EOGNET: A Novel Deep Learning Model for Sleep Stage Classification Based on Single-Channel EOG Signal

Fan, Jiahao; Sun, Chenglu; Meng, Long; Chen, Chen; Chen, Wei

doi:10.3389/fnins.2021.573194

Cited by 42 publications

(21 citation statements)

References 22 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The automatic sleep staging model presented in this study performed on a similar level as other state-of-the-art automatic methods utilizing only single-channel EOG signal (accuracies between 76%-91%) [32]- [34]. It also reached similar F1scores as what has been reported with another U-time-derived network i.e., the U-sleep utilizing one EOG channel combined with one forehead EEG channel (F1-scores=0.76-0.77) [15].…”

Section: A Performance Of the Modelmentioning

confidence: 91%

Generalizable Deep Learning-based Sleep Staging Approach for Ambulatory Textile Electrode Headband Recordings

Rusanen

Huttunen

Korkalainen

et al. 2023

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Reliable, automated, and user-friendly solutions for the identification of sleep stages in home environment are needed in various clinical and scientific research settings. Previously we have shown that signals recorded with an easily applicable textile electrode headband (FocusBand, T 2 Green Pty Ltd) contain characteristics similar to the standard electrooculography (EOG, E1-M2). We hypothesize that the electroencephalographic (EEG) signals recorded using the textile electrode headband are similar enough with standard EOG in order to develop an automatic neural network-based sleep staging method that generalizes from diagnostic polysomnographic (PSG) data to ambulatory sleep recordings of textile electrode-based forehead EEG. Standard EOG signals together with manually annotated sleep stages from clinical PSG dataset (n=876) were used to train, validate, and test a fully convolutional neural network (CNN). Furthermore, ambulatory sleep recordings including a standard set of gel-based electrodes and the textile electrode headband were conducted for 10 healthy volunteers at their homes to test the generalizability of the model. In the test set (n=88) of the clinical dataset, the model's accuracy for 5-stage sleep stage classification was 80% (κ=0.73) using only the single-channel EOG. The model generalized well for the headband-data, reaching 82% (κ=0.75) overall sleep staging accuracy. In comparison, accuracy of the model was 87% (κ=0.82) in home recordings using the standard EOG. In conclusion, the CNN model shows potential on automatic sleep staging of healthy individuals using a reusable electrode headband in a home environment.

show abstract

Section: A Performance Of the Modelmentioning

confidence: 91%

Generalizable Deep Learning-based Sleep Staging Approach for Ambulatory Textile Electrode Headband Recordings

Rusanen

Huttunen

Korkalainen

et al. 2023

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

show abstract

“…The first group depends on traditional machine learning such as logical rules, threshold, and artificial neural networks (ANN) [13][14][15][16][17][18][19]. The second one uses deep learning models such as convolutional neural networks (CNN) [20][21][22][23].…”

Section: Literature Reviewsmentioning

confidence: 99%

“…Three feature types are extracted: statistical features, the derived AR coefficients from the Burg method, and estimated PSD using the Yule-Walker method. For classification, Artificial neural Fan Jiahao et al[22] designed a new approach to determine the automatic gradient of sleep. The electrooculogram (EOG) signals are used instead of relying on electroencephalographic (EEG) signals.…”

mentioning

confidence: 99%

Developing a Method for Classifying Electro-Oculography (EOG) Signals Using Deep Learning

Hossieny

Tantawi

Shedeed

et al. 2022

IJICIS

View full text Add to dashboard Cite

Recently, a significant increase appears in the number of patients with severe motor disabilities even though the cognitive parts of their brains are intact. These disabilities prevent them from being able to move all their limbs except for the movement of their eyes. This creates great difficulty in carrying out the simplest daily activities, as well as difficulty in communicating with their surrounding environment. With the advent of Human Computer Interfaces (HCI), a new method of communication has been found based on determining the direction of eye movement. The eye movement is recorded by Electro-oculogram (EOG) using a set of electrodes placed around the eye horizontally and vertically. In this work, The horizontal and vertical EOG signals are filtered and analyzed to determine six eye movement directions (Right, left, up, down, center, and double blinking). The deep learning models namely Residual network and ResNet-50 network have been examined. The experimental results show that the ResNet-50 network gives the best average accuracy 95.8%.

show abstract

“…Most studies chose EEG as the primary modality (Supratak et al, 2017 ). Some studies selected EOG signals which could be more convenient to acquire than EEG signals (Fan et al, 2021 ). Other studies also adopted EMG signals with more distinguishable features between the W and REM stages (Li et al, 2022 ).…”

Section: Introductionmentioning

confidence: 99%

MMASleepNet: A multimodal attention network based on electrophysiological signals for automatic sleep staging

et al. 2022

View full text Add to dashboard Cite

Pandemic-related sleep disorders affect human physical and mental health. The artificial intelligence (AI) based sleep staging with multimodal electrophysiological signals help people diagnose and treat sleep disorders. However, the existing AI-based methods could not capture more discriminative modalities and adaptively correlate these multimodal features. This paper introduces a multimodal attention network (MMASleepNet) to efficiently extract, perceive and fuse multimodal features of electrophysiological signals. The MMASleepNet has a multi-branch feature extraction (MBFE) module followed by an attention-based feature fusing (AFF) module. In the MBFE module, branches are designed to extract multimodal signals' temporal and spectral features. Each branch has two-stream convolutional networks with a unique kernel to perceive features of different time scales. The AFF module contains a modal-wise squeeze and excitation (SE) block to adjust the weights of modalities with more discriminative features and a Transformer encoder (TE) to generate attention matrices and extract the inter-dependencies among multimodal features. Our MMASleepNet outperforms state-of-the-art models in terms of different evaluation matrices on the datasets of Sleep-EDF and ISRUC-Sleep. The implementation code is available at: https://github.com/buptantEEG/MMASleepNet/.

show abstract

EOGNET: A Novel Deep Learning Model for Sleep Stage Classification Based on Single-Channel EOG Signal

Cited by 42 publications

References 22 publications

Generalizable Deep Learning-based Sleep Staging Approach for Ambulatory Textile Electrode Headband Recordings

Generalizable Deep Learning-based Sleep Staging Approach for Ambulatory Textile Electrode Headband Recordings

Developing a Method for Classifying Electro-Oculography (EOG) Signals Using Deep Learning

MMASleepNet: A multimodal attention network based on electrophysiological signals for automatic sleep staging

Contact Info

Product

Resources

About