以數學模型應用於舉重選手之膝關節振動信號分析

Appling mathematical modeling for knee joint vibration signal analysis of weightlifters

李峰誌(Feng-Chih Lee)

指導教授 ： 李枝宏

共同指導教授 ： 江清泉(Ching-Chuan Chiang)

國立臺灣大學/電機資訊學院/電信工程學研究所/碩士(2004年)

https://doi.org/10.6342/NTU.2004.00068

摘要

運動員之膝關節既不屬於正常膝關節，亦不屬於病變膝關節，此因運動員在經歷一連串運動訓練後，身體結構產生適應性變化。本研究旨在探討不同運動表現之舉重選手膝關節於比賽前、後之差異性，及探討正常者、籃球及舉重選手之間的差異性。
本論文利用關節振動測量術（vibration arthrometry；VAM）進行研究，並配合自回歸模型（autoregressive model）及可適性區段（adaptive segmentation）等信號處理技術。VAM為非侵襲性、簡單而價廉的臨床檢查工具，且不受時空的限制，亦不影響運動員之作息及訓練，是適合研究運動員膝關節的診斷工具。
研究對象有正常者27名、籃球及舉重選手各20名，以相依樣本單因子變異數分析（repeated measure one-way ANOVA）考驗舉重選手於比賽前、後是否有顯著差異，若達顯著則進行杜凱氏（Tukey）事後考驗；以混合設計二因子變異數分析（two-way ANOVA mixed design）考驗肌力比高低與比賽前後是否有交互作用，若交互作用達顯著，則進行杜凱氏（Tukey）事後考驗；以獨立樣本單因子變異數分析（one-way ANOVA）考驗正常者、籃球及舉重選手之間的差異，若達顯著則進行雪費（Scheffe）事後考驗。
從信號處理、膝關節生理特性及運動訓練法三方面綜合分析解釋研究結果。實驗結果顯示均方根值可有效區分正常者與運動（籃球、舉重）選手，此解釋為運動選手長期接受運動訓練以致軟骨磨損；主極點必v比可區分高肌力比舉重選手在比賽前、後的差異，而研究結果顯示選手腿部肌力的改變影響主極點必v比的變化。

關鍵字

可適性區段 ； 獨立樣本單因子變異術分析 ； 混合設計二因子變異術分析 ； 相依樣本單因子變異術分析 ； 膝關節 ； 關節振動量測術 ； 自回歸模型 ； 舉重選手 ； 肌力比

並列摘要

The athletes’ knee joint is neither normal knee joint nor abnormal knee joint because the exercise training program causes structural adjustment of the athletes’ body. This thesis is divided into two topics. The first topic is aimed to discuss difference between two groups of weightlifters before and after competition, which have different muscle strength ratio. The second topic considers the difference of knee joint among normal people, basketball players and weightlifters.
We apply vibration arthrometry(VAM) to collect the vibration signals of the patellar and use adaptive segmentation and autoregressive model to analyze the vibration signals. The VAM is a noninvasive, simple and cheap diagnostic tool. It is a suitable diagnostic tool for athletes because it is not restricted by the time and space and doesn’t influence the training schedule of athletes.
The subjects of this thesis were 27 normal people, 20 basketball players and 20 weightlifters. Repeated measure one-way ANOVA was used to compare mean difference of weightlifters at before and after the competitions. Turkey post hoc t-comparison was used to analyze the data if there was a significant difference. Two-way ANOVA mixed design was used to analyze the interaction between muscle strength ratio and the time before and after competition. Turkey post hoc t-comparison was used to analyze the data if there was a significant difference. One-way ANOVA was used to compare mean difference among normal people, basketball players and weightlifters.
We use signal processing, physiological characteristics of knee joint and sport science and training to explain experiment results. The experiment results show that RMS value could distinguish normal people and athletes (basketball players and weightlifters) because of the cartilage erosion of athletes’. The spectral power ratio of dominant pole could distinguish weightlifters with high muscle strength ratio at before and after competition. In this thesis, we observe that the change of muscular strength of leg of weightlifters has an influence on spectral power ratio of dominant pole.

並列關鍵字

repeated measure one-way ANOVA ； adaptive segmentation ； one-way ANOVA ； two-way ANOVA mixed design ； autoregressive model(AR model) ； weightlifters ； muscle strength ratio ； knee joint ； vibration arthrometry(VAM)

參考文獻

[5] S. Tavathia, R. M. Rangayyan, C. B. Frank, G. D. Bell, K. O. Ladly, and Y. T. Zhang,“Analysis of knee vibration signals using linear prediction,”IEEE Transactions on Biomedical Engineering, vol. 39, no. 9, pp. 959-970, Sept. 1992.

[6] C. B. Frank, R. M. Rangayyan, and G. D. Bell,“Analysis of knee joint sound signals for noninvasive diagnosis of cartilage pathology,”IEEE Engineering on Medicine and Biology Magazine, vol. 9, no. 1, pp. 65-68, March 1990.

[7] Z. M. K. Moussavi, R. M. Rangayyan, G. D. Bell, C. B. Frank, K. O. Ladly, and Y. T. Zhang,“Screening of vibroarthrographic signals via adaptive segmentation and linear prediction modeling,”IEEE Transactions on Biomedical Engineering, vol. 43, no. 1, pp. 15-23, Jan. 1996.

[8] S. Krishnan, R. M. Rangayyan, G. D. Bell, C. B. Frank, and K. O. Ladly,“Recursive least-squares lattice-based adaptive segmentation and autoregressive modeling of knee joint vibroarthrographic signals,”in Proc. Canadian conf. Electrical and Computer Engineering, Calgary, Alta., Canada, vol. 1, pp. 339-342, May 1996.

[9] Y. T. Zhang, C. B. Frank, R. M. Rangayyan, and G. D. Bell,“Mathematical modeling and spectrum analysis of the physiological patello-femoral pulse train produced by slow knee movement,”IEEE Transactions on Biomedical Engineering, vol. 39, no. 9, pp. 971-979, Sept. 1992.