V. A. Metody raspoznavaniya. M.: Vysshaya shkola. 1977. 222 s. (In Russian).
13. Bryuhoveckij A., Bugaev J., Suetenko A. Image and stage identification during laser remote sounding objects of an organic origin. Proc. SPIE. 2008. V. 7027. 702714. P. 141–155. DOI:10.1117/12.822516.
14. Bryuhoveckij A.P., Bugaev Yu.N., Zhutyaeva T.S. Lazernaya lokaciya i distancionnaya diagnostika materialov ob"ektov i sred. M.: MEI. 2009. 38 s. (In Russian).
15. Zondy InPhotonics dlya spektroskopii kombinacionnogo rasseyaniya: http://www.oemoptic.ru/probes_raman-rip.php (In Russian).
16. Sushchinskij M.M. Kombinacionnoe rasseyanie sveta i stroenie veshchestva. M.: Nauka. 1981. 182 s. (In Russian).
17. Ishchenko A.A. Spektral'nye metody analiza: Ucheb. posobie. M.: Izd-vo MITHT. 2013. 167 s. (In Russian).
18. Kovba V.M. Kvantovo-mekhanicheskaya model' molekuly. M.: MGU. 2011. 37 s. (In Russian).
19. Tugarina D.O., Bryuhoveckij A.P. Lazernaya diagnostika biologicheskih ob"ektov . Radiotekhnicheskie tetradi. 2016. №56. S. 69–73 (In Russian).
20. Shrajner R., F'yuzon R., Kertin D., Morrill T. Identifikaciya organicheskih soedinenij. M.: Mir. 1983. 704 s. (In Russian).
21. Software for Matlab and Octave [Elektronnyj resurs]. URL: https://www.physionet.org/physiotools/matlab/, svobodnyj.
22. Smolencev N.K. Osnovy teorii Vejvletov. Vejvlety v Matlab. M.: DMK Press. 2005. 304 s. (In Russian).
23. Bryukhovetsky A., Makarov A., Ivanov V. Chernoiarova E. On Raman Scattering and the Study of an Object’s Physical and Chemical Parameters: 2025. in book: System Design in Software Engineering. P. 174–197. DOI:10.1007/978-3-031-94770-4_16.