膝關節振動訊號無線測量系統之開發與膝關節退化性程度之分析

Development of a System for Measuring Knee Joint Vibration Signals and Analysis of Osteoarthritis Degree of Knee Joints

湯雅茗(Ya-Ming Tang)

指導教授 ： 王家鍾

義守大學/醫學院/生物醫學工程學系/碩士(2014年)

https://doi.org/10.6343/ISU.2014.00267

摘要

在骨科中最常見的疾病為退化性關節炎，所以早期發現早期治療顯得格外重要。本論文中改良實驗室已開發之無線式多通道膝關節量測裝置，此裝置可同時量測三通道振動訊號(外側髁突、髕骨中央及內側髁突)及單通道角度訊號，為驗證此裝置之可行性，在義大醫院骨科部進行人體試驗，已完成128例的人體試驗(控制組N=29，退化性關節炎組N=99)之五種運動試驗項目(主動擺動、被動擺動、上下台楷、正常行走及蹲立動作)，將量測到的訊號以頻域及時域的分析方法進行分析。
頻域分析結果顯示，針對量測內側髁突的位置，在主動擺動(無負重)時，Group 1(Control 組)之頻段0-200 Hz的振幅能量比例顯著小於Group 4(OA-Ⅲ&Ⅳ組)(38.49±13.02% vs 48.99±13.90%, p=0.008)，而Group 1之頻段400-1K Hz的振幅能量比例顯著大於Group 4 (25.42±6.78% vs 20.57±6.87%, p=0.031)；在正常行走(具負重)時，Group 1之頻段0-200Hz的振幅能量比例顯著小於Group 4(46.63±9.77% vs 55.98±9.30%, p=0.004)，而Group 1之頻段400-1K Hz的振幅能量比例顯著大於Group 4 (21.79±4.75% vs 17.04±4.09%, p=0.004)。時域分析結果中顯示，針對量測內側髁突的位置，在正常行走(具負重)之膝關節彎曲角度區間於90~0度時，Group 1之參數F0(平均頻率)平均值顯著大於Group 4 (552.44±58.23Hz vs 459.24±76.61Hz , p=0.000)；在角度區間於0~90度時，Group 1之參數F0平均值顯著大於Group 4 (559.35±79.20Hz vs 486.48±99.38Hz , p=0.009)。由頻域分析整體結果得知，頻段0-200之振幅能量比例特徵隨OA嚴重程度增加有上升的趨勢，而頻段400-1K及1K-2.5K Hz之振幅能量比例特徵則隨著OA嚴重程度增加有下降的趨勢，另由時域分析整體結果得知，參數F0之平均數值隨OA嚴重程度增加有下降的趨勢，可結合上述三項特徵條件作為膝關節是否病變之判斷條件，協助醫師快速判斷門診病患膝關節病變的退化程度。

關鍵字

退化性關節炎 ； 振動訊號 ； 膝關節 ； 身體負重 ； 頻域分析

並列摘要

Analysis of the orthopedic clinic diseases shows that there are two-thirds of orthopedic patients with the knee degeneration or lesions. Therefore, how to diagnose the degeneration of knee joint in time is very important and urgent. The purpose of the thesis was to develop a wireless measuring system for non-invasively recording the vibration signals from the surface of knee joints with osteoarthritis (OA), and to validate the feasibility of the measuring system. In human trials, twenty-nine normal subjects and ninety-nine OA patients in the Department of Orthopedics, E-Da Hospital, were asked to perform the active and passive swings with no body-weight loading, and the step-forward and –backward motions, walking movement, and squatting movement with body-weight loading. During these experiments, the vibration and angle signals from the surface of knee joints were recorded with the measuring system. The vibration signals generated by the normal and osteoarthritis joints were first segmented according to different maneuvers, and then the segmented signals were further analyzed by using the frequency- and time-domain techniques.
In the frequency-domain analysis, the result showed that the energy ratio of 0-200 Hz in the vibration signals recorded from the medial condyle during the active swing (without loading) in Group 1 (control group) was significantly smaller than that in Group 4 (OA-Ⅲ&Ⅳpatients) (38.49±13.02% vs 48.99±13.90%, p=0.008). The energy ratio of 400-1KHz in Group 1 was significantly greater than that in Group 4 (25.42±6.78% vs 20.57±6.87%, p=0.031). The vibration signals recorded from the medial condyle during the walking movement (with loading), the energy ratio of 0-200 Hz in Group 1 was significantly smaller than that in Group 4 (46.63±9.77% vs 55.98±9.30%, p=0.004). The energy ratio of 400-1KHz in Group 1 was significantly greater than that in Group 4 (21.79±4.75% vs 17.04±4.09%, p=0.004). The time-domain analysis showed that the vibration signals recorded from the medial condyle during the walking movement (with loading), during the 90~0∘angle range, the mean value of F0 in Group 1 was significantly greater than that in Group 4 (552.44±58.23Hz vs 459.24±76.61Hz, p=0.000). During the 0~90∘angle range, the mean value of F0 in Group 1 was significantly greater than that in Group 4 (559.35±79.20Hz vs 486.48±99.38Hz , p=0.009).
In summary, the energy ratio of 0-200 Hz in severe OA patients was higher than mild OA patients. In addition, the energy ratio of 400-1K and 1K-2.5K Hz in severe OA patients was lower than that in mild OA patients. In addition, the mean value of F0 in severe OA patients was lower than that in mild OA patients. Therefore, we may design a software program, based on on these findings, to help the orthopedist effectively diagnose the severity of OA.

並列關鍵字

Osteoarthritis ； Vibration signal ； Knee joint ； Body-weight loading ； Frequency-domain analysis

參考文獻

[3] 江清泉，「膝關節病變之膝關節振動訊號的數學模式，NSC89-2320- B002-109」，89年度(第一期)國科會專題研究計畫成果報告。

[4] D. Pereira, B. Peleteiro, J. Araújo, J. Branco, R. A. Santos, and E.Ramos, "The effect of osteoarthritis definition on prevalence and incidence estimates: a systematic review. Osteoarthritis Cartilage," Osteoarthritis and Cartilage, 19(11):1270-85, 2011.

[6] W. E. Blodgett, "Auscultation of the knee joint, "Boston Med. Surg J., 146(3):63-66, 1902.

[7] M. L. Chu, I. A. Gradisar, and R. Mostardi, "A noninvasive electroacousticale valution technique of cartilage damage in pathological kneejoints, "Med. Biol. Eng. Comput., 16(4):437-42, 1978.

[9] S. Derrett, G. D. Walley, S. A. Bridgman, P. Richards, and N. Maffulli, "Magnetic resonance imaging, knee arthroscopy, and clinical decision making: a descriptive study of five surgeons," Int J Technol Assess Health Care, 25(4):577-83, 2009.