基於感壓器陣列之動脈壓力波形量測系統設計

Arterial Pressure Waveform Monitoring System Design Based on Sensor Array

黃植鈺(Chih-Yu Huang)

指導教授 ： 葉丙成

國立臺灣大學/電機資訊學院/電信工程學研究所/碩士(2013年)

https://doi.org/10.6342/NTU.2013.00411

摘要

血壓測量是評估心血管疾病的重要方法之一， 傳統上測量的目標主要包括收縮壓與舒張壓，以及脈搏壓力波形。 其中收縮壓與舒張壓的數值是高血壓疾病最普遍的指標， 而從脈搏壓力波形中也可以分析出許多關於心血管的資訊。 此外，脈搏壓力傳播速度亦是個評估血管硬化的重要指標。 當血管變硬時，壓力波就會隨之變快。 然而，能夠測量這些資訊的儀器都相當龐大且昂貴。 在這篇論文中，我們致力於開發一個可攜性的動脈血壓波形監控系統， 並希望此系統能夠讓使用者自行在家中使用，如同血壓計一般。 系統的功能主要針對連續的量測脈搏壓力波形，以及計算脈搏壓力傳播速度。 本系統是由壓力感測器陣列、類比轉數位晶片、和信號處理程式所組成。 壓力波形是從感測器陣列所量到的多個信號回復而來，而壓力傳播速度則是以計算不同感測器所量到的信號時間差的方式所得。 對於以上這兩個目標我們各別實作了一些演算法， 並且將結果與現成的醫療儀器進行比較。 為了要確認回復信號的品質，我們從波形中計算一些用於波形分析的數值並和現有的醫病儀器所測量到的信號做比較， 而比較結果顯現了高度的一致性。 接著，我們將測量到的局部傳播速度與醫療儀器所量到的速度數值進行比較，結果發現我們所提出的演算法具有高度的相關性及精密性。 根據結果，我們的成果是一個有效的可攜式壓力波形監控系統的原型。 未來只要將感測器陣列和我們提出的信號處理演算法整合進目前的血壓計當中，相信這就能成為可攜式的壓力波形量測儀器。

關鍵字

脈博波傳導速度 ； 數位信號處理 ； 盲信號分離 ； 信號對齊

並列摘要

Blood pressure measurement is an essential evaluation of cardiovascular disease. Traditionally, the main objective of the measurement include the systolic and diastolic pressure and the pulse pressure waveform. The blood pressure is an general index to determine the hypertension, and many information can be retrieved from the pulse pressure waveform. Moreover, the pulse wave velocity (PWV) is an important index to assess the stiffness of vessels as well. While the vessel becomes stiff, the pressure wave will transmit faster. However, the current clinical device to measure these data is bulky and expensive. In this thesis, we aim to develop a portable arterial pressure waveform monitoring system, which can be used at home like a sphygmometer. The main feature of the system is to record the pressure waveform continuously and estimate the local PWV. This system is composed of a pressure sensor array, an analog-to-digital chip, and a signal processing program. With the help of the sensor array, the pressure waveform is recovered from multiple signals and the pulse wave velocity is calculated by the time difference of signals measured by different sensors. The results are evaluated by comparing with a current clinical device. First, to evaluate the quality of the recovered pressure waveform, we extract and compare several indices from the waveform. The result shows that the recovered waveform has high similarity comparing to the reference. Second, we compare the value of the PWV, and the results show high positive correlation with the reference, and it also has very great precision. According to the results, our work is a good prototype of a portable pressure waveform monitoring system, and it has similar performance with the current clinical de- vice. While integrating a sensor array and our signal processing algorithm into the current sphygmometer, it can evolve to a portable waveform monitoring device.

並列關鍵字

Pulse pressure velocity ； digital signal processing ； blind signal separation ； signal alignment

參考文獻

[6] Peter J Brands, Jean M Willigers, Le ́on AF Ledoux, Robert S Reneman, and Arnold PG Hoeks, “A noninvasive method to estimate pulse wave velocity in ar- teries locally by means of ultrasound,” Ultrasound in medicine & biology, vol. 24, no. 9, pp. 1325–1335, 1998.

[7] PierreJallonandAntoineChevreuil,“Separationofinstantaneousmixturesofcyclo- stationary sources,” Signal Processing, vol. 87, no. 11, pp. 2718–2732, 2007.

[8] Anthony J Bell and Terrence J Sejnowski, “An information-maximization approach to blind separation and blind deconvolution,” Neural computation, vol. 7, no. 6, pp. 1129–1159, 1995.

[9] Adel Belouchrani, Karim Abed-Meraim, J-F Cardoso, and Eric Moulines, “A blind source separation technique using second-order statistics,” Signal Processing, IEEE Transactions on, vol. 45, no. 2, pp. 434–444, 1997.

[10] SI McNeill, “An analytic formulation for blind modal identification,” Journal of Vibration and Control, vol. 18, no. 14, pp. 2111–2121, 2012.