L.A. Manilo1, V.I. Menshikova2, A.P. Nemirko3, Z.M. Yuldashev4, A.A. Tatarinova5, D.A. Stepanov6

1–4 St. Petersburg State Electrotechnical University "LETI" (Saint Petersburg, Russia)
5, 6Federal State Budgetary Institution "NMIC n. a. V. A. Almazov" of the Ministry of Health of the Russian Federation (St. Petersburg, Russia)
1 lmanilo@yandex.ru, 2 vasiliy3me3@yandex.ru, 3 apn-bs@yandex.ru

Sudden cardiac death is a fatal and irreversible process, the underlying cause of which is still not fully understood. The chance to save a person who is in a state of clinical death is minimal, but it exists. This work is devoted to predicting the occurrence of sudden cardiac death (SCD) based on heart rate variability (HRV). The development of new forecasting methods will make it possible to identify such a dangerous condition long before the initial symptoms appear. Timely measures taken increase the chance of saving a person.

The aim of the work is to evaluate the possibility of predicting SCD using nonlinear heart rate analysis.

The paper examines an approach to predicting SCD based on the analysis of the following HRV indicators: standard deviation of RR intervals, Shannon entropy, approximated entropy, and multiscale variation. The parameters were estimated based on 10-minute fragments of the signal selected at different time intervals from the moment of occurrence of the SCD (from 10 minutes to 70 minutes). Using two certified PhysioNet ECG record databases, experiments were conducted to recognize two conditions: the norm and the condition before the onset of SCD. The classification was carried out by the Fischer linear discriminant method. The recognition efficiency indicators for different segments of the rhythmogram, which differ in delay time relative to the moment of SCD, are evaluated.

The results obtained are important for expanding the functions of ECG monitoring systems in order to prevent possible critical conditions of the patient.

Manilo L.A., Menshikova V.I., Nemirko A.P., Yuldashev Z.M., Tatarinova A.A., Stepanov D.A. Prediction of sudden cardiac death by heart rate variability indicators. Biomedicine Radioengineering. 2025. V. 28. № 2. P. 5–13. DOI: https:// doi.org/10.18127/j15604136-202502-01 (In Russian)

1. Bojcov S.A., Linchak R.M., Nedbajkin A.M., Semencova E.V., Yusova I.A., Strukova I.V. Epidemiologiya vnezapnoj serdechnoj smerti: chto my znaem segodnya?. Klinicheskaya praktika. 2014. № 4 (20). S. 13–15 (In Russian).
2. Bokeriya O.L., Ahobekov A.A. Vnezapnaya serdechnaya smert': mekhanizmy vozniknoveniya i stratifikaciya riska. Ann. aritm. 2012. № 3. S. 6–7 (In Russian).
3. Sudden Cardiac Death Holter Database. PhysioNet. URL: https://physionet.org/content/sddb/1.0.0/
4. MIT-BIH Normal Sinus Rhythm Database. PhysioNet. URL: https://physionet.org/content/nsrdb/1.0.0/
5. Yang, J., Sun, Z., Zhu, W. et al. Intelligent prediction of sudden cardiac death based on multi-domain feature fusion of heart rate variability signals. EURASIP J. Adv. Signal Process. 2023. 32.
6. Stepanov D.A., Tatarinova A.A. EKG-stratifikaciya riska vnezapnoj serdechnoj smerti i zhizneugrozhayushchih zheludochkovyh aritmij. Vestnik aritmologii. 2024 № 31(1). S. 1–4 (In Russian).
7. Van Hoogenhuyze D., Martin G., Weiss J., Schaad J., Singer D. Spectrum of heart rate variability. Proc. Comput. Cardiol. 1989. P. 1–10.
8. Nemirko A.P., Manilo L.A., Kalinichenko A.N. Matematicheskij analiz biomedicinskih signalov i dannyh. M.: Fizmatlit. 2017. 248 s. (In Russian).
9. Costa M.D., Goldberger A.L., Peng C.-K. Multiscale entropy analysis of biological signals. Phys Rev. E. Stat. Nonlin. Soft Matter Phys. 2005. № 71.
10. Shi M., He H., Geng W., et al. Early detection of sudden cardiac death by using ensemble empirical mode decomposition-based entropy and classical linear features from heart rate variability signals. Frontires in Physiology. 2020. № 11. С. 118.