Abstract
Sleep apnea (SA) is a sleep-related breathing disorder characterized by breathing pauses during sleep. A person’s sleep schedule is significantly influenced by that person’s hectic lifestyle, which may include unhealthy eating habits and their line of work. Polysomnographic (PSG) sleep studies examine sleep-related disorders by recording various biosignals from the human body. However, SA classification methods could be more robust in terms of performance because they rely on feature-engineering strategies or employ a particular signal from PSG recording for diagnosis. This study aims to classify the severity of SA according to the apnea–hypopnea index (AHI) into normal, mild, and moderate-to-severe groups using oxygen saturation (SpO2), electroencephalogram (EEG), and electrocardiogram (ECG) signals. The proposed deep neural network (DNN)-bidirectional long short-term memory (DNN-BiLSTM) framework addresses the issue of low detection accuracy in analysis. The DNN-BiLSTM approach employs features extracted from a multiscale dilation attention 1D convolutional neural network (MSDA-1DCNN) as input for detection and classification purposes. The MSDA-1DCNN network extracts deep features from processed SpO2, EEG, and ECG signals. The developed firefly combined electric fish optimization (FCEFO) algorithm improves performance by optimizing the hidden neuron count of the DNN and the learning rate of the BiLSTM framework. The performance measures proved the effectiveness of the model over conventional machine and deep learning approaches. With the integration of deep features, the proposed DNN-BiLSTM model provides enhanced performance in terms of accuracy and precision. Thus, the proposed approach is progressive and can be used for medical diagnosis.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00521-024-09889-3/MediaObjects/521_2024_9889_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00521-024-09889-3/MediaObjects/521_2024_9889_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00521-024-09889-3/MediaObjects/521_2024_9889_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00521-024-09889-3/MediaObjects/521_2024_9889_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00521-024-09889-3/MediaObjects/521_2024_9889_Fig5_HTML.png)
Similar content being viewed by others
Data availability
Data supporting the findings of this study are available in [Physionet] at https://doi.org/https://doi.org/10.13026/C23K5S.
References
Quan SF, Gillin JC et al (1999) Sleep-related breathing disorders in adults: recommendationsfor syndrome definition and measurement techniques in clinical research. Editorials Sleep 22:662–689
Kato M, Adachi T, Koshin Y et al (2009) Obstructive sleep apnea and cardiovascular disease. Circ J 73(8):1363–1370. https://doi.org/10.1253/circj.cj-09-0364
Young T, Palta M, Dempsey J et al (1993) The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 328:1230–1235
Kline LR (2023) Clinical-presentation-and-diagnosis-of-obstructive-sleep-apnea-in-adults. UpToDateInc. https://www.uptodate.com. Accessed 13 Jan 2023
Iber C, Ancoli Israel S et al (2007) The AASM manual for the scoring of sleep and associated events: Rules, terminology and technical specifications. American Academy of Sleep Medicine, Westchester
Hafezi M et al (2020) Sleep apnea severity estimation from tracheal movements using deep learning model. IEEE Access 8:22641–22649
Vaquerizo F et al (2021) A convolutional neural network architecture to enhance oximetry ability to diagnose pediatric obstructive sleep apnea. IEEE J Biomed Health Inform 25(8):2906–2916
Hui Yu, Liu D, Zhao J et al (2022) A sleep apnea-hypopnea syndrome automatic detection and subtype classification method based on LSTM-CNN. Biomed Signal Process Control. https://doi.org/10.1016/j.bspc.2021.103240
Mendez MO, Bianchi AM et al (2009) Sleep apnea screening by autoregressive models from a single ECG lead. IEEE Trans Biomed Eng 56:2838–2850
Khandoker AH, Gubbi J et al (2009) Automated scoring of obstructive sleep apnea and hypopneaevents using short-term electrocardiogram recordings. Inf Technol Biomed IEEE Trans 13:1057–1067
Penzel T, McNames J, De Chazal P et al (2002) Systematic comparison of different algorithms for apnoea detection based on electrocardiogram recordings. Med Biol Eng Comput 40:402–407
Nikkonen S, Afara IO et al (2019) Artificial neural network analysis of the oxygen saturation signal enables accurate diagnostics of sleep apnea. Sci Rep. https://doi.org/10.1038/s41598-019-49330-7
Leino A, Nikkonen S et al (2021) Neural network analysis of nocturnal SpO2 signal enables easy screening of sleep apnea in patients with acute cerebrovascular disease. Sleep Med 79:71–78
ElMoaqet H, Eid M, Glos M et al (2020) Deep recurrent neural networks for automatic detection of sleep apnea from single channel respiration signals. Sensors 20:5037. https://doi.org/10.3390/s20185037
Yue H, Lin Y et al (2021) Deep learning for diagnosis and classification of obstructive sleep apnea: a nasal airflow-based multi-resolution residual network. Nat Sci Sleep 13:361–373. https://doi.org/10.2147/NSS.S297856
Dey D, Chaudhuri S, Munshi S (2018) Obstructive sleep apnoea detection using convolutional neural network based deep learning framework. Biomed Eng Lett 8(1):95–100. https://doi.org/10.1007/s13534-017-0055-y
Shen Q, Qin H et al (2021) Multiscale deep neural network for obstructive sleep apnea detection using RR interval from single-lead ECG signal. IEEE Trans Instrum Meas 70:1–13
Albuhayri MA (2022) CNN model for sleep apnea detection based on SpO2 signal. Comput Inf Sci. https://doi.org/10.5539/cis.v16n1p39
Zarei A, Beheshti H et al (2022) Detection of sleep apnea using deep neural networks and single-lead ECG signals. Biomed Signal Process Control 71(Part A):103125. https://doi.org/10.1016/j.bspc.2021.103125
Erdenebayar U, Kim YJ, Park J-U, Joo EY et al (2019) Deep learning approaches for automatic detection of sleep apnea events from an electrocardiogram. Comput Methods Programs Biomed. https://doi.org/10.1016/j.cmpb.2019.105001
Rajabrundha A et al (2022) Analysis of sleep apnea considering electrocardiogram data using deep learning algorithms. J Phys: Conf Ser 2318:012009
Li K, Pan W, Li Y et al (2018) A method to detect sleep apnea based on deep neural network and hidden Markov model using single-lead ECG signal. Neurocomputing 294:94–101
Tanmoy P et al (2022) ECG and SpO2 signal-based real-time sleep apnea detection using feed-forward artificial neural network. AMIA Annu Symp Proc 2022:379–385
Abasi AK, Aloqaily M, Guizani M (2023) Optimization of CNN using modified honey badger algorithm for sleep apnea detection. Expert Syst Appl 229(Part A):120484. https://doi.org/10.1016/j.eswa.2023.120484
Gaspar A, Oliva D, Cuevas E, Zaldívar D, Pérez M, Pajares G (2021) Hyperparameter optimization in a convolutional neural network using metaheuristic algorithms. In: Metaheuristics in machine learning: theory and applications. Springer International Publishing, Cham, pp 37–59. https://doi.org/10.1007/978-3-030-70542-8_2
Mostafa SS, Mendonça F, Ravelo-Garcia AG, Gabriel Juliá-Serdá G, Morgado-Dias F (2020) Multi-objective hyperparameter optimization of convolutional neural network for obstructive sleep apnea detection. IEEE Access 8:129586–129599. https://doi.org/10.1109/ACCESS.2020.3009149
MIT-BIH database. https://doi.org/10.13026/C23K5S
Alsalamah M, Amin S et al (2018) Detection of obstructive sleep apnea using deep neural network. In: Applications of big data analytics: trends, issues, and challenge, pp 97–120
Mahmud T, Khan IA et al (2021) Sleep apnea detection from variational mode decomposed EEG signal using a hybrid CNN-BiLSTM. IEEE Access 9:102355–102367. https://doi.org/10.1109/ACCESS.2021.3097090
Yilmaz S, Sen S (2020) Electric fish optimization:a new heuristic algorithm inspired by electrolocation. Neural Comput Appl 32:11543–11578
Gandomi AH, Yang XS et al (2013) Firefly algorithm with chaos. Commun Nonlinear Sci Numer Simul 18(1):89–98
Heidari AA et al (2019) Harris hawks optimization: algorithm and applications. Future Gener Comput Syst 97:849–872
Gandomi AH, Alavi AH (2012) Krill herd: a new bio-inspired optimization algorithm. Commun Nonlinear Sci Numerical Simul 17(12):4831–4845
Fatimah B, Singh P et al (2020) Detection of apnea events from ECG segments using Fourier decomposition method. Biomed Signal Process Control. https://doi.org/10.1016/j.bspc.2020.102005
Taghizadegan Y et al (2021) Obstructive sleep apnea event prediction using recurrence plots and convolutional neural networks from polysomnographic signals. Biomed Signal Process Control. https://doi.org/10.1016/j.bspc.2021.102928
Huang G, Ma F (2021) ConCAD: contrastive learning-based cross attention for sleep apnea detection. Machine learning and knowledge discovery in databases. Lect Notes Comput Sci 12979:68–84
Golrou A, Keivan M et al (2015) Automatic sleep apnea detection using fuzzylogic. In: IEEE signal processing in medicine and biology symposium. Philadelphia. https://doi.org/10.1109/SPMB.2015.7405469
Acknowledgements
The Department of Science and Technology supported this work under Grant [No. SEED/TIDE/2018/20/G]. We thank Dr. Somusivabalan, a pulmonologist and senior consultant at the Rela Institute in Chennai, for his consistent guidance and assistance in completing this work.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors express no potential conflict of interest relevant to this content.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Raja Brundha, A., Lakshmi Sangeetha, A. & Balajiganesh, A. Deep learning framework for automatic detection and classification of sleep apnea severity from polysomnography signals. Neural Comput & Applic (2024). https://doi.org/10.1007/s00521-024-09889-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00521-024-09889-3