本研究提出了基於卷積式類神經網路及資料不平衡資料集處理策略的睡眠階層自動辨識演算法開發。卷積式類神經網路在處理影像資料有其高度的準確性及強健性，本研究使用多通道睡眠生理訊號中的腦波訊號（Fpz-Oz及Pz-Oz頻道）以及眼動圖訊號從時域透過連續小波轉換轉換到時頻域頻譜，並將兩個腦波訊號頻道以及一個演動圖的時頻域頻譜結合併成一張影像，然後以卷積式類神經網路進行辨識。由於睡眠階層的資料分布之自然特性屬於資料不平衡資料集，因此本論文使用欠抽樣(under-sampling)以及過抽樣(over-sampling)兩種常見的資料不平衡處理方式來進行處理，並使用k-fold交叉驗證(k=5,10)來進行測試。最後在原始資料集得到91.0%的正確率、在欠採樣的資料集下得到了90.7%的正確率、在過採樣下的資料集下得到了99.7%的正確率。

This research proposed an over-/under-sampling framework for the CNN-based sleep recognition algorithm via PSG spectrogram fusion. The convolutional neural network (CNN) is good at image classification with high accuracy and robustness. The raw EEG signal (Fpz-Oz and Pz-Oz channels) and EOG channel from polysomnography (PSG) are selected to transform the time-frequency spectrogram and combined as the input of the CNN. Owing to the fact that numbers of various sleep stages distributed unequally in nature. The data imbalanced processing techniques under-and over-sampling are utilized in this study. K-fold cross-validation is adopted as the cross-validation in this study. The classification accuracy of raw-data, under-sampling, over-sampling are 91.0%, 90.7%, 99.7% in this study.

[1] Z. Cui, et al. “Automatic Sleep Stage Classification Based on Convolutional Neural Network and Fine-Grained Segments,” Complexity, vol. 2018, 2018.
[2] S. Chambon, M. N. Galtier, P. J. Arnal, G. Wainrib and A. Gramfort, “A Deep Learning Architecture for Temporal Sleep Stage Classification Using Multivariate and Multimodal Time Series,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 26, no. 4, pp. 758-769, 2018.
[3] A. Supratak, H. Dong, C. Wu, and Y. Guo, “DeepSleepNet: A model for automatic sleep stage scoring based on raw single-channel EEG,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 25, no. 11, pp. 1998-2018, 2017.
[4] O. Tsinalis, P. M. Matthews, Y. Guo, and S. Zafeiriou, “Automatic sleep stage scoring with single-channel EEG using convolutional neural networks,” arXiv preprint arXiv:1610.01683, 2016.
[5] C. Sun, J. Fan, C. Chen, W. Li, and W. Chen, “A Two-Stage Neural Network for Sleep Stage Classification Based on Feature Learning,” Sequence Learning, and Data Augmentation. IEEE Access, vol. 7, pp. 109386-109397 2019.
[6] H. Dong, A. Supratak, W. Pan, C. Wu, P. M.Matthews, and Y. Guo, “Mixed neural network approach for temporal sleep stage classification,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 26, no. 2, pp. 324-333, 2017.
[7] Y. Jeon et al. “Pediatric Sleep Stage Classification Using Multi-Domain Hybrid Neural Networks,” IEEE Access, vol. 7, pp. 96495-96505, 2019.
[8] H. Phan, F. Andreotti, N. Cooray, O. Y. Chén, and M. De Vos, “Joint classification and prediction CNN framework for automatic sleep stage classification,” IEEE Transactions on Biomedical Engineering, vol. 66, no. 5, pp. 1285-1296, 2018.
[9] H. Phan, F. Andreotti, N. Cooray, O. Y. Chén, and M. De Vos, “SeqSleepNet: end-to-end hierarchical recurrent neural network for sequence-to-sequence automatic sleep staging,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 27, no. 3, pp. 400-410, 2019.
[10] P. Piñero, P. Garcia, L. Arco, A. Álvarez, M. M. Garcı́a, and R. Bonal, “Sleep stage classification using fuzzy sets and machine learning techniques,” Neurocomputing, vol. 58, pp. 1137-1143, 2004.
[11] M. Längkvist, L. Karlsson, and A. Loutfi, “Sleep stage classification using unsupervised feature learning,” Advances in Artificial Neural Systems, vol. 2012, no. 5, 2012.
[12] S. Crisler, M. J. Morrissey, A. M. Anch, and D. W. Barnett, “Sleep-stage scoring in the rat using a support vector machine,” Journal of neuroscience methods, vol. 168, no. 2, pp. 524-534, 2008.
[13] S. Güneş, K. Polat, and Ş. Yosunkaya, “Efficient sleep stage recognition system based on EEG signal using k-means clustering based feature weighting,” Expert Systems with Applications, vol. 37, no. 12, pp. 7922-7928, 2010.
[14] T. Lajnef et al. “Learning machines and sleeping brains: automatic sleep stage classification using decision-tree multi-class support vector machines,” Journal of neuroscience methods, vol. 250, pp. 94-105, 2015.
[15] O. Tsinalis, P. M. Matthews, and Y.Guo, “Automatic sleep stage scoring using time-frequency analysis and stacked sparse autoencoders, ”Annals of biomedical engineering, vol. 44, no. 5, pp. 1587-1597, 2016.
[16] M. Peker, “An efficient sleep scoring system based on EEG signal using complex-valued machine learning algorithms,” Neurocomputing, vol. 207, pp. 165-177, 2016.
[17] X. Li, L. Cui, S. Tao, J. Chen, X. Zhang, and G. Q. Zhang, “Hyclasss: A hybrid classifier for automatic sleep stage scoring,” IEEE journal of biomedical and health informatics, vol. 22, no. 2, pp. 375-385, 2017.
[18] T. Willemen et al. “An evaluation of cardiorespiratory and movement features with respect to sleep-stage classification,” IEEE journal of biomedical and health informatics, vol. 18, no. 2, pp. 661-669, 2013.
[19] S. M. Isa, I. Wasito, and A. M. Arymurthy, “Kernel dimensionality reduction on sleep stage classification using ECG signal,” International Journal of Computer Science Issues (IJCSI), vol. 8, no. 4, pp. 115, 2011.
[20] J. Zhang and Y. Wu, “A new method for automatic sleep stage classification,” IEEE transactions on biomedical circuits and systems, vol. 11, no. 5, pp. 1097-1110, 2017.
[21] S. Mousavi, F. Afghah, and U. R. Acharya, “SleepEEGNet: Automated sleep stage scoring with sequence to sequence deep learning approach,” PloS one, vol. 14, no. 5, pp. e0216456, 2019.
[22] J. Zhang and Y. Wu, “Complex-valued unsupervised convolutional neural networks for sleep stage classification,” Computer methods and programs in biomedicine, vol. 164, pp. 181-191, 2018.
[23] X. Long, J. Foussier, P. Fonseca, R. Haakma, and R. M. Aarts, “Analyzing respiratory effort amplitude for automated sleep stage classification,” Biomedical Signal Processing and Control, vol. 14, pp. 197-205, 2014.
[24] R. Boostani, F. Karimzadeh, and M. Nami, “A comparative review on sleep stage classification methods in patients and healthy individuals,” Computer methods and programs in biomedicine, vol. 140, pp. 77-91, 2017.
[25] O. Tsinalis, P. M. Matthews, Y. Guo, and S. Zafeiriou, “Automatic sleep stage scoring with single-channel EEG using convolutional neural networks,” arXiv preprint arXiv:1610.01683, 2016.
[26] O.Tsinalis, P. M. Matthews, and Y. Guo, “Automatic sleep stage scoring using time-frequency analysis and stacked sparse autoencoders,” Annals of biomedical engineering, vol. 44, no. 5, pp. 1587-1597, 2016.
[27] A. Supratak, H. Dong, C. Wu, and Y. Guo, “DeepSleepNet: A model for automatic sleep stage scoring based on raw single-channel EEG,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 25, no. 11, pp. 1998-2008, 2017.
[28] T. L. da Silveira, A. J. Kozakevicius, and C. R. Rodrigues, “Automated drowsiness detection through wavelet packet analysis of a single EEG channel,” Expert Systems with Applications, vol. 55, pp. 559-565, 2016.
[29] T. L. da Silveira, A. J. Kozakevicius, and C. R. Rodrigues, “Single-channel EEG sleep stage classification based on a streamlined set of statistical features in wavelet domain,” Medical & biological engineering & computing, vol. 55, no. 2, pp. 343-352, 2017.
[30] M. M. Rahman, M. I. H. Bhuiyan, and A. R. Hassan, “Sleep stage classification using single-channel EOG,” Computers in biology and medicine, vol. 102, pp. 211-220, 2018.
[31] J. Zhang, Y. Wu, J. Bai, and F. Chen, “Automatic sleep stage classification based on sparse deep belief net and combination of multiple classifiers,” Transactions of the Institute of Measurement and Control, vol. 38, no. 4, pp. 435-451, 2016.
[32] American Academy of Sleep Medicine. “Standards for Accreditation.” Internet: http://www.stop-being-tired.com/tiredness-and-sleep/sleep-stages-and-cycles Nov., 2016 [Dec. 01, 2019].
[33] M. Sandsten, “Time-frequency analysis of time-varying signals and non-stationary processes,” Lund University, 2016.
[34] J. P. Antoine, P. Carrette, R. Murenzi, and B. Piette, “Image analysis with two-dimensional continuous wavelet transform,” Signal processing, vol. 31, no. 3, pp. 241-272, 1993.
[35] MATLAB Mathworks. “jet.” Internet: https://la.mathworks.com/help/matlab/ref/jet.html [Dec. 01, 2019].
[36] MATLAB Mathworks. “parula.” Internet: https://la.mathworks.com/help/matlab/ref/parula.html [Dec. 01, 2019].
[37] MATLAB Mathworks. “hsv.” Internet: https://la.mathworks.com/help/matlab/ref/hsv.html [Dec. 01, 2019].
[38] Z. Cui, X. Zheng, X. Shao, and L. Cui, “Automatic Sleep Stage Classification Based on Convolutional Neural Network and Fine-Grained Segments,” Complexity, 2018.
[39] Wikipedia, the free encyclopedia. “Oversampling and undersampling in data analysis.” Internet: https://en.wikipedia.org/wiki/Oversampling_and_undersampling_in_data_analysis#cite_note-:0-5 [Dec. 12, 2019].
[40] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” In Advances in neural information processing systems, pp. 1097-1105, 2012.
[41] M. Khoshdeli, and B. Parvin, “Feature-based representation improves color decomposition and nuclear detection using a convolutional neural network,” IEEE Transactions on Biomedical Engineering, vol. 65, no. 3, pp. 625-634, 2017.
[42] H. Yang, “Multiscale recurrence quantification analysis of spatial cardiac vectorcardiogram signals,” IEEE Transactions on Biomedical Engineering, vol. 58, no. 2, pp. 339-347, 2010.