臺灣博碩士論文加值系統

人體的前庭系統是負責感知頭部運動的重要系統，可為大腦提供有關空間和平衡方面的訊息。前庭功能衰弱引起的症狀包含頭暈、眩暈和身體不平衡，可能會伴隨凝視不穩定和視力模糊，甚至導致噁心、嘔吐和注意力不集中，進而降低病患的活動水平和社交互動。臨床上，前庭功能減退通常使用前庭復健療法來緩解頭暈和身體不平衡的症狀。在此療法中，病患通常會被要求在家中執行前庭復健運動，例如凝視穩定度的訓練。然而，在居家環境中，臨床治療師很難追蹤病患在前庭復健運動中的表現，進而影響復健進程和效果，因此，開發一個基於居家環境的前庭復健系統以追蹤和評估動作的表現是非常重要的。
本研究的目標是開發一個基於居家環境的前庭復健評估系統，以協助病患及臨床人員有效地追蹤復健進程並提供更多客觀的量化資訊。研究分為兩個部分：第一部分利用穿戴式感測器收集受試者的前庭復健動作資料，透過運動學特徵的擷取評估受試者的身體控制能力，進而對受試者平衡能力和表現進行統計分析和分類；接著，第二部分是針對在統計分析結果上有顯著差異的運動學特徵，利用機器學習演算法，分辨前庭復健運動是由病患或健康受試者完成的，用以輔助辨識前庭復健功能衰弱的病患。統計分析的結果顯示，在運動學特徵方面，健康受試者和患者之間存在顯著差異；而前庭復健運動功能性辨識的結果顯示，選用統計分析中，有顯著差異的特定運動學特徵可以有效的區分受試者的組別，這有助於提供臨床人員客觀的評估病患狀況並制定有效的前庭復健進程。
該系統提供一個可靠、有效和可行的前庭復健評估系統，以促進病患復健過程的效果和提高臨床人員的工作效率。

Vestibular system is responsible for sensing head movement and providing information about space and balance to the brain. Vestibular hypofunction can cause symptoms include dizziness, vertigo, and body imbalance, accompanied by unstable gaze, blurred vision, nausea, vomiting, and reduced attention, ultimately reducing patients' activity levels and social interaction. Vestibular rehabilitation therapy (VRT) is commonly used to alleviate these symptoms. Involving to perform vestibular rehabilitation exercises, such as gaze stabilization exercise, at home environment. However, tracking patients' performance in a home environment is challenging for therapists, thereby impacting the rehabilitation progress and outcomes. Therefore, a home-based vestibular rehabilitation system for tracking and evaluating movement performance is crucial.
This study aims to develop a home-based vestibular rehabilitation assessment system to effectively evaluate the rehabilitation progress and providing more objective quantitative information. It consists of two parts: First, the kinematic features are extracted from the vestibular rehabilitation movement data collected from participants and performs statistical analysis. Second, machine learning algorithms are utilized to classify movements as performed by patients or healthy participants based on significant kinematic differences The statistical analysis reveals significant differences in kinematic features between healthy participants and patients. Moreover, functional classification of vestibular rehabilitation movements effectively distinguishes participant groups with only the specific kinematic features showing significant differences in the statistical analysis. These aids professionals in providing an objective assessment of patients' conditions and formulating effective VRT programs.
The proposed system offers a reliable, effective, and feasible vestibular rehabilitation assessment system to enhance the effectiveness of patients' rehabilitation process and improve the efficiency of clinical professionals.

致謝 i
摘要 ii
Abstract iii
目錄 iv
圖目錄 vi
表目錄 vii
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機 1
1.3 研究目的 2
1.4 論文架構 3
第二章 文獻回顧 4
2.1 前庭復健運動種類 4
2.2 前庭復健運動系統 8
2.2.1 數位健康輔助（Digital health intervention）應用程式 8
2.2.2 遊戲式前庭復健運動輔助系統（Game-based vestibular rehabilitation system） 8
2.2.3 穿戴式感測器系統（Wearable-based sensor system） 12
2.3 機器學習演算法 14
2.3.1 支援向量機 14
2.3.2 K最近鄰 14
2.3.3 隨機森林 15
2.3.4 極限梯度提升 15
第三章 材料與方法 16
3.1 系統架構 16
3.2 實驗設計 17
3.2.1 受試者 17
3.2.2 實驗設備 18
3.2.3 實驗流程 18
3.3 手動切割訊號與標記 22
3.4 運動學訊號特徵統計分析 22
3.4.1 資料集 22
3.4.2 運動學訊號特徵提取 22
3.4.3 統計分析 23
3.5 前庭復健運動功能性辨識 24
3.5.1 特徵提取 24
3.5.2 機器學習分類器 24
3.5.3 系統表現評估 25
第四章 研究結果 27
4.1 前庭復健運動動作訊號手動切割 27
4.2 運動學訊號特徵統計分析 29
第五章 討論 38
5.1 運動學訊號特徵統計分析 38
5.2 前庭復健運動功能性辨識 40
第六章 結論 42
參考文獻 43

[1] K. G. Gassmann and R. Rupprecht, ‘Dizziness in an older community dwelling population: A multifactorial syndrome’, J. Nutr. Health Aging, vol. 13, no. 3, pp. 278–282, Mar. 2009, doi: 10.1007/s12603-009-0073-2.
[2] E. Dannenbaum, N. Paquet, G. Chilingaryan, and J. Fung, ‘Clinical Evaluation of Dynamic Visual Acuity in Subjects With Unilateral Vestibular Hypofunction’, Otol. Neurotol., vol. 30, no. 3, pp. 368–372, Apr. 2009, doi: 10.1097/MAO.0b013e31819bda35.
[3] S. J. Herdman, M. C. Schubert, V. E. Das, and R. J. Tusa, ‘Recovery of Dynamic Visual Acuity in Unilateral Vestibular Hypofunction’, Arch. Otolaryngol. Neck Surg., vol. 129, no. 8, p. 819, Aug. 2003, doi: 10.1001/archotol.129.8.819.
[4] S. L. Hillier and M. McDonnell, ‘Vestibular rehabilitation for unilateral peripheral vestibular dysfunction’, in Cochrane Database of Systematic Reviews, The Cochrane Collaboration, Ed., Chichester, UK: John Wiley & Sons, Ltd, 2011, p. CD005397.pub3. doi: 10.1002/14651858.CD005397.pub3.
[5] F. Joseph M, ‘Practical management of the balance disorder patient’, Audiol. Hamilt., vol. 36, no. 4, p. 237, Aug. 1997.
[6] M. Strupp and T. Brandt, ‘Vestibular Neuritis’, Semin. Neurol., vol. 29, no. 05, pp. 509–519, Nov. 2009, doi: 10.1055/s-0029-1241040.
[7] S. J. Herdman, C. D. Hall, M. C. Schubert, V. E. Das, and R. J. Tusa, ‘Recovery of Dynamic Visual Acuity in Bilateral Vestibular Hypofunction’, Arch. Otolaryngol. Neck Surg., vol. 133, no. 4, p. 383, Apr. 2007, doi: 10.1001/archotol.133.4.383.
[8] H. S. Cohen, ‘Disability and rehabilitation in the dizzy patient’, Curr. Opin. Neurol., vol. 19, no. 1, pp. 49–54, Feb. 2006, doi: 10.1097/01.wco.0000194373.08203.33.
[9] T. Cawthorne, ‘Vestibular Injuries’, Proc. R. Soc. Med., vol. 39, no. 5, pp. 270–273, Mar. 1946.
[10] T. Cawthorne, ‘The physiological basis for head exercises’, Chart. Soc Physiothe, vol. 29, pp. 106–107, 1944.
[11] F. Cooksey, ‘Physical medicine’, Practitione, vol. 155, pp. 300–305, 1945.
[12] F. S. Cooksey, ‘Rehabilitation in Vestibular Injuries’, Proc. R. Soc. Med., vol. 39, no. 5, pp. 273–278, Mar. 1946.
[13] H. C. Hecker, C. O. Haug, and J. W. Herndon, ‘Treatment of the vertiginous patient using cawthorne’s vestibular exercises: TREATMENT OF THE VERTIGINOUS PATIENT’, The Laryngoscope, vol. 84, no. 11, pp. 2065–2078, Nov. 1974, doi: 10.1002/lary.5540841121.
[14] H. S. Song and J. S. Kim, ‘Vestibular Rehabilitation Therapy: Review of Indications, Mechanisms, and Key Exercises’, J. Clin. Neurol., vol. 7, no. 4, pp. 184–196, Dec. 2011, doi: 10.3988/jcn.2011.7.4.184.
[15] E. Murray et al., ‘Evaluating Digital Health Interventions’, Am. J. Prev. Med., vol. 51, no. 5, pp. 843–851, Nov. 2016, doi: 10.1016/j.amepre.2016.06.008.
[16] D. Meldrum et al., ‘Toward a Digital Health Intervention for Vestibular Rehabilitation: Usability and Subjective Outcomes of a Novel Platform’, Front. Neurol., vol. 13, p. 836796, Mar. 2022, doi: 10.3389/fneur.2022.836796.
[17] S. J. Herdman, ‘Vestibular rehabilitation’:, Curr. Opin. Neurol., vol. 26, no. 1, pp. 96–101, Feb. 2013, doi: 10.1097/WCO.0b013e32835c5ec4.
[18] R. A. Roller and C. D. Hall, ‘A speed-based approach to vestibular rehabilitation for peripheral vestibular hypofunction: A retrospective chart review’, J. Vestib. Res., vol. 28, no. 3–4, pp. 349–357, Nov. 2018, doi: 10.3233/VES-180633.
[19] P.-Y. Chen, W.-L. Hsieh, S.-H. Wei, and C.-L. Kao, ‘Interactive wiimote gaze stabilization exercise training system for patients with vestibular hypofunction’, J. NeuroEngineering Rehabil., vol. 9, no. 1, p. 77, Dec. 2012, doi: 10.1186/1743-0003-9-77.
[20] P. Hovareshti et al., ‘VestAid: A Tablet-Based Technology for Objective Exercise Monitoring in Vestibular Rehabilitation’, Sensors, vol. 21, no. 24, p. 8388, Dec. 2021, doi: 10.3390/s21248388.
[21] S. Obdrzalek et al., ‘Accuracy and robustness of Kinect pose estimation in the context of coaching of elderly population’, in 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, CA: IEEE, Aug. 2012, pp. 1188–1193. doi: 10.1109/EMBC.2012.6346149.
[22] P. Bonato, ‘Wearable Sensors and Systems’, IEEE Eng. Med. Biol. Mag., vol. 29, no. 3, pp. 25–36, May 2010, doi: 10.1109/MEMB.2010.936554.
[23] S. C. Mukhopadhyay, ‘Wearable Sensors for Human Activity Monitoring: A Review’, IEEE Sens. J., vol. 15, no. 3, pp. 1321–1330, Mar. 2015, doi: 10.1109/JSEN.2014.2370945.
[24] Y. Delahoz and M. Labrador, ‘Survey on Fall Detection and Fall Prevention Using Wearable and External Sensors’, Sensors, vol. 14, no. 10, pp. 19806–19842, Oct. 2014, doi: 10.3390/s141019806.
[25] W. Tao, T. Liu, R. Zheng, and H. Feng, ‘Gait Analysis Using Wearable Sensors’, Sensors, vol. 12, no. 2, pp. 2255–2283, Feb. 2012, doi: 10.3390/s120202255.
[26] L. Freeman, G. Gera, F. B. Horak, M. T. Blackinton, M. Besch, and L. King, ‘Instrumented Test of Sensory Integration for Balance: A Validation Study’, J. Geriatr. Phys. Ther., vol. 41, no. 2, pp. 77–84, Apr. 2018, doi: 10.1519/JPT.0000000000000110.
[27] M. Ghislieri, L. Gastaldi, S. Pastorelli, S. Tadano, and V. Agostini, ‘Wearable Inertial Sensors to Assess Standing Balance: A Systematic Review’, Sensors, vol. 19, no. 19, p. 4075, Sep. 2019, doi: 10.3390/s19194075.
[28] S. Patel, H. Park, P. Bonato, L. Chan, and M. Rodgers, ‘A review of wearable sensors and systems with application in rehabilitation’, J. NeuroEngineering Rehabil., vol. 9, no. 1, p. 21, Dec. 2012, doi: 10.1186/1743-0003-9-21.
[29] L. Parrington, D. Jehu, P. Fino, S. Pearson, M. El-Gohary, and L. King, ‘Validation of an Inertial Sensor Algorithm to Quantify Head and Trunk Movement in Healthy Young Adults and Individuals with Mild Traumatic Brain Injury’, Sensors, vol. 18, no. 12, p. 4501, Dec. 2018, doi: 10.3390/s18124501.
[30] C. J. Todd, P. P. Hubner, P. Hubner, M. C. Schubert, and A. A. Migliaccio, ‘StableEyes—A Portable Vestibular Rehabilitation Device’, IEEE Trans. Neural Syst. Rehabil. Eng., vol. 26, no. 6, pp. 1223–1232, Jun. 2018, doi: 10.1109/TNSRE.2018.2834964.
[31] D. Martini, N. Pettigrew, J. Wilhelm, L. Parrington, and L. King, ‘Wearable Sensors for Vestibular Rehabilitation: A Pilot Study’, vol. 5, Jul. 2021.
[32] I. El Naqa and M. J. Murphy, ‘What Is Machine Learning?’, in Machine Learning in Radiation Oncology, I. El Naqa, R. Li, and M. J. Murphy, Eds., Cham: Springer International Publishing, 2015, pp. 3–11. doi: 10.1007/978-3-319-18305-3_1.
[33] S. Rauschert, K. Raubenheimer, P. E. Melton, and R. C. Huang, ‘Machine learning and clinical epigenetics: a review of challenges for diagnosis and classification’, Clin. Epigenetics, vol. 12, no. 1, p. 51, Dec. 2020, doi: 10.1186/s13148-020-00842-4.
[34] C. J. C. Burges, ‘A Tutorial on Support Vector Machines for Pattern Recognition’, SUPPORT VECTOR Mach..
[35] N. Cristianini and J. Shawe-Taylor, An Introduction to Support Vector Machines and Other Kernel-based Learning Methods, 1st ed. Cambridge University Press, 2000. doi: 10.1017/CBO9780511801389.
[36] B. Schölkopf and A. J. Smola, Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. The MIT Press, 2018. doi: 10.7551/mitpress/4175.001.0001.
[37] C. Cortes and V. Vapnik, ‘Support-vector networks’, Mach. Learn., vol. 20, no. 3, pp. 273–297, Sep. 1995, doi: 10.1007/BF00994018.
[38] N. S. Altman, ‘An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression’, Am. Stat., vol. 46, no. 3, p. 175, Aug. 1992, doi: 10.2307/2685209.
[39] T. Cover and P. Hart, ‘Nearest neighbor pattern classification’, IEEE Trans. Inf. Theory, vol. 13, no. 1, pp. 21–27, Jan. 1967, doi: 10.1109/TIT.1967.1053964.
[40] L. Breiman, ‘Random Forests’, Mach. Learn., vol. 45, no. 1, pp. 5–32, 2001, doi: 10.1023/A:1010933404324.
[41] A. Liaw and M. Wiener, ‘Classification and Regression by randomForest’, vol. 2, 2002.
[42] T. Chen and T. He, ‘xgboost: eXtreme Gradient Boosting’.
[43] T. Chen and C. Guestrin, ‘XGBoost: A Scalable Tree Boosting System’, in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco California USA: ACM, Aug. 2016, pp. 785–794. doi: 10.1145/2939672.2939785.
[44] J. H. Friedman, ‘Greedy function approximation: A gradient boosting machine.’, Ann. Stat., vol. 29, no. 5, Oct. 2001, doi: 10.1214/aos/1013203451.
[45] ‘Mobility Lab User Guide’. APDM Inc., Nov. 2020. [Online]. Available: https://share.apdm.com/documentation/MobilityLabUserGuide.pdf
[46] C.-C. Liu, K.-C. Ting, C.-Y. Hsieh, K.-C. Liu, and C.-T. Chan, ‘Wearable-based Vestibular Rehabilitation Monitoring Syste’, presented at the Global Conference on Biomedical Engineering (GCBME), Taiwan, Dec. 2022.
[47] D. C. Montgomery, ‘Design and Analysis of Experiments’.
[48] A. Bulling, U. Blanke, and B. Schiele, ‘A tutorial on human activity recognition using body-worn inertial sensors’, ACM Comput. Surv., vol. 46, no. 3, pp. 1–33, Jan. 2014, doi: 10.1145/2499621.
[49] D. Berrar, ‘Cross-Validation’, in Encyclopedia of Bioinformatics and Computational Biology, Elsevier, 2019, pp. 542–545. doi: 10.1016/B978-0-12-809633-8.20349-X.
[50] P. E. Taylor, G. J. M. Almeida, J. K. Hodgins, and T. Kanade, ‘Multi-label classification for the analysis of human motion quality’, in 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, CA: IEEE, Aug. 2012, pp. 2214–2218. doi: 10.1109/EMBC.2012.6346402.