Determination of sleep spindles with time series approach

Year 2025, Volume: 67 Issue: 2, 107 - 124, 24.12.2025

Levent Özbek , Sinan Yetkin , Asuhan Zupan , Hakkı Gökhan İlk , Nihan Nur Ceran Gürlek

https://doi.org/10.33769/aupse.1613459

https://izlik.org/JA69SK23PU

Abstract

The graphoelements of electroencephalography (EEG), which form the sleep structure, are one of the important research areas of electrophysiology and cognitive science. Sleep structuring is done by experienced sleep experts according to standardized rules. It is performed by visual evaluation according to the graphoelements in the EEG recorded during the night's sleep. This process is subjective, requires time and susceptible for human errors. Among these graphoelements, sleep spindles are the characteristic EEG wave activities. They belong to the non-rapid eye movement (NREM) N2 sleep period. They play a role in sleep structure and cognitive functions. Therefore, it has recently been the subject of research as a clinical determining factor in neuropsychiatric disorders. The initial stage of detection and then further analysis of the waveforms of sleep spindles and the determination of the time and duration of the spindles have gained importance in research. However, since sleep spindles have transient properties compared to background EEG signals, they are difficult to visually analyze. In this study, the sleep EEG was modeled with the time-varying parameter Autoregressive AR (9) to determine the sleep spindles encountered in the sleep EEG. In dynamic linear models given with AR (9), the parameter vector is accepted as a random walk process. Therefore, it can be written as a state space model. By using the adaptive Kalman filter (AKF), the model parameters that can be easily estimated are the time variable parameter and the variance of the white noise process. The focal point of our analysis revealed that the estimated values showed significant changes in the parts with the spindles. We therefore propose that the estimated values of the time varying variance of the white noiseprocess will provide a valuable decision support to sleep experts working in the field of sleep scoring. It can be considered that the method discussed in our study can be combined with existing automatic scoring algorithms and can also be used as a control mechanism in manual scoring for inexperienced sleep scorers.

Keywords

EEG , sleep , sleep spindle , time series , adaptive Kalman filter.

References

• Berger, H., Ueber das elektroenkephalogramm des Menschen, J. Psychol. Neurol., 40 (1930), 160-179.
• Rechtschaffen, A., Kales, A., Brain Information Service, and Brain Research Institute UoC, A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects, US Government Printing Office, US Public Health Service, 1968.
• Yücelbaş, C., Yücelbaş, Ş., Özşen, S., et al., Automatic detection of sleep spindles with the use of STFT, EMD and DWT methods, Neural Comput. Applic., 29 (2018),17-33, https://doi.org/10.1007/s00521-016-2445-y.
• Olbrich, E., Achermann, P., Oscillatory events in the human sleep EEG- detection and properties, Neurocomputing, 58 (2004), 129-135, https://doi.org/10.1016/j.neucom. 2004.01.033.
• Kaplan, A., Röschke, J., Darkhovsky, B., Fell, J., Macrostructural EEG characterization based on nonparametric change point segmentation: application to sleep analysis, J. Neurosci. Methods, 106 (2001), 81-90, https://doi.org/10.1016/s0165-0270(01)00331-4.
• Olbrich, E., Achermann, P., Meier, P. F., Dynamics of human sleep EEG, Neurocomputing, 52 (2003), 857-862, https://doi.org/10.1016/S0925-2312(02)00816-0.
• Pardey, J., Robert, S., Tarassenko, L. A., Review of parametric modelling techniques for EEG analysis, Med. Eng. Phys., 18 (1) (1995), 2-11, https://doi.org/10.1016/13504533(95)00024-0.
• Perumalsamy, V., Sankaranarayanan, S., Rajamony, S., Sleep spindles detection from human sleep EEG signals using autoregressive (AR) model: A surrogate data approach, J. Biomed. Sci. Eng., 2 (2009), 294-303, https://doi.org/10.4236/jbise.2009.25044.
• Atyabi, A., Chis, F., Naples, A., Mixture of autoregressive modeling orders and its implication on single trial EEG classification, 15 (2016), 164-180, https://doi.org/10.1016/j.eswa.2016.08.044.
• Ghimatgar, H., Kazemi, K., Helfroush, M. S., Aarabi, A., An automatic single-channel EEG-based sleep stage scoring method based on hidden Markov model, J. Neurosci. Methods, 324 (2019), 108320, https://doi.org/10.1016/j.jneumeth.2019.108320.
• Altan, A., Hacıoğlu, R., Model predictive control of three-axis gimbal system mounted on UAV for real-time target tracking under external disturbances, Mech. Syst. Signal Process., 138 (2020), 106548, https://doi.org/10.1016/j.ymssp.2019.106548.
• Karasu, S., Altan, A., Recognition model for solar radiation time series based on random forest with feature selection approach, 2019 11th International Conference on Electrical and Electronics Engineering (ELECO), (2019), 8-11, https://doi.org/10.23919/ELECO47770.2019.8990664.
• Özbek, L., Sütçigil, L., Aydın, H., Yetkin, S., Özgen, F., A statistical overview on sleep scoring, Commun. Fac. Sci. Univ. Ank. Series A2-A3, 62 (2020), 1-8, https://doi.org/10.33769/aupse.745571.
• Altan, A., Karasu, A, Bekiros, S., Digital currency forecasting with chaotic meta heuristic bio-inspired signal processing techniques, Chaos Soliton Fract., 126 (2019), https://doi.org/10.1016/j.chaos.2019.07.011325-336.
• Athanasios, T., Clifford, G. D., Stage-independent, single lead EEG sleep spindle detection using the continuous wavelet transform and local weighted smoothing, Front. Human Neurosci., 9 (2015), 181, https://doi.org/10.3389/fnhum.2015.00181.
• Mensen, A., Poryazova, R., Huber, R., et al., Individual spindle detection and analysis in high-density recordings across the night and in thalamic stroke, Sci. Rep., 8 (2018), 17885, https://doi.org/10.1038/s41598-018-36327-x.
• Zriouil, I., Regragui, F., Hamzaoui, E. M., Himmi, M. M., Mounach, J., Detection of sleep spindles using frequency components tracking in raw EEG data, Proceedings of the Mediterranean Conference on Information and Communication Technologies 2015, Lecture Notes in Electrical Engineering, (2016), 380, https://doi.org/10.1007/978-3-31930301-7_46.
• Hernandez-Pereira, E., Fernandez-Varela, I., Moret-Bonillo, V., A comparison of performance of sleep spindle classification methods using wavelets, Innovation in Medicine and Healthcare 2016, Smart Innovation, Systems and Technologies, (2016), 60, https://doi.org/10.1007/978-3-319-39687-3_6.
• Hassan, A. R., Bhuiyan, M. I. H., Automated identification of sleep states from EEG signals by means of ensemble empirical mode decomposition and random under sampling boosting, Comput. Methods Prog. Biomed., 140 (2017), 201-210, https://doi.org/10.1016/j.cmpb.2016.12.015.
• Hassan, A. R., Bhuiyan M. I. H., An automated method for sleep staging from EEG signals using normal inverse Gaussian parameters and adaptive boosting, Neurocomputing, 219 (2017), 76-87, https://doi.org/10.1016/j.neucom.2016.09.011.
• Aboalayon, K. A. I., Faezipour, M., Almuhammadi, W. S. and Moslehpour, S., Sleep stage classification using EEG signal analysis: A comprehensive survey and new investigation, Entropy, 18 (9) (2016), 272, https://doi.org/10.3390/e18090272.
• Purcell, S. M., Manoach, D. S., Demanuele, C., Cade, B. E., Mariani, S., Cox, R., Panagiotaropoulou, G., Saxena, R., Pan, J. Q., Smoller, J. W., Redline, S., Stickgold, R., Characterizing sleep spindles in 11,630 individuals from the National Sleep Research Resource, Nat. Commun., 8 (2017), 15930, https://doi.org/10.1038/ncomms15930.
• Clawson, B. C., Durkin, J., Aton, S. J., Form and function of sleep spindles across the lifespan, Neural. Plast., (2016) 6936381, https://doi.org/10.1155/2016/6936381.
• Iber, C., Ancoli-Israel, S., Chesson, A., Quan, S., The AASM manual for the scoring of sleep and associated events: Rules, terminology and technical specifications. American Academy of Sleep Medicine, 2007.
• Warby, S., Wendt, L., Welinder, P., et al., Sleep-spindle detection: crowdsourcing and evaluating performance of experts, non-experts and automated methods, Nat. Methods, 11 (4) (2014), 385-392, https://doi.org/10.1038/nmeth.2855.
• Wallant, D. C., Maquet, P., Phillips, C., Sleep spindles as an electrographic element: Description and automatic detection methods, Neural Plast., (2016), 6783812, https://doi.org/10.1155/2016/6783812.
• Christian, O., Nielsen, T., Automatic sleep spindle detection: benchmarking with fine temporal resolution using open science tools, Front. Human Neurosci., 9 (2015), 353.
• Chen, E., Chen, D., Zhang, L., Tang, Y., Li, X., Automated sleep spindle detection with mixed EEG features, Biomed. Signal Process. Cont., 70 (2021), 103026.
• Kaulen, L., Schwabedal, J. T. C., Schneider, J. et al., Advanced sleep spindle identification with neural networks, Sci. Rep., 12 (2022), 7686.
• Li, Y., Song, K., Zhang, Y., Karray, F., Method and system for automated detection of sleep spindles using a single EEG channels based TEO and EMD, Expert. Syst. Appl., 249 B, (2024), 123661, https://doi.org/10.1016/j.eswa.2024.123661.
• Kalman, R. E., A new approach to linear filtering and prediction problems, J. Basic Eng., 82 (1960), 35-45, https://doi.org/10.1115/1.3662552.
• Ljung, L., Söderström, T., Theory and Practice of Recursive Identification, The MIT Press, Cambridge, 1983.
• Goodwin, G. C., Sin K. S., Adaptive Filtering Prediction and Control, Prentice Hall, Englewood Cliffs, 1985.
• Anderson, B. D. O., Moore, J. B., Optimal Filtering, Prentice Hall, Englewood Cliffs, 1979.
• Özbek, L., Kalman Filtresi, Akademisyen Yayınevi, Ankara, 2017.
• Arnold, M., Miltner, W. H. R., Witte, Bauer, H. R. and Braun, C., Adaptive AR modeling of nonstationary time series by means of Kalman filtering, IEEE Trans. Biomed. Eng., 45, (1998), 553.
• C. Anderson, Classification of Electroencephalogram (EEG) Signals, http://www. cs. colostate.edu/anderson/res/eeg/.
• Özbek, L., Aliev, F. A., Comments on adaptive fading Kalman filter with an application, Automatica, 34 (12) (1998), 1663-1664.
• Özbek, L., Efe, M., An adaptive extended Kalman filter with application to compartment models, Commun. Stat. Simul. Comput., 33 (1) (2004), 145-158, https://doi.org/10.1081/SAC-120028438.
• Isaksson, A., On system identification in one and two dimensions with signal processing applications, Linköping Stud. Sci. Tech., 196 (1988), 250.