肌肉質量的多寡會影響肌肉的耐受度與疲勞的產生，而肌肉質量會隨者年齡的增長而逐漸下降，其中下肢肌肉質量的流失，會使下肢更易疲勞、平衡能力下降、步態不穩定性上升造成跌倒風險增加。於工業安全與長者安全上，將是現代社會不容忽視的問題。而對於動態平衡與步行移動之下肢肌群中，以小腿後側腓腸肌最為重要。本研究針對雙側腓腸肌疲勞前後，搭配表面肌電訊號之中位頻率變化確認疲勞狀態，並透過慣性感測元件（Inertial Measurement Unit, IMU）蒐集身體6個部位於步態中的之加速度與角速度等資訊特徵。使用深度學習技術辨識下肢肌肉疲勞之步態，找出對於疲勞步態辨識最為準確之部位。期許未來能進一步應用至工業或醫療場域中，將有助於提升工業和日常生活的安全性。
本研究之對象為16名年齡範圍20至30歲之健康成年人，男性與女性各8名。將IMU黏貼於雙腳腳跟與腳趾、薦椎、頭部共6個部位後，請受試者於70公尺的走廊行走。實驗分二日進行，間隔七日。首日受試者分別於下肢疲勞誘發前後於長廊進行自覺舒適速度之步態及疲勞步態測試。第二日則請受使者以首日疲勞後之步頻行走，以模擬疲勞後之步態，幫助確認肌肉疲勞對於步態的影響。研究結果顯示，透過IMU進行步態的辨識，於疲勞前、疲勞後即模擬步頻三種情況下；模擬步頻之辨識準確率於各IMU組合下，皆高於另二種情況，顯示腓腸肌的疲勞對於步態模式的影響確實對於非疲勞時的情況下有明顯的差異。於步態三種情況下之辨識準確度，身體各部位中辨識準確度以腳趾最高，頭部為最低；而於各IMU組合之下，腳趾與薦椎的搭配之下有最高之辨識準確度，顯示薦椎與足部的相互變化對於步態的辨識有較佳的學習效果。
本研究之結果可以得知，使用深度學習技術LSTM及透過單一部位之IMU即可辨識此步態是屬於何種狀態。一般正常生理狀態，或是小腿肌肉疲勞狀態，亦或是模擬疲勞步頻下，肌肉未疲勞的狀態。未來研究中將加入更多不同年齡層之受試者，及搭配不同活動的疲勞步態，幫助增加未來應用上之可行性，並讓能應用之場所更加多元。

Sarcopenia is a type of muscle mass loss that occurs with aging. The amount of muscle mass affects the tolerance of a muscle towards fatigue. Muscle mass loss and fatigue, especially in lower limbs, induce walking imbalance and gait instability, increasing the risk of falling. Lower limb muscles, such as gastrocnemius muscles, are important for human locomotion and daily activities. The study aims to use a deep learning technique to recognize the data collecting from inertial measurement units (IMUs) for fatigued gait. Second, the study aims to reveal the location or the number of IMUs can have the best performance.
Sixteen healthy adults (24.4 ± 2.07) participated in the experiment. Six IMUs were attached to each subject’s heels, toes, sacrum, and head. The research is a two-day process with a week interval. On the first day, the subjects were instructed to walk along a hallway before and after the fatigue protocol as a non-fatigued and fatigue gait. On the second day, the subjects were instructed to walk along a hallway following the beat of their fatigue gait cadence measured on the first day. The result revealed that the LSTM model could recognize the gait of simulated cadence with the highest accuracy among these three conditions (non-fatigued, fatigue, and simulated cadence). For the IMU location, the result showed that the IMUs attached to the toes had the highest accuracy. For multiple IMUs combinations, the result showed that the IMUs of toes and sacrum achieved the highest accuracy among other combinations.
In conclusion, the deep learning technique of LSTM with one or more IMUs can recognize the gait under normal, physical fatigue, or simulated cadence without muscle fatigue. For future researches and applications, more subjects from different ages would be expected to collect and along with other walking conditions to improve the feasibility in multiple fields such as industrial and elderly safety.

摘要I
AbstractII
目錄III
圖目錄V
表目錄VI
第一章緒論1
1.1.研究背景與動機1
1.2.研究範圍與目的4
1.3.研究架構與流程4
第二章文獻回顧5
2.1.疲勞5
2.1.1疲勞之種類5
2.1.2肌肉的收縮機制6
2.1.3肌肉疲勞與肌少症8
2.2.步態9
2.2.1步態中下肢肌肉之作用10
2.2.2下肢疲勞對於平衡與步態的影響12
2.3.步態參數的獲取14
2.3.1慣性感測元件14
2.3.2表面肌電訊號14
2.4.基於步態參數之疲勞辨識15
2.5.小結17
第三章研究方法19
3.1.實驗對象與流程19
3.1.1疲勞誘發20
3.2.實驗設備22
3.3.前導實驗24
3.4.訊號處理28
3.4.1步態週期切割28
3.4.2肌電訊號分析30
3.5.模型架構與測試30
第四章實驗結果34
4.1.受試者疲勞誘發運動34
4.2.疲勞狀態確認34
4.3.步態參數變化35
4.4.模型辨識成效38
4.4.1身體各部位辨識成效39
4.4.2各IMU組合辨識成效43
第五章討論48
5.1.疲勞後肌電訊號48
5.2.疲勞後與模擬步頻之步態參數48
5.3.模型成效49
5.4.模擬步頻辨識準確率50
5.5.IMU 擺放位置和數量影響51
5.5.1身體部位選擇51
5.5.2IMU數量組合選擇53
5.6.潛在應用54
5.7.研究限制55
第六章結論與未來方向57
參考文獻58
附錄一基本步態參數數值63
附錄二研究倫理審查通知書64

1. 吳蔓君. （2015）. 肌少症簡介. 家庭醫學與基層醫療, 30（4）, 103-107.
2. 蔡政龍. （2006）. 肌電圖強度與速度分析於機器手臂控制之應用. 國立交通大學電機與控制工程系所碩士論文.
3. Anwary, A. R., Yu, H., & Vassallo, M. (2018). Optimal foot location for placing wearable IMU sensors and automatic feature extraction for gait analysis. IEEE Sensors Journal, 18(6), 2555-2567.
4. Baghdadi, A., Megahed, F. M., Esfahani, E. T., & Cavuoto, L. A. (2018). A machine learning approach to detect changes in gait parameters following a fatiguing occupational task. Ergonomics, 61(8), 1116-1129.
5. Barbieri, F. A., Dos Santos, P. C. R., Lirani-Silva, E., Vitório, R., Gobbi, L. T. B., & Van Diëen, J. H. (2013). Systematic review of the effects of fatigue on spatiotemporal gait parameters. Journal of back and musculoskeletal rehabilitation, 26(2), 125-131.
6. Bogdanis, G. C. (2012). Effects of physical activity and inactivity on muscle fatigue. Frontiers in physiology, 3, 142.
7. Boyas, S., Hajj, M., & Bilodeau, M. (2013). Influence of ankle plantarflexor fatigue on postural sway, lower limb articular angles, and postural strategies during unipedal quiet standing. Gait & posture, 37(4), 547-551.
8. Chen, L.-K., Woo, J., Assantachai, P., Auyeung, T.-W., Chou, M.-Y., Iijima, K., . . . Kim, S. (2020). Asian Working Group for Sarcopenia: 2019 consensus update on sarcopenia diagnosis and treatment. Journal of the American Medical Directors Association, 21(3), 300-307. e302.
9. Chen, W.-M., Park, J., Park, S.-B., Shim, V. P.-W., & Lee, T. (2012). Role of gastrocnemius–soleus muscle in forefoot force transmission at heel rise—A 3D finite element analysis. Journal of biomechanics, 45(10), 1783-1789.
10. Cruz-Jentoft, A. J., Baeyens, J. P., Bauer, J. M., Boirie, Y., Cederholm, T., Landi, F., . . . Schneider, S. M. (2010). Sarcopenia: European consensus on definition and diagnosisReport of the European Working Group on Sarcopenia in Older PeopleA. J. Cruz-Gentoft et al. Age and ageing, 39(4), 412-423.
11. Dadashi, F., Mariani, B., Rochat, S., Büla, C. J., Santos-Eggimann, B., & Aminian, K. (2014). Gait and foot clearance parameters obtained using shoe-worn inertial sensors in a large-population sample of older adults. Sensors, 14(1), 443-457.
12. Dimitrova, N., Arabadzhiev, T., Hogrel, J.-Y., & Dimitrov, G. (2009). Fatigue analysis of interference EMG signals obtained from biceps brachii during isometric voluntary contraction at various force levels. Journal of Electromyography and Kinesiology, 19(2), 252-258.
13. Gadaleta, M., & Rossi, M. (2018). Idnet: Smartphone-based gait recognition with convolutional neural networks. Pattern Recognition, 74, 25-37.
14. Gallagher, D., Visser, M., De Meersman, R. E., Sepúlveda, D., Baumgartner, R. N., Pierson, R. N., . . . Heymsfield, S. B. (1997). Appendicular skeletal muscle mass: effects of age, gender, and ethnicity. Journal of applied physiology, 83(1), 229-239.
15. Gard, S. A., & Childress, D. S. (1999). The influence of stance-phase knee flexion on the vertical displacement of the trunk during normal walking. Archives of physical medicine and rehabilitation, 80(1), 26-32.
16. Gimmon, Y., Riemer, R., Oddsson, L., & Melzer, I. (2011). The effect of plantar flexor muscle fatigue on postural control. Journal of Electromyography and Kinesiology, 21(6), 922-928.
17. Hirasaki, E., Moore, S. T., Raphan, T., & Cohen, B. (1999). Effects of walking velocity on vertical head and body movements during locomotion. Experimental brain research, 127(2), 117-130.
18. Hlavackova, P., & Vuillerme, N. (2012). Do somatosensory conditions from the foot and ankle affect postural responses to plantar-flexor muscles fatigue during bipedal quiet stance? Gait & posture, 36(1), 16-19.
19. Hunt, M. A., & Hatfield, G. L. (2017). Ankle and knee biomechanics during normal walking following ankle plantarflexor fatigue. Journal of Electromyography and Kinesiology, 35, 24-29.
20. Hwang, A.-C., Liu, L.-K., Lee, W.-J., Peng, L.-N., & Chen, L.-K. (2018). Calf circumference as a screening instrument for appendicular muscle mass measurement. Journal of the American Medical Directors Association, 19(2), 182-184.
21. Hwang, T.-H., Reh, J., Effenberg, A. O., & Blume, H. (2018). Real-time gait analysis using a single head-worn inertial measurement unit. IEEE Transactions on Consumer Electronics, 64(2), 240-248.
22. Janssen, I., Heymsfield, S. B., Wang, Z., & Ross, R. (2000). Skeletal muscle mass and distribution in 468 men and women aged 18–88 yr. Journal of applied physiology.
23. Karthick, P., Ghosh, D. M., & Ramakrishnan, S. (2018). Surface electromyography based muscle fatigue detection using high-resolution time-frequency methods and machine learning algorithms. Computer methods and programs in biomedicine, 154, 45-56.
24. Kavanagh, J., Barrett, R., & Morrison, S. (2006). The role of the neck and trunk in facilitating head stability during walking. Experimental brain research, 172(4), 454.
25. Kavanagh, J. J., Barrett, R. S., & Morrison, S. (2005). Age-related differences in head and trunk coordination during walking. Human movement science, 24(4), 574-587.
26. Kim, S., Soangra, R., Grant-Beuttler, M., & Aminian, A. (2019). Wearable Sensor-based Gait Classification in Idiopathic Toe Walking Adolescents. Biomedical sciences instrumentation, 55(2), 178.
27. Köse, A., Cereatti, A., & Della Croce, U. (2012). Bilateral step length estimation using a single inertial measurement unit attached to the pelvis. Journal of neuroengineering and rehabilitation, 9(1), 1-10.
28. Kuo, Y.-H., Wang, T.-F., Liu, L.-K., Lee, W.-J., Peng, L.-N., & Chen, L.-K. (2019). Epidemiology of sarcopenia and factors associated with it among community-dwelling older adults in Taiwan. The American Journal of the Medical Sciences, 357(2), 124-133.
29. Langley, P. (1994). Selection of relevant features in machine learning. Proceedings of the AAAI Fall symposium on relevance,
30. Latt, M. D., Menz, H. B., Fung, V. S., & Lord, S. R. (2008). Walking speed, cadence and step length are selected to optimize the stability of head and pelvis accelerations. Experimental brain research, 184(2), 201-209.
31. Lewis, C. L., Laudicina, N. M., Khuu, A., & Loverro, K. L. (2017). The human pelvis: variation in structure and function during gait. The Anatomical Record, 300(4), 633-642.
32. Lim, Y. P., Lin, Y. C., & Pandy, M. G. (2013). Muscle function during gait is invariant to age when walking speed is controlled. Gait & posture, 38(2), 253-259.
33. Lu, Y., Wang, H., Qi, Y., & Xi, H. (2021). Evaluation of classification performance in human lower limb jump phases of signal correlation information and LSTM models. Biomedical Signal Processing and Control, 64, 102279.
34. MacIntosh, B. R., Holash, R. J., & Renaud, J.-M. (2012). Skeletal muscle fatigue–regulation of excitation–contraction coupling to avoid metabolic catastrophe. Journal of cell science, 125(9), 2105-2114.
35. Malmstrom, T. K., & Morley, J. E. (2013). SARC-F: a simple questionnaire to rapidly diagnose sarcopenia. Journal of the American Medical Directors Association, 14(8), 531-532.
36. Mannion, A. F., & Dolan, P. (1994). Electromyographic median frequency changes during isometric contraction of the back extensors to fatigue. Spine, 19(11), 1223-1229.
37. Marcora, S. M., Staiano, W., & Manning, V. (2009). Mental fatigue impairs physical performance in humans. Journal of applied physiology, 106(3), 857-864.
38. Maslivec, A., Bampouras, T., Dewhurst, S., Vannozzi, G., Macaluso, A., & Laudani, L. (2018). Mechanisms of head stability during gait initiation in young and older women: a neuro-mechanical analysis. Journal of Electromyography and Kinesiology, 38, 103-110.
39. Mason, J. E., Traoré, I., & Woungang, I. (2016). Machine Learning Techniques for Gait Biometric Recognition. Springer.
40. Mehta, R. K., & Parasuraman, R. (2014). Effects of mental fatigue on the development of physical fatigue: a neuroergonomic approach. Human factors, 56(4), 645-656.
41. Millar, C., Siddique, N., & Kerr, E. (2020). LSTM Classification of sEMG Signals For Individual Finger Movements Using Low Cost Wearable Sensor. 2020 International Symposium on Community-centric Systems (CcS),
42. Murad, A., & Pyun, J.-Y. (2017). Deep recurrent neural networks for human activity recognition. Sensors, 17(11), 2556.
43. Neptune, R. R., Kautz, S., & Zajac, F. (2001). Contributions of the individual ankle plantar flexors to support, forward progression and swing initiation during walking. Journal of biomechanics, 34(11), 1387-1398.
44. Nilwik, R., Snijders, T., Leenders, M., Groen, B. B., van Kranenburg, J., Verdijk, L. B., & van Loon, L. J. (2013). The decline in skeletal muscle mass with aging is mainly attributed to a reduction in type II muscle fiber size. Experimental gerontology, 48(5), 492-498.
45. Olson, R. L., Chang, Y.-K., Brush, C. J., Kwok, A. N., Gordon, V. X., & Alderman, B. L. (2016). Neurophysiological and behavioral correlates of cognitive control during low and moderate intensity exercise. NeuroImage, 131, 171-180.
46. Parijat, P., & Lockhart, T. E. (2008). Effects of quadriceps fatigue on the biomechanics of gait and slip propensity. Gait & posture, 28(4), 568-573.
47. Puetz, T. W. (2006). Physical activity and feelings of energy and fatigue. Sports medicine, 36(9), 767-780.
48. Requiao, L., Nadeau, S., Milot, M., Gravel, D., Bourbonnais, D., & Gagnon, D. (2005). Quantification of level of effort at the plantarflexors and hip extensors and flexor muscles in healthy subjects walking at different cadences. Journal of Electromyography and Kinesiology, 15(4), 393-405.
49. Rieser, J. J., Pick, H. L., Ashmead, D. H., & Garing, A. E. (1995). Calibration of human locomotion and models of perceptual-motor organization. Journal of Experimental Psychology: Human Perception and Performance, 21(3), 480.
50. Roerdink, M., Hlavackova, P., & Vuillerme, N. (2011). Effects of plantar-flexor muscle fatigue on the magnitude and regularity of center-of-pressure fluctuations. Experimental brain research, 212(3), 471-476.
51. Santos, P. C. R. d., Barbieri, F. A., Zijdewind, I., Gobbi, L. T. B., Lamoth, C., & Hortobágyi, T. (2019). Effects of experimentally induced fatigue on healthy older adults’ gait: A systematic review. PLoS One, 14(12), e0226939.
52. Sherratt, F., Plummer, A., & Iravani, P. (2021). Understanding LSTM network behaviour of IMU-based locomotion mode recognition for applications in prostheses and wearables. Sensors, 21(4), 1264.
53. Steuer, I., & Guertin, P. A. (2019). Central pattern generators in the brainstem and spinal cord: an overview of basic principles, similarities and differences. Reviews in the Neurosciences, 30(2), 107-164.
54. Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles, SENIAM. http://www.seniam.org/
55. Svoboda, Z., Janura, M., Kutilek, P., & Janurova, E. (2016). Relationships between movements of the lower limb joints and the pelvis in open and closed kinematic chains during a gait cycle. Journal of human kinetics, 51(1), 37-43.
56. Tunca, C., Pehlivan, N., Ak, N., Arnrich, B., Salur, G., & Ersoy, C. (2017). Inertial sensor-based robust gait analysis in non-hospital settings for neurological disorders. Sensors, 17(4), 825.
57. Ulloa, J. G. (2018). Applied Biomechatronics Using Mathematical Models. Academic press.
58. Wan, J.-j., Qin, Z., Wang, P.-y., Sun, Y., & Liu, X. (2017). Muscle fatigue: general understanding and treatment. Experimental & molecular medicine, 49(10), e384-e384.
59. Wang, R., Fukuda, D. H., Stout, J. R., Robinson, E. H., Miramonti, A. A., Fragala, M. S., & Hoffman, J. R. (2015). Evaluation of electromyographic frequency domain changes during a three-minute maximal effort cycling test. Journal of sports science & medicine, 14(2), 452.
60. Wu, C.-C., Chen, Y.-J., Hsu, C.-S., Wen, Y.-T., & Lee, Y.-J. (2020). Multiple Inertial Measurement Unit Combination and Location for Center of Pressure Prediction in Gait. Frontiers in Bioengineering and Biotechnology, 8, 1252.
61. Xing, X., Zhong, B., Luo, H., Rose, T., Li, J., & Antwi-Afari, M. F. (2020). Effects of physical fatigue on the induction of mental fatigue of construction workers: A pilot study based on a neurophysiological approach. Automation in Construction, 120, 103381.
62. Yang, K., Ahn, C. R., & Kim, H. (2020). Deep learning-based classification of work-related physical load levels in construction. Advanced Engineering Informatics, 45, 101104.
63. Zhang, J., Lockhart, T. E., & Soangra, R. (2014). Classifying lower extremity muscle fatigue during walking using machine learning and inertial sensors. Annals of biomedical engineering, 42(3), 600-612.
64. Zhao, A., Qi, L., Dong, J., & Yu, H. (2018). Dual channel LSTM based multi-feature extraction in gait for diagnosis of Neurodegenerative diseases. Knowledge-Based Systems, 145, 91-97.
65. Zhu, X., & Wu, X. (2004). Class noise vs. attribute noise: A quantitative study. Artificial intelligence review, 22(3), 177-210.