根據台灣睡眠醫學學會(Taiwan Society of Sleep Medicine)在2017年的調查，全台有1/10人口受慢性失眠症所苦，盛行率為11.3%，意同大約有超過兩百萬人以上，都有睡眠方面的障礙。除了身心機能的疾病之外，造成失眠的一大主因則是合併身體的慢性病，這一個族群常見到的，就是睡眠呼吸中止症(Sleep apnea, SA)。對睡眠障礙患者而言，目前最好的檢驗方式是使用多通道生理紀錄儀 (Polysomnography, PSG) 來紀錄病患整晚的生理訊號，之後再由睡眠技師進行人工判讀睡眠階段，供醫師診斷參考。由於近年來國人失眠比率及民眾正確就醫知識均增加，醫院睡眠中心每月都有近百筆資料需要睡眠技師判讀。人工判讀是一項非常主觀且耗時間的工作，故已有許多自動睡眠判讀的演算法相繼被提出。然而目前許多自動睡眠判讀系統雖然在準確率上有不錯的表現，多數系統除了僅適用在正常睡眠的族群上，也只能應用特定的資料群組，不同的群組必須重新訓練及調整系統。因此，目前醫院主要仍是利用睡眠技師人工判讀的方式來處理資料，市面上的自動判讀系統仍然無法對他們有實質上的幫助。
本研究提出一套適用於睡眠中心的自動睡眠判讀系統，系統的主要分類方式是使用能結合專家知識的法則式決策樹，以美國睡眠醫學學會AASM(American Academy of Sleep Medicine)訂定的標準為主，與睡眠技師討論其判讀依據及需求為輔。因為在階段轉換處睡眠訊號特徵會較不明顯，非AASM訂定的標準訊號，睡眠技師會考慮患者與正常族群之間差異性，依照自身的經驗及看法做判讀標準上的調整，雖然每位技師間結果不盡相同，但都屬合理的範圍內，因此本研究開發了結合人機協同的判讀介面，將系統自動判讀的結果中信賴度較低的部分標示出來，技師可針對這些有爭議性的部分做修正，有效且實際的運用自動判讀系統來降低判讀所花的時間。系統與專家的睡眠階段在訓練集(25筆平均SE為85%以上的OSA患者)總體準確率可達86.18%，測試集(25筆平均SE為85%以上的OSA患者) 總體準確率可達85.01%，其中除了N1階段，其餘各階段準確率皆可達到75%以上。此外，本研究開發的介面，除了顯示AASM規範基本的睡眠訊號及階段之外，亦提供相關睡眠指標及資訊，讓專家作為判讀時的重要參考。希望本系統能扮演輔助睡眠技師的角色，有效的減輕其工作量，並利用系統做定期測試，提升睡眠技師之間的判讀一致性。

According to a survey conducted by the Taiwan Society of Sleep Medicine in 2017, 1 in 10 people in the whole Taiwan suffers from chronic insomnia, with a prevalence rate of 11.3%. An intention of more than 2 million people has sleep disorders. In addition to physical and mental diseases, a major cause of insomnia is the combination of chronic diseases of the body. This common group is sleep apnea (SA). For patients with sleep disorders, the best test method is to use Polysomnography (PSG) to record the patient's physiological signal throughout the night, and then the sleep experts manually score the sleep stage for the doctor's diagnosis. Due to the increase of the insomnia rate of people in Taiwan and the correct medical knowledge of the people in recent years, the sleep center has nearly 100 pieces of data every month that requires the sleep experts to score. Manual scoring is a very subjective and time-consuming task, so many algorithms for automatic sleep interpretation have been proposed. However, although many automatic sleep scoring systems have good performance in accuracy, most systems can only apply the specific datasets except for healthy groups. Different groups must retrain and adjust the system. Therefore, at present, the hospital still mainly uses the method of manual scoring by the sleep expert to process the data, and the automatic scoring system on the market still cannot substantially help them.
This study proposes an automatic sleep scoring system for sleep centers. The main classification of the system is to use a rule-based decision tree that can be combined with expert knowledge. Based on the standards set by the American Academy of Sleep Medicine, we discuss with the sleep experts the basis and needs of their scoring. Because the sleep signal characteristics are less obvious at the stage transition, different to the standard signal set by AASM, the sleep experts consider the difference between the patient and the normal group at first and make adjustments based on their own experience and opinions. Although the results of each expert are not the same, they are all within a reasonable range. Therefore, this study developed a computer-aided sleep scoring interface, which marked the low-reliability parts of the results of the automatic scoring system. The experts only need to correct these debatable parts, effective and practical use of the automatic scoring system to reduce the time spent on scoring. The overall agreement of sleep stages between the system and the experts in the training dataset (25 OSA patients with an average SE of 85% or more) can reach 86.18%, and the overall agreement of the testing dataset (25 OSA patients with an average SE of 85% or more) is 85.01%. Except for the N1 stage, the agreement of the other stages can reach more than 75%. In addition, the interface developed in this study, apart from showing sleep signals and stages, the basic specifications of the AASM, it also provides relevant sleep index and information, making experts an important reference for scoring. It is hoped that the system can play the role of assisting sleep experts, effectively reducing the workload. Alongside this, experts can use the system to do regular tests to improve the consistency of scoring between them.

Agarwal, R., & Gotman, J. (2001). Computer-assisted sleep staging. IEEE Transactions on Biomedical Engineering, 48(12), 1412-1423.
Anderer, P., Gruber, G., Parapatics, S., Woertz, M., Miazhynskaia, T., Klösch, G., . . . Danker-Hopfe, H. (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.
Chang, T.-H. (2017). An Automatic Sleep Scoring System for Accurate Estimation of Various Sleep Measures. (Master), National Cheng Kung University, Taiwan.
Dimitriadis, S. I., Salis, C., & Linden, D. (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.
Duman, F., Erdamar, A., Erogul, O., Telatar, Z., & Yetkin, S. (2009). Efficient sleep spindle detection algorithm with decision tree. Expert Systems with Applications, 36(6), 9980-9985.
Hassan, A. R., Bashar, S. K., & Bhuiyan, M. I. H. (2015). Automatic classification of sleep stages from single-channel electroencephalogram. Paper presented at the India Conference (INDICON), 2015 Annual IEEE.
Iber, C. (2007). The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specifications: American Academy of Sleep Medicine.
Imtiaz, S. A., & Rodriguez-Villegas, E. (2015). Automatic sleep staging using state machine-controlled decision trees. Paper presented at the Engineering in Medicine and Biology Society (EMBC), 2015 37th Annual International Conference of the IEEE.
Jansen, B. H., & Dawant, B. M. (1989). Knowledge-based approach to sleep EEG analysis-a feasibility study. IEEE Transactions on Biomedical Engineering, 36(5), 510-518.
Kuo, C.-E., Liang, S.-F., Lee, Y.-C., Cherng, F.-Y., Lin, W.-C., Chen, P.-Y., . . . Shaw, F.-Z. (2014). An EOG-based automatic sleep scoring system and its related application in sleep environmental control. Paper presented at the International Conference on Physiological Computing Systems.
Li, X., Cui, L., Tao, S., Chen, J., Zhang, X., & Zhang, G.-Q. (2017). HyCLASSS: A Hybrid Classifier for Automatic Sleep Stage Scoring. IEEE Journal of Biomedical and Health Informatics.
Liang, S.-F., Kuo, C.-E., Hu, Y.-H., & Cheng, Y.-S. (2012). A rule-based automatic sleep staging method. Journal of neuroscience methods, 205(1), 169-176.
Mignot, E. (2008). Why we sleep: the temporal organization of recovery. PLoS Biol, 6(4), e106.
Norman, R. G., Pal, I., Stewart, C., Walsleben, J. A., & Rapoport, D. M. (2000). Interobserver agreement among sleep scorers from different centers in a large dataset. Sleep, 23(7), 901-908.
Park, H., Park, K., & Jeong, D.-U. (2000). Hybrid neural-network and rule-based expert system for automatic sleep stage scoring. Paper presented at the Engineering in Medicine and Biology Society, 2000. Proceedings of the 22nd Annual International Conference of the 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.
Rechtschaffen, A., & Kales, A. (1968). A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects.
Richard B.Berry, M. (2017). The AASM Manual for the Scoring of Sleep and Associated Events: rules, terminology and technical specifications (version 2.4 ed.): American Academy of Sleep Medicine.
Rodríguez-Sotelo, J. L., Osorio-Forero, A., Jiménez-Rodríguez, A., Cuesta-Frau, D., Cirugeda-Roldán, E., & Peluffo, D. (2014). Automatic sleep stages classification using eeg entropy features and unsupervised pattern analysis techniques. Entropy, 16(12), 6573-6589.
Rosenberg, R. S., & Van Hout, S. (2013). The American Academy of Sleep Medicine inter-scorer reliability program: sleep stage scoring. J Clin Sleep Med, 9(1), 81-87.
Schaltenbrand, N., Lengelle, R., Toussaint, M., Luthringer, R., Carelli, G., Jacqmin, A., . . . Macher, J.-P. (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., Negin, M., & Nevis, A. H. (1969). Automatic analysis of sleep electroencephalograms by hybrid computation. IEEE transactions on systems science and cybernetics, 5(4), 278-284.
Sun, Y., Wang, B., Jin, J., & Wang, X. (2018). Deep Convolutional Network Method for Automatic Sleep Stage Classification Based on Neurophysiological Signals. Paper presented at the 2018 11th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI).
Tian, J., & Liu, J. (2006). Automated sleep staging by a hybrid system comprising neural network and fuzzy rule-based reasoning. Paper presented at the 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.
Whitney, C. W., Gottlieb, D. J., Redline, S., Norman, R. G., Dodge, R. R., Shahar, E., . . . Nieto, F. J. (1998). Reliability of scoring respiratory disturbance indices and sleep staging. Sleep, 21(7), 749-757.
Zadeh, L. A. (1965). Fuzzy sets. Information and control, 8(3), 338-353.