尿失禁的復健治療方式分為侵入式與非侵入式的治療方式，侵入式的治療方式須將儀器侵入人體之中探測骨盆底肌肉的收縮程度，此方式是直接性的針對骨盆底肌肉的收縮程度進行分析與探討，而此復健方式需要侵入人體造成許多人無法接受；而非侵入式的治療方式主要以凱格爾運動為主，藉由提肛向上的力量收縮陰道、尿道口及肛門的肌肉來訓練骨盆底肌肉群，相較來說非侵入式的復健治療方式較大多數人接受，但無加裝任何感測器來感測患者數據。
在與醫師的合作下，已開發的無線感測復健輔具進行對病患的復健教學。進一步的希望病患將復健儀器帶回進行長期復健追蹤，因此在病患將儀器帶回家的過程中，經過病患建議與回饋後將解決下列問題，1.儀器較重與材質上堅硬導致在掉落時造成二次傷害。2.需要透過行動裝置才能觀看到復健的狀況，造成年長者的患者操作上的不便性。3.凱格爾復健的成效無法立即性的顯現出來，造成在復健過程中因此中斷。因此本論文將解決針對上述三個主要的問題進行改善，提升患者的接受度，並且預測分析患者的復健狀況，給予患者在復健期間的復健目標，並進行療程追蹤及分析。

In the incontinence treatment of rehabilitation isdivided to invasive therapy and non-invasive therapy. In the invasive therapytreatment, body will be invaded by instrument and detected the contraction. The method is analysis and discussion pelvic floor muscles contraction. The rehabilitation is needed to invade the human body caused most of people would not accept. In the non-invasive therapy treatment, the rehabilitation is based on Kegel exercise. By force of contraction of the levatorani up the vagina, urethra and anus muscle groups to train the pelvic floor muscles. The non-invasive therapy treatment is acceptable than invasive therapy treatment, but it isn’t embeded any sensors.
In collaboration withthe doctor, we had been developedwireless sensor network assistive device and taught patients. We want to track with patients for a long time, and we need to solve the problem as follows. First, because the assistive device is so heavy and hard, patients will hurt themselves by dropping. Second, she can view her achievement if she have mobile device. The older patients are difficult to use. Last, the Kegelexercise don’t show the effect immediately. Because of reasons, rehabilitationwill be interrupted. This thesissolves three problems of the assistive device, reducing the complexity of the operation and adding prediction patient exercisestatus to the achievement in the rehabilitation.

目錄
第一章緒論			1
1.1研究背景			2
1.2研究動機與目的			3
1.3論文架構			5
第二章相關研究			6
2.1尿失禁簡介			6
2.1.1尿失禁種類			7
2.1.2尿失禁治療方式		7
2.1.2.1藥物治療			8
2.1.2.2復健治療			9
2.1.2.3手術治療			10
2.2時間序列預測法			13
2.2.1時間序列定義及目的		13
2.2.2時間序列的性質		14
2.2.3時間序列的資料組成		15
2.2.4預估方法			17
第三章研究方法			21
3.1系統架構			22
3.1.1凱格爾運動訓練方式		23
3.1.2凱格爾復健輔具適用對象		24
3.2智慧型無線感測輔具系統		26
3.2.1坐姿凱格爾輔具設計		26
3.2.2無線感測模組			28
3.3互動回饋系統			30
3.4分析回饋系統			32
3.4.1復健數據分析-復健治療效果定義	32
3.4.2復健數據分析-短期預估成效	33
3.4.3復健數據分析-長期預估成效	35
第四章實作結果			38
4.1實作環境			38
4.2智慧型無線感測輔具系統		39
4.2.1坐姿凱格爾輔具設計		39
4.2.2硬體架構			40
4.2.3第一代vs第二代復健輔具		43
4.2.4追蹤方式			44
4.3互動回饋系統			46
4.4分析回饋系統			48
4.4.1短期預估成效			48
4.4.2長期預估成效			50
第五章結論與未來方向		55
5.1結論				55
5.2未來方向			56
參考文獻				57
附錄一英文論文			60

圖目錄
圖1第一代凱格爾訓練器材	4
圖2TVT手術示意圖	11
圖3平滑法預測圖	18
圖4趨勢法預測圖	19
圖5季節修正法預測趨勢線圖	20
圖6智慧型無線感測復健系統架構圖	23
圖7站姿凱格爾造成腿部及腰部負擔示意圖	24
圖8躺姿復健前後造成負擔示意圖	25
圖9坐姿凱格爾運動示意圖	26
圖10復健輔具設計示意圖	27
圖11復健輔具模擬示意圖	28
圖12壓力感測目標	29
圖13復健動作分析狀態圖	32
圖14復健狀況斜率等級圖	34
圖15短期預估成效	35
圖16最小平方法示意圖	36
圖17長期預估成效	37
圖18輔具設計過程	40
圖19智慧型無線感測硬體架構圖	41
圖20實體電路概要圖	43
圖21智慧型無線感測系統實體圖(A)智慧型無線感測輔具(B)復健系統(C)分析系統與回饋	45
圖22正確動作判定圖	47
圖23錯誤動作-收縮與放鬆時間皆不足圖	47
圖24患者一，實際追蹤短期預估成效	49
圖25患者二，實際追蹤短期預估成效	50
圖26患者三，實際追蹤短期預估成效	50
圖27患者一，實際追蹤26天時的復健趨勢圖	51
圖28患者一，實際追蹤69天時的復健趨勢圖	52
圖29患者二，實際追蹤58天時的復健趨勢圖	53
圖30患者三，實際追蹤58天時的復健趨勢圖	53

表目錄
表1動作判定正確與錯誤情況	30
表2實作環境	39
表3第一代與第二代復健輔具比較表	44
表4追蹤患者資料	48

[1]
經建會人力規劃處, “全球人口老化之現況與趨勢, ”人口老化專輯, 2013
[2]
陳浚琥, “膀胱之肌肉力量訓練器結構改良,”中華民國專利：M409083,2011.8.11
[3]
Huang, Ching-Hsiang, “Design and implementation of noninvasive electronic rehabilitation device for pelvic floor muscle training,”Tamkang University Tamsui, 2013
[4]
Amanda K. Berry, “Methylphenidate for Giggle Incontinence,” The Journal of Urology, Vol. 182, Issue 4, pp. 2028–2032, October 2009
[5]
RoselienHerderschee, “Feedback or biofeedback to augment pelvic floor muscle training for urinary incontinence in women,” The Cochrane Library, 2011
[6]
Natalia Price, “Pelvic floor exercise for urinary incontinence: A systematic literature review,”Maturitas, Vol. 67, Issue 4, pp. 309–315, Dec 2010
[7]
Kari,“Pelvic floor muscle training in treatment of female stress urinary incontinence, pelvic organ prolapse and sexual dysfunction,” World Journal of Urology, Vol. 30, Issue 4, pp. 437-443, Aug 2012
[8]
LadislavKrofta, “TVT and TVT-O for surgical treatment of primary stress urinary incontinence: prospective randomized trial,” International Urogynecology Journal, Vol. 21, Issue 2, pp. 141-148, Feb 2010
[9]
KobiStav, “Repeat Synthetic Mid Urethral Sling Procedure for Women With Recurrent Stress Urinary Incontinence,” The Journal of Urology, Vol. 183, Issue 1, pp. 241–246, Jan. 2010
[10]
Gautier Chene, Benjamin Cotte, Anne-Sylvie Tardieu, Denis Savary, AslamMansoor, “Clinical and ultrasonographic correlations following three surgical anti-incontinence procedures (TOT, TVT and TVT-O),”International 
Urogynecology Journal,2008
[11]
Giacomo Novara, “Updated Systematic Review and Meta-Analysis of the Comparative Data on Colposuspensions, Pubovaginal Slings, and Midurethral Tapes in the Surgical Treatment of Female Stress Urinary Incontinence,” European Urology, Vol. 58, Issue 2, pp. 218–238, Aug. 2010
[12]
Antonio M. Lopes, J.A. Tenreiro Machado, “Analysis of temperature time-series: Embedding dynamics into the MDS method,”Communications in Nonlinear Science and Numerical Simulation,Vol. 19,Issue 4, pp.851–871, 2014
[13]
Li Wang, Huiyan Zhang, Hong Xue, “Life Prediction based on Degradation Amount Distribution using Composite Time Series Analysis,”SciVerseScienceDirect, pp.217 – 224, 2012
[14]
Peng Guo, David Infield, and Xiyun Yang, “Wind Turbine Generator Condition-MonitoringUsing Temperature Trend Analysis,” IEEE Transactions on Sustainable Energy, vol. 3, no. 1, Jan.2012
[15]
Ida Wolden Bache, Tommy Sveen, and KjerstiNassTorstensen, “Revisiting the importance of non-tradable goods’ prices in cyclical real exchange rate fluctuations,” European Economic Review, 2013
[16]
Pasquale Rossi, DomenicoVecchio, GianlucaNeglia, Rossell Di Palo, Bianca Gasparrini,  Michael J. D’Occhio, and Giuseppe Campanile, “Seasonal fluctuations in the response of Italian Mediterranean buffaloes to synchronization of ovulation and timed artificial insemination,”Theriogenology, 2014
[17]
K. Rehfeld, and J. Kurths, “Similarity estimators for irregular and age-uncertain time series,” Climate of the Past, 2014
[18]
J. Durbin, “Time Series Analysis by State Space Methods,”Library of Congress Cataloging, 2012
[19]
Rolf Larsson, “A likelihood ratio type test for invertibility in moving average processes,” Computational Statistics & Data Analysis, Vol. 76, pp. 489–501, August 2014
[20]
SevilSenturk, NihalErginel, Ihsan Kaya, CengizKahraman, “Fuzzy exponentially weighted moving average control chart for univariate data with a real case application,” Applied Soft Computing, 2014
[21]
Yan-Fang Sang, Zhonggen Wang, and Changming Liu “Comparison of the MK test and EMD method for trend identification in hydrological time series,” Journal of Hydrology, Vol. 510, pp. 293–298, 14 March 2014
[22]
Mohammad Sayemuzzaman and Manoj K. Jha, “Seasonal and annual precipitation time series trend analysis in North Carolina, United States,” Atmospheric Research, Vol. 137, pp. 183–194, Feb. 2014
[23]
ChantaleDumoulin, “Pelvic floor muscle training versus no treatment, or inactive control treatments, for urinary incontinence in women,” The Cochrane Library, 2010