睡眠在人的一生中約佔據三分之一的時間，然而全球有數以千萬計的人都面臨著失眠、睡不好等問題。目前最為標準的睡眠檢測儀器為多通道生理訊號接收器(PSG)，並利用該機器所接收睡眠生理訊號，來做為專家判讀睡眠之依據。但該儀器操作複雜，容易干擾使用者睡眠之外，也需要專人進行穿戴操作，並無法進行長時間觀測。此外PSG訊號須經由專家判讀，需耗費許時間並無法即時分析，然而失眠現象是需要進行長時間的觀測，才能有效指出問題所在。因此我們便開發了一套眼罩與手機裝置之睡眠分析系統，擁有著高度的便利性，能夠即時分析資料的特性同時，仍然擁有高度的準確度，並能夠進行長時間的追蹤觀測。且除了追求睡眠階段的準確度之外，也著重於臨床所使用的睡眠指標，讓整體的分析結果更加趨近於臨床睡眠分析報告。而在本研究中，我們使用圖片辨識模型進行四階段睡眠分類以及睡眠分析。利用實驗中所收取的17筆整夜睡眠資料來訓練模型以及8筆資料進行系統測試。我們利用眼罩接收腦波以及眼動訊號後，經由眼罩內部的韌體進行濾波並輸出FFT資料，再藉由行動裝置端接收轉換為時頻圖，最後進入MobileNetV2圖形辨識網路進行運算得到睡眠階段結果。不僅整體的過程皆為即時的，且整體測試及資料與專家判讀的準確度高達了86.72 %，並在四階段中都取得了很好的成績。針對各個睡眠指標的分析中，表現也都十分優異。更在SE (睡眠效率)這項指標中，與專家之平均誤差只有1.6%。證實我們系統擁有高度的準確性同時，睡眠指標誤差也能控制於極小的範圍內。未來也希望能讓這套系統應在不同族群的使用者身上，讓整套系統更加準確，並讓更多使用者接觸。

Sleep occupies about one-third of a person's life, but tens of millions of people around the world are facing problems such as insomnia and poor sleep. At present, the most standard sleep detection instrument is Polysomnography (PSG), and uses the sleep physiological signal received by the device as a basis for experts to interpret sleep. However, the instrument is complicated to operate and easily interferes with the user's sleep. It also requires a dedicated person to wear it and cannot perform long-term observations. In addition, the PSG signal must be interpreted by experts, which takes a lot of time and cannot be analyzed in real time. However, insomnia requires long-term observation to effectively point out the problem. Therefore, we have developed a sleep analysis system for eye masks and mobile devices, which has a high degree of convenience, has a high degree of accuracy, can analyze data in real time, and can be tracked for a long time. In addition to pursuing the accuracy of sleep stages, it also focuses on the sleep index used clinically, making the overall analysis results closer to the clinical sleep analysis report. In this study, we used an image classification model to perform four-stage sleep classification and sleep analysis. Use 17 all-night sleep data received in the experiment to train the model and 8 pieces of data for system testing. We use the eyemask to receive EEG signal and EEG signal, filter it through the firmware inside the eyemask and output FFT data, and then convert it to a time-frequency image by the mobile device, and finally enter the MobileNetV2 image recognition network to calculate the sleep stage result. Not only is the overall real-time process, but the accuracy of the overall test and data and expert interpretation is as high as 86.72%, and also good results have been achieved in the four stages. In the analysis of various sleep index, the performance is also very pretty. Even in the SE (Sleep Efficiency) index, the average error with scorer is only 1.6%. Confirm that our system has a high degree of accuracy. In the future, we hope that this system can be applied to users of different ethnic groups, so that the whole system is more accurate and more users can get in touch.

Agarwal, R. and J. Gotman (2001). "Computer-assisted sleep staging." IEEE Transactions on Biomedical Engineering 48(12): 1412-1423.
Anderer, P., et al. (2005). "An E-health solution for automatic sleep classification according to Rechtschaffen and Kales: validation study of the Somnolyzer 24× 7 utilizing the Siesta database." Neuropsychobiology 51(3): 115-133.
Bajaj, V. and R. B. Pachori (2013). "Automatic classification of sleep stages based on the time-frequency image of EEG signals." Computer methods and programs in biomedicine 112(3): 320-328.
Dimitriadis, S. I., et al. (2018). "A novel, fast and efficient single-sensor automatic sleep-stage classification based on complementary cross-frequency coupling estimates." Clinical Neurophysiology 129(4): 815-828.
Fraiwan, L., et al. (2012). "Automated sleep stage identification system based on time–frequency analysis of a single EEG channel and random forest classifier." Computer methods and programs in biomedicine 108(1): 10-19.
Hassan, A. R., et al. (2015). Automatic classification of sleep stages from single-channel electroencephalogram. India Conference (INDICON), 2015 Annual IEEE, IEEE.
Hsu, Y.-L., et al. (2013). "Automatic sleep stage recurrent neural classifier using energy features of EEG signals." Neurocomputing 104: 105-114.
Imtiaz, S. A. and E. Rodriguez-Villegas (2015). Automatic sleep staging using state machine-controlled decision trees. Engineering in Medicine and Biology Society (EMBC), 2015 37th Annual International Conference of the IEEE, IEEE.
Jansen, B. H. and B. M. Dawant (1989). "Knowledge-based approach to sleep EEG analysis-a feasibility study." IEEE Transactions on Biomedical Engineering 36(5): 510-518.
Kelly, J. M., et al. (2012). "Recent Developments in Home Sleep-Monitoring Devices." ISRN neurology 2012.
Li, X., et al. (2017). "HyCLASSS: A Hybrid Classifier for Automatic Sleep Stage Scoring." IEEE Journal of Biomedical and Health Informatics.
Liang, S.-F., et al. (2012). "A rule-based automatic sleep staging method." Journal of neuroscience methods 205(1): 169-176.
Liang, S.-F., et al. (2015). "Development of an EOG-based automatic sleep-monitoring eye mask." IEEE Transactions on Instrumentation and Measurement 64(11): 2977-2985.
Michielli, N., et al. (2019). "Cascaded LSTM recurrent neural network for automated sleep stage classification using single-channel EEG signals." Computers in biology and medicine 106: 71-81.
Norman, R. G., et al. (2000). "Interobserver agreement among sleep scorers from different centers in a large dataset." Sleep 23(7): 901-908.
Park, H., et al. (2000). Hybrid neural-network and rule-based expert system for automatic sleep stage scoring. Engineering in Medicine and Biology Society, 2000. Proceedings of the 22nd Annual International Conference of the IEEE, IEEE.
Peker, M. (2016). "A new approach for automatic sleep scoring: Combining Taguchi based complex-valued neural network and complex wavelet transform." Computer methods and programs in biomedicine 129: 203-216.
Rodríguez-Sotelo, J. L., et al. (2014). "Automatic sleep stages classification using eeg entropy features and unsupervised pattern analysis techniques." Entropy 16(12): 6573-6589.
Sandler, M., et al. (2018). Mobilenetv2: Inverted residuals and linear bottlenecks. Proceedings of the IEEE conference on computer vision and pattern recognition.
Sateia, M. J. (2014). "International classification of sleep disorders." Chest 146(5): 1387-1394.
Schaltenbrand, N., et al. (1996). "Sleep stage scoring using the neural network model: comparison between visual and automatic analysis in normal subjects and patients." Sleep 19(1): 26-35.
Smith, J. R., et al. (1969). "Automatic analysis of sleep electroencephalograms by hybrid computation." IEEE transactions on systems science and cybernetics 5(4): 278-284.
Sun, Y., et al. (2018). Deep Convolutional Network Method for Automatic Sleep Stage Classification Based on Neurophysiological Signals. 2018 11th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), IEEE.
Tian, J. and J. Liu (2006). Automated sleep staging by a hybrid system comprising neural network and fuzzy rule-based reasoning. 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference, IEEE.
Whitney, C. W., et al. (1998). "Reliability of scoring respiratory disturbance indices and sleep staging." Sleep 21(7): 749-757.