從單一訊號源萃取心電圖、胃電圖與呼吸訊號

Extraction of ECG, EGG and respiratory signal from single composite abdominal signal

邱少禹(Shao-Yu Chiu)

指導教授 ： 翁昭旼

共同指導教授 ： 蔣以仁(I-Jen Chiang)

國立臺灣大學/工學院/醫學工程學研究所/碩士(2009年)

https://doi.org/10.6342/NTU.2009.01521

摘要

過去生理監控系統缺乏一台整合性的儀器，使每位住院病人身上須貼附多張電極貼片才能擷取各種生理訊號，而過多的貼片可能會造成病人的不適外，各儀器間也可能會互相影響。另外，某些待在家中的病人突然發生身體不適時，最好也可以透過簡單的操作方式就可以將多種生理訊號做記錄，並經由網路將資訊傳給相關醫護人員。
因此，期望能設計一套儀器能藉由少量的貼片及電極線即能擷取受測者的生理訊號。在我們的系統裡將三張電極片貼於受測者腹部，透過合適的電路將訊號擷取至電腦上，並依各種生理訊號的特性不同，從一個訊號源分別將心電圖、胃電圖及呼吸訊號萃取出來。
本論文實作了一個結合心電圖、胃電圖與呼吸的生理訊號擷取系統，包括硬體上的製作、與訊號的儲存。在心電圖方面利用動態視窗的基頻訊號漂移修正法解決了呼吸造成的漂移問題。利用臨床使用之儀器同步記錄來驗證實作系統之可靠性在17筆受測者之心電圖訊號驗證中不論是長時間(1小時)或短時間(5分鐘)分析皆可得到與商用心電圖儀器很好的一致性。在胃電圖方面利用特殊的貼片設計，可與臨床用機器同步取得相同訊號源，在10筆受測者之長時間(1小時)或短時間(20分鐘)的胃電圖驗證裡，也皆得到很好的一致性。而在10位受測者的呼吸訊號驗證上，利用二次取樣與二次濾波器方式可以取得與臨床儀器一致的呼吸訊號。
總結以上，從腹部放置的電極貼片裡取得一組訊號，經過我們提供的硬體設計架構，能將擁有高頻特性的心電圖訊號或中頻的呼吸訊號或低頻且振幅小的胃電圖記錄下來。並依照它們在振幅、與頻率上不同的特性，使用我們的分析方法將它們各別分離出來。最後經由臨床應用儀器的驗證結果得知這樣的方法是可行的。

關鍵字

心電圖 ； 胃電圖 ； 呼吸訊號 ； 基頻訊號漂移修正 ； 數位濾波器

並列摘要

The lack of integrated bio-signal detection instruments made monitor patients’ multiple physiology parameters rather complicated in the past. Many electrodes need be applied to the body surface at the same time. Those recording devices may have interference with by each other. In addition, patients at home may have sudden attack of discomfort, an easy implemented device that can record a variety of essential physiological signals through simple operation will be extremely helpful. These signals can also be transferred through the network to health care specialists.
For above purposes, we implemented a portable device using few electrodes on abdominal wall to measure various patients’ electrophysiology signals simultaneously. The signals were acquisited through three electrodes placed on abdomen wall and were separated into Electrocardiogram (ECG), Electrogastrogram (EGG) and respiratory rhythm according to their individual rhythmic characters.
In this thesis, it set up a combinatory ECG, EGG and respiratory signal system which includes the hardware for data acquisition and storage. In ECG signal processing, dynamic window with the baseline wandering fitting algorithm was noted to solve the drifting problem caused by respiration. The validation of our combinatory monitoring system was verified by synchronous recording using commercial available individual system. Good ECG correlation was demonstrated in 17 subjects in a long duration (1 hour) or short time (5 minutes) analysis. In EGG signal processing, a special designed electrode was used to ensure simultaneously recording. In a 10 subjects study, a long duration (1 hour) or short time(20 minutes) analysis are both show good correlation. The respiratory signal component was verified by twice down-sampling processing and the usage of twice filtering. A good respiratory signal correlation was demonstrated in 10 subjects.
In brief. We had set up a system which can accurately record three sets of physiological signals with three electrodes on upper abdomen. High frequency high amplitude ECG signals and low frequency low amplitude ECG signals in accompany with respiratory movement signal can be simultaneously recorded. The mixed tracing can then be separated according to their characteristics. This simple design is very user friendly and can be applied to ambulatory physiological monitoring especially for the purpose of symptom correlation.

並列關鍵字

ECG ； EGG ； Respiratory Signal ； Baseline wandering fitting ； Digital Filtering

參考文獻

[1] Callaerts, D., Vanderschoot, J. Vandewalle, J., Sansen, W.,Vantrappen, G., Janssens, J., “An adaptive on-line method for the extraction of the’complete fetal electrocardiogram from cutaneous multilead recordings”, Journal of Perinatal Medicine, vol. 14, no. 6, pp. 421-433, 1986.

[2] L. Zhukov, D. Weinstein, and C. Johnson, “Independent component analysis for EEG source localization”, IEEE Eng. Med. Biol. Mag.,vol. 19, pp. 87–96, 2000.

[3] Van Oosrerom A , “Spatial filtering of the FECG”, J. Perinar. Med. 14 41 1-9, 1986.

[4] P.P. Kanjilal , S. Palit and G Saha, “Fetal ECG extraction from single-channel maternal ECG using singular value decomposition”, IEEE Transactions on Biomedical engineering , vol. 44, no. 1, January 1997.

[5] Denoising Sameni, R., Shamsollahi, M.B., Jutten, C., Clifford, G.D., “A Nonlinear Bayesian Filtering Framework for ECG”, IEEE Transactions on Biomedical Engineering, vol 54, Issue 12, Page(s):2172 – 2185, Dec. 2007.