I.I. Titova1

1 St. Petersburg State Electrotechnical University (St. Petersburg, Russia)
1 iititova@etu.ru

The development of modern monitoring methods is necessitated by the profound global burden of epilepsy, which affects approximately 50 million people worldwide and is associated with significantly elevated mortality rates due to accidents, trauma, and sudden unexpected death in epilepsy (SUDEP). The cornerstone of epilepsy management – unpredictability of seizures – creates constant psychological burden and limits patients' daily activities. While video-electroencephalography monitoring remains the gold standard for diagnosis, it is inherently unsuitable for long-term continuous monitoring in ambulatory settings due to its stationary nature and discomfort. Existing wearable solutions are predominantly unimodal (either accelerometry-based or heart rate monitoring), demonstrating limited sensitivity for non-motor or focal seizures and high false alarm rates.

The aim of the study is to develop a scientifically grounded method for multimodal analysis of electrophysiological signals (electroencephalogram, photoplethysmogram, electrocardiogram) enabling early detection of epileptic seizure episodes in wearable monitoring conditions.

A systematic analysis of seizure pathophysiology was conducted, confirming the causal link between cortical discharges and autonomic responses, with autonomic changes often preceding electroencephalographic manifestations. A set of informative features for each signal type was identified (spectral EEG components, heart rate variability parameters, pulse wave morphology), suitable for real-time automated processing. A review of current wearable technologies revealed a lack of integrated solutions combining EEG and physiological sensors in a comfortable form factor. Based on these findings, a concept for a hybrid wearable device (a headband with dry EEG electrodes and a reflectance PPG sensor) was formulated, along with architectural requirements.

The use of developed method is firmly grounded in a theoretical framework that substantiates both the feasibility and clinical relevance of multimodal seizure detection. This framework not only supplies the algorithmic basis but also delineates precise architectural specifications for a hybrid wearable device—specifically, a headband configuration that integrates dry EEG electrodes with a reflectance PPG sensor positioned at the temple, augmented by an inertial measurement unit for motion artifact correction. Consequently, this research establishes the foundation for a domestic hardware-software platform aimed at personalized epilepsy monitoring. Such a system holds the potential to enhance detection accuracy, mitigate the risk of SUDEP, and improve patients' quality of life by enabling timely intervention and alleviating psychological burden.

Titova I.I. Epileptic Seizure Monitoring Method Based on Multimodal Analysis of Human Electrophysiological Parameters. Biomedicine Radioengineering. 2026. V. 29. № 3. P. 131–135. DOI: https:// doi.org/10.18127/ j15604136-202603-23 (In Russian)

1. Fisher R.S. et al. Epileptic seizures and epilepsy: definitions proposed by the International League Against Epilepsy (ILAE). Epilepsia. 2005. V. 46. № 4. P. 470–472.
2. Ryvlin P. et al. Incidence and mechanisms of cardiorespiratory arrests in epilepsy monitoring units (MORTEMUS). The Lancet Neurology. 2013. V. 12. № 10. P. 966–977.
3. Baumgartner C., Koren J.P., Rothmayer M. Automatic Computer-Based Detection of Epileptic Seizures. Frontiers in Neurology. 2018. V. 9. P. 639.
4. Onorati F. et al. Multicenter clinical assessment of improved wearable multimodal convulsive seizure detectors. Epilepsia. 2017. V. 58. № 11. P. 1870–1879.
5. Beniczky S. et al. Automated seizure detection with noninvasive wearable devices: A systematic review and meta-analysis. Epilepsia. 2022. V. 63. № 8. P. 1930–1941.
6. Guerrero-Aranda A. et al. Bridging the Gap in Tonic Seizure Detection: A Systematic Review and Meta-Analysis of Automatic Detection Systems. Journal of Clinical Neurophysiology. 2025. V. Publish Ahead of Print.
7. Devinsky O. et al. Epilepsy and the autonomic nervous system. Epilepsy & Behavior. 2019. V. 101. Pt B. P. 106511.
8. Leutmezer F. et al. Electrocardiographic changes at the onset of epileptic seizures. Epilepsia. 2003. V. 44. № 3. P. 348–354.
9. Zijlmans M. et al. Heart rate changes and ECG abnormalities during epileptic seizures: prevalence and definition of an objective clinical sign. Epilepsia. 2002. V. 43. № 8. P. 847–854.
10. Leung H. et al. Ictal tachycardia: The head–heart connection. Seizure. 2014. V. 23. № 7. P. 496–505.
11. Cheung F., Pearl P.L., Stamoulis C. Novel Seizure Biomarkers in Continuous Electrocardiograms from Pediatric Epilepsy Patients. Conference Proceedings. 2026.
12. Persson H. et al. Changes in heart rate variability during tonic-clonic seizures. Epilepsy Research. 2012. V. 100. № 1–2. P. 109–113.