阻塞型睡眠呼吸中止症（Obstructive Sleep Apnea, OSA）是一種常見且具有潛在風險的睡眠障礙，其特徵為睡眠期間反覆發生上呼吸道塌陷與阻塞，導致間歇性低氧與睡眠中斷。若未妥善治療，OSA 不僅影響睡眠品質與日常功能，長期更可能增加高血壓、心律不整、腦中風及冠心病等心血管疾病的發生風險。其標準治療方式為持續正壓呼吸器（Continuous Positive Airway Pressure, CPAP）。儘管 CPAP 已被廣泛證實具有良好療效，但許多患者因配戴不適或長期依從性不佳而難以持續使用，因此可能轉而選擇手術作為替代治療選項。由於手術具有不可逆性，且其治療成效存在個體差異，若能在術前有效預測療效，將有助於臨床決策與病患溝通，提升治療的整體品質與效率。
本研究提出一種整合加權可見度圖與模糊推論系統的多階段機器學習方法，旨在提升 OSA 術後成效的預測準確性。方法上，透過從鼻壓訊號中萃取圖論與時域特徵，結合支援向量機（SVM）進行特徵選擇與模型訓練，並導入模糊推論系統（Fuzzy Inference System, FIS），以提供更具彈性與可解釋性的預測結果。實驗結果顯示，本方法在二元分類與術後呼吸中止指數（Apnea-Hypopnea Index, AHI）之迴歸預測上皆展現良好表現，且模糊推論系統有效補足傳統硬性標籤在臨床應用上的侷限。未來可望作為術前風險評估的輔助工具，協助醫師制定更個別化的治療策略。

Obstructive Sleep Apnea (OSA) is a common and potentially high-risk sleep disorder, characterized by repeated collapse and obstruction of the upper airway during sleep, leading to intermittent hypoxia and sleep fragmentation. If not properly treated, OSA can not only affect sleep quality and daily functioning but also increase the long-term risk of cardiovascular diseases such as hypertension, arrhythmia, stroke, and coronary artery disease. The standard treatment for OSA is Continuous Positive Airway Pressure (CPAP), which has been widely proven to be clinically effective. However, due to discomfort or poor long-term adherence, many patients find it difficult to continue using CPAP and may instead opt for surgical treatment as an alternative. Since surgery is irreversible and its therapeutic outcomes vary among individuals, effective preoperative prediction can greatly assist clinical decision-making and patient communication, thereby improving the overall quality and efficiency of treatment.
This study proposes a multi-stage machine learning approach that integrates Weighted Visibility Graph and Fuzzy Inference System to improve the accuracy of postoperative outcome prediction in OSA. The method involves extracting graph-theoretical and time-domain features from nasal pressure signals, performing feature selection and model training using Support Vector Machine (SVM), and incorporating a Fuzzy Inference System (FIS) to provide more flexible and interpretable prediction results. Experimental results show that the proposed method performs well in both binary classification and regression prediction of postoperative Apnea-Hypopnea Index (AHI), and the FIS effectively compensates for the limitations of traditional rigid labeling in clinical applications. This approach is expected to serve as a preoperative risk assessment tool to support clinicians in formulating more personalized treatment strategies.

[1] P. Lévy et al., “Obstructive sleep apnoea syndrome,” Nature Reviews Disease Primers, vol. 1, Art. no. 15015, Jun. 25, 2015.
[2] C. V. Senaratna et al., “Prevalence of obstructive sleep apnea in the general population: A systematic review,” Sleep Medicine Reviews, vol. 34, pp. 70–81, 2017.
[3] I. Fietze et al., “Prevalence and association analysis of obstructive sleep apnea with gender and age differences - Results of SHIP-Trend,” Journal of Sleep Research, vol. 28, no. 5, Art. no. e12770, 2019.
[4] A. R. Schwartz, S. P. Patil, A. M. Laffan, V. Polotsky, H. Schneider, and P. L. Smith, “Obesity and obstructive sleep apnea: pathogenic mechanisms and therapeutic approaches,” Proceedings of the American Thoracic Society, vol. 5, no. 2, pp. 185–192, 2008.
[5] S. Jehan et al., “Obstructive sleep apnea and obesity: implications for public health,” Sleep Medicine and Disorders: International Journal, vol. 1, no. 4, Art. no. 00019, 2017.
[6] S. P. Patil, H. Schneider, A. R. Schwartz, and P. L. Smith, “Adult obstructive sleep apnea: pathophysiology and diagnosis,” Chest, vol. 132, no. 1, pp. 325–337, 2007.
[7] R. L. Horner, “Pathophysiology of obstructive sleep apnea,” Journal of Cardiopulmonary Rehabilitation and Prevention, vol. 28, no. 5, pp. 289–298, Sep.–Oct. 2008.
[8] K. Gunhan, “Pathophysiology of obstructive sleep apnea,” in Sleep Apnea: Impacts, Treatment and Management, Springer, 2013, pp. 313–329.
[9] M. S. Badr, “Pathophysiology of upper airway obstruction during sleep,” Clinics in Chest Medicine, vol. 19, no. 1, pp. 21–32, 1998.
[10] B. McGinley, A. R. Schwartz, H. Schneider, J. Kirkness, P. L. Smith, and S. P. Patil, “Upper airway neuromuscular compensation during sleep is defective in obstructive sleep apnea,” Journal of Applied Physiology, vol. 105, no. 1, pp. 197–205, 2008.
[11] S. P. Patil, H. Schneider, J. J. Marx, E. Gladmon, A. R. Schwartz, and P. L. Smith, “Neuromechanical control of upper airway patency during sleep,” Journal of Applied Physiology, vol. 102, no. 2, pp. 547–556, 2007.
[12] E. S. Katz and D. P. White, “Genioglossus activity in children with obstructive sleep apnea during wakefulness and sleep onset,” American Journal of Respiratory and Critical Care Medicine, vol. 168, no. 6, pp. 664–670, 2003.
[13] G. A. Silva, H. H. Sander, A. L. Eckeli, R. M. F. Fernandes, E. B. Coelho, and F. Nobre, “Conceitos básicos sobre síndrome da apneia obstrutiva do sono Basic concepts about obstructive sleep apnea,” Revista Brasileira de Hipertensão, vol. 16, no. 3, pp. 150–157, 2009.
[14] S. M. Caples, W. M. Anderson, K. Calero, M. Howell, and S. D. Hashmi, “Use of polysomnography and home sleep apnea tests for the longitudinal management of obstructive sleep apnea in adults: An American Academy of Sleep Medicine position paper,” Clinical and Experimental Otorhinolaryngology vol. 17, no. 7, pp. 1503–1507, 2021.
[15] R. B. Berry et al., “AASM Scoring Manual Updates for 2017 (Version 2.4),” Journal of Clinical Sleep Medicine, vol. 13, no. 5, pp. 665–666, 2017.
[16] T. Mitterling et al., “Sleep and respiration in 100 healthy Caucasian sleepers—A polysomnographic study according to American Academy of Sleep Medicine Standards,” Sleep, vol. 38, no. 6, pp. 867–875, 2015.
[17] D. J. Gottlieb and N. M. Punjabi, “Diagnosis and management of obstructive sleep apnea: A Review,” JAMA, vol. 323, no. 14, pp. 1389–1400, Apr. 2020.
[18] S. P. Patil, L. J. Ayappa, L. J. Caples, D. P. Kimoff, K. M. Patel, and T. G. Harrod, “Treatment of adult obstructive sleep apnea with positive airway pressure: an American Academy of Sleep Medicine systematic review, meta-analysis, and GRADE assessment,” Journal of Clinical Sleep Medicine, vol. 15, no. 2, pp. 301–334, 2019.
[19] H. S. Lin et al., “Treatment compliance in patients lost to follow-up after polysomnography,” Otolaryngology—Head and Neck Surgery: Official Journal of American Academy of Otolaryngology-Head and Neck Surgery, vol. 136, no. 2, pp. 236–240, 2007.
[20] T. E. Weaver and R. R. Grunstein, “Adherence to continuous positive airway pressure therapy: the challenge to effective treatment,” Proceedings of the American Thoracic Society, vol. 5, no. 2, pp. 173–178, Feb. 2008.
[21] K. Ramar, L. C. Dort, S. G. Katz, C. J. Lettieri, C. G. Harrod, S. M. Thomas, and R. D. Chervin, “Clinical practice guideline for the treatment of obstructive sleep apnea and snoring with oral appliance therapy: An update for 2015,” Journal of Clinical Sleep Medicine, vol. 11, no. 7, pp. 773–827, Jul. 2015.
[22] S. Y. Liu, R. W. Riley, and M. S. Yu, “Surgical algorithm for obstructive sleep apnea: an update,” Clinical and Experimental Otorhinolaryngology, vol. 13, no. 3, pp. 215–224, Aug. 2020.
[23] P. Ish, V. Rathi, S. Sharma, and S. Mitra, “Myofunctional therapy for obstructive sleep apnea: The ignored adjunct,” Indian Journal of Sleep Medicine, 2024.
[24] Y. Xu, R. Yang, M. Yu, and X. Gao, “Efficacy of myofunctional therapy for obstructive sleep apnea: A systematic review and network meta-analysis,” The Journal of Evidence-Based Dental Practice, vol. 25, 2025.
[25] J. Y. Kim et al., “Machine learning-based preoperative datamining can predict the therapeutic outcome of sleep surgery in OSA subjects,” Scientific Reports, vol. 11, no. 1, p. 14911, Jul. 2021.
[26] J. You et al., “Postoperative apnea‐hypopnea index prediction of velopharyngeal surgery based on machine learning,” OTO Open, vol. 9, no. 1, p. e70061, Jan. 2025.
[27] T. Wen, H. Chen, and K. H. Cheong, “Visibility graph for time series prediction and image classification: a review,” Nonlinear Dynamics, vol. 110, pp. 2979–2999, 2022.
[28] N. Gupta, H. Singh, and J. Singla, “Fuzzy logic-based systems for medical diagnosis: a review,” in Proc. 3rd Int. Conf. Electronics and Sustainable Communication Systems (ICESC), pp. 1125–1130, 2022.
[29] M. G. Crowson et al., “A contemporary review of machine learning in otolaryngology-head and neck surgery,” The Laryngoscope, vol. 130, no. 1, pp. 45–51, 2020.
[30] L. Lacasa, B. Luque, F. Ballesteros, J. Luque, and J. C. Nuño, “From time series to complex networks: The visibility graph,” Proceedings of the National Academy of Sciences of the United States of America, vol. 105, no. 13, pp. 4972–4975, 2008.
[31] J. Zhang, J. Xia, X. Liu, and J. Olichney, “Machine learning on visibility graph features discriminates the cognitive event-related potentials of patients with early Alzheimer’s disease from healthy aging,” Brain Sciences, vol. 13, no. 5, p. 770, May 2023.
[32] G. Zhu, Y. Li, and P. Wen, “Analysis and classification of sleep stages based on difference visibility graphs from a single-channel EEG signal,” IEEE Journal of Biomedical and Health Informatics, vol. 18, no. 6, pp. 1813–1821, Jan. 2014.
[33] L. A. Zadeh, “Fuzzy sets,” Information and Control, vol. 8, no. 3, pp. 338–353, 1965
[34] E. H. Mamdani and S. Assilian, “An experiment in linguistic synthesis with a fuzzy logic controller,” International Journal of Man-Machine Studies, vol. 7, no. 1, pp. 1–13, 1975.
[35] G. Ahmad, M. A. Khan, S. Abbas, A. Athar, B. S. Khan, and M. S. Aslam, “Automated diagnosis of hepatitis B using multilayer Mamdani fuzzy inference system,” Journal of Healthcare Engineering, vol. 2019, pp. 1–11, Feb. 2019.
[36] M. Shatnawi, A. Shatnawi, Z. AlShara, and G. Husari, “Symptoms-based fuzzy-logic approach for COVID-19 diagnosis,” International Journal of Advanced Computer Science and Applications (IJACSA), vol. 12, no. 4, 2021.
[37] I. Iancu, “Heart disease diagnosis based on mediative fuzzy logic,” Artificial Intelligence in Medicine, vol. 89, pp. 51–60, Jul. 2018.
[38] S. Supriya, S. Siuly, H. Wang, J. Cao, and Y. Zhang, “Weighted visibility graph with complex network features in the detection of epilepsy,” IEEE Access, vol. 4, pp. 6554–6566, 2016.
[39] J.-P. Onnela et al., “Structure and tie strengths in mobile communication networks,” Proceedings of the National Academy of Sciences of the United States of America, vol. 104, no. 18, pp. 7332–7336, May 2007.
[40] M. E. J. Newman, “Analysis of weighted networks,” Physical Review E, vol. 70, no. 5, p. 056131, Nov. 2004.
[41] C.-T. Lin, “Characterization of polysomnogram signals for surgical response in obstructive sleep apnea using machine learning,” M.S. thesis, Department of Biomedical Engineering, National Cheng Kung University, Tainan, Taiwan, 2024.
[42] C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, no. 3, pp. 273–297, Sep. 1995.
[43] A. J. Smola and B. Schölkopf, “A tutorial on support vector regression,” Statistics and Computing, vol. 14, no. 3, pp. 199–222, Aug. 2004.
[44] R. Rodríguez-Pérez and J. Bajorath, “Evolution of support vector machine and regression modeling in chemoinformatics and drug discovery,” Journal of Computer-Aided Molecular Design, vol. 36, no. 5, pp. 1–8, May 2022.
[45] I. Guyon and A. Elisseeff, "An Introduction to Variable and Feature Selection," Journal of Machine Learning Research, vol. 3, pp. 1157–1182, 2003.
[46] A. E. Sher, K. B. Schechtman, and J. F. Piccirillo, “The efficacy of surgical modifications of the upper airway in adults with obstructive sleep apnea syndrome,” Sleep, vol. 19, no. 2, pp. 156–177, Mar. 1996.