以加速規為基礎之人體姿態穩定量測系統

Abstract

本研究是以三軸加速規為基礎開發出低成本且高效能的「人體姿態穩定量測系統」，本系統包括個人電腦、微處理機與三個三軸加速規，三個加速規分別放置於受試者腰部，大腿和小腿的位置，微處理機能將類比加速度訊號轉換為數位訊號，接著將數位訊號依通訊協定編碼為數位資料並且透過RS232通訊介面傳回個人電腦，量測系統之取樣頻率達655Hz。 比對以下三種不同的實驗方法：(1)觀看靜態影像時的身體晃動；(2)觀看暈眩動態影像時的身體晃動；(3)進行數學運算時的身體晃動。 量測訊號的分析程序主要分為五個部份：(1)以一維小波去噪的方式濾除白雜訊；(2)使用零相位巴特沃斯低通濾波器濾除60Hz電力線干擾與50Hz以上的高頻干擾；(3)使用零相位巴特沃斯高通濾波器濾除2Hz以下的低頻干擾；(4)使用多變量多尺度熵分析濾波訊號(MMSE)；(5)將尺度1至尺度10計算出的MMSE加總得到多變量多尺度熵指標(MMEI)。 本實驗總共有30位20至26歲的健康受試者參與，經過MMEIs計算後的結果顯示出在比對觀看靜態影像與進行數學運算時的實驗結果有顯著差異，且進行數學運算時的MMEIs計算結果小於觀看靜態影像結果，本研究顯示出本系統可評估身體潛在的平衡與姿態穩定能力。 關鍵字：身體晃動、加速規、零相位、多變量多尺度熵指標In this study, we develop an inexpensive, efficient system for the measurement of postural stability using tri-axial accelerometers. This system consists of a personal computer, a microprocessor and 3 tri-axial accelerometers located on waist, thigh and ankle of a subject. The accelerometers were used to measure the accelerations caused by postural sway of the subjects. The microprocessor converts the analog acceleration signals into digital signals and then transmits the digital data to a personal computer via RS232 interface. The sampling rate of this measure system is up to 655Hz. Three different testing methods were compared: (1) measurement of postural sway during watching a still picture; (2) measurement of postural sway during watching a dizzy picture; (3) measurement of postural sway during doing mental calculations. The analysis procedures of the measured data consists of five major parts: (1) removing the sensor noise by using wavelet denoise method; (2) removing the power line inference (60Hz) by using a zero phase Butterworth low pass filter with the cutoff frequency at 50Hz; (3) removing the low frequency inference (60Hz) by using a zero phase Butterworth high pass filter with the cutoff frequency at 2Hz; (4) Analyzing the filtered signals by using the multivariate multiscale entropy (MMSE). (5) multivariate multiscale entropy index (MMEI) were obtained by summing the MMSEs from scale 1 to scale 10. There are thirty healthy subjects aged 20-26 years participate in this research project. The experimental results indicate that there exists a significant difference of MMEIs between watching-still-picture tests and doing-mental-calculations tests. The MMEIs obtained from doing-mental-calculations are smaller than that obtained from watching-still-picture tests. These studies show the potential of this system in the assessment of balance and postural stability. Keywords: postural sway, accelerometer, zero phase, multivariate multiscale entropy index.