SingleChannelNet: A Model for Automatic Sleep Stage Classification with Raw Single-Channel EEG

Sleep is a natural form of rest for humans. People need sleep to perform their daily functions. Insufficient or unstable sleep may adversely affect the function of many systems in human body. Sleep disorders can be seen common and cause serious health problems that affect quality of life. From past to present, it has become imperative to classify sleep stages in order to accurately analyze and diagnose these disorders. This classification is made by people who are experts in the field of sleep. However, this process is a very laborious task that requires high attention, and since it is done by a human, it is quite normal to make wrong classifications. As a solution to this, it is possible to make these classifications with machine learning techniques to obtain more accurate results. In this study, we compared different classification methods with each other and examined the channel-based accuracy of the method that gives the highest accuracy based on channels. The accuracy of the Fine Gaussian SVM Method was 98.9% and the F1-score was 98.95, the accuracy of the Weighted KNN Method was 97.9% and the F1-score was 97.89, the accuracy of the Wide Neural Network Method was 97.4% and the F1-score was 97.09, the accuracy of the Cubic SVM Method was 96.2% and the F1-score was 96.36. When we examine the Fine Gaussian SVM Method with the highest accuracy based on channels, we found accuracy of only Fpz-CZ channel is 98.1%, accuracy of only Pz-Oz channel is 94.5%.

show abstract

“…The results are evaluated by precision, recall, accuracy, F1score values that are computed for the test data [15]:…”

Section: Resultsmentioning

confidence: 99%

Classification of Sleep Stages via Machine Learning Algorithms

2022

View full text Add to dashboard Cite

show abstract

“…Some of the recently proposed DL based works are: convolutional neural network (CNN) ( 65 , 66 ), IITNet, a CNN and Recurrent Neural Networks (RNN) based network ( 67 ), SingleChannelNet (SCNet), a CNN based model. The DL based approaches achieved 83.9 to 92.9% for a standard five-class classification ( 68 ), SleepStageNet ( 69 ), Long short-term memory (LSTM)-RNN ( 70 ). The Figure 1 illustrates the overview of sleep data extraction and analysis pipeline using cloud computing.…”

Section: Role Of Ai In Sleep Stage Classificationmentioning

confidence: 99%

Efficacy of Single-Channel EEG: A Propitious Approach for In-home Sleep Monitoring

Radhakrishnan

Kirubakaran

Jebadurai

et al. 2022

Front. Public Health

View full text Add to dashboard Cite

ZleepAnlystNet: a novel deep learning model for automatic sleep stage scoring based on single-channel raw EEG data using separating training

Jirakittayakorn,

Wongsawat,

Mitrirattanakul

2024

Sci Rep

View full text Add to dashboard Cite

Numerous models for sleep stage scoring utilizing single-channel raw EEG signal have typically employed CNN and BiLSTM architectures. While these models, incorporating temporal information for sequence classification, demonstrate superior overall performance, they often exhibit low per-class performance for N1-stage, necessitating an adjustment of loss function. However, the efficacy of such adjustment is constrained by the training process. In this study, a pioneering training approach called separating training is introduced, alongside a novel model, to enhance performance. The developed model comprises 15 CNN models with varying loss function weights for feature extraction and 1 BiLSTM for sequence classification. Due to its architecture, this model cannot be trained using an end-to-end approach, necessitating separate training for each component using the Sleep-EDF dataset. Achieving an overall accuracy of 87.02%, MF1 of 82.09%, Kappa of 0.8221, and per-class F1-socres (W 90.34%, N1 54.23%, N2 89.53%, N3 88.96%, and REM 87.40%), our model demonstrates promising performance. Comparison with sleep technicians reveals a Kappa of 0.7015, indicating alignment with reference sleep stags. Additionally, cross-dataset validation and adaptation through training with the SHHS dataset yield an overall accuracy of 84.40%, MF1 of 74.96% and Kappa of 0.7785 when tested with the Sleep-EDF-13 dataset. These findings underscore the generalization potential in model architecture design facilitated by our novel training approach.

show abstract

SingleChannelNet: A Model for Automatic Sleep Stage Classification with Raw Single-Channel EEG

Cited by 3 publications

References 45 publications

Classification of Sleep Stages via Machine Learning Algorithms

Classification of Sleep Stages via Machine Learning Algorithms

Efficacy of Single-Channel EEG: A Propitious Approach for In-home Sleep Monitoring

ZleepAnlystNet: a novel deep learning model for automatic sleep stage scoring based on single-channel raw EEG data using separating training

Contact Info

Product

Resources

About