基於高齡化的家庭醫療保健技術

Home Healthcare Technologies for Active Ageing

趙彥揚(Yen-Yang Chao)

指導教授 ： 馬席彬

國立清華大學/電機資訊學院/電機工程學系/碩士(2013年)

摘要

近年來醫療保健系統是一個新興的話題。這個系統可以為患者節省時間，改善患者的生活，並且減少濫用的醫療成本。這項研究目標是開發一個攜帶型心電圖即時監測系統並且結合具有能區分日常生活動作和跌倒的偵測裝置。在本篇論文分為兩個部分，一個是心電圖即時監測，另一個是應用模糊理論的跌倒偵測系統。該系統包含一個無線感測節點和一個中繼站，無線感測節點用來擷取身體的生物信號，中繼站的功能則是把截取後的生理訊號即時顯示在螢幕上以及利用3G或WiFi網路傳送到遠端的伺服器。無線感測節點包含一個類比前端放大器、微處理器、類比數位轉換器以及低耗電藍芽4.0版本。我們也利用Android平台作為一個資料中繼站，將數據傳輸到遠端的伺服器。在跌倒偵測的部分，我們使用三軸加速器來建立跌倒時的特徵。為了提高容錯和準確率，我們提出一個行為估計方法，其中包含加速度的變化和模糊理論來估算實驗對象的行為。最後我們使用MATLAB內建的模糊理論工具箱來進行模擬及驗證。結果顯示我們可以從160次的日常生活動作姿態中區分跌倒，敏感性為95%和特異性為97.5%。

關鍵字

心電圖 ； 跌倒偵測 ； 模糊理論

並列摘要

Healthcare system is an emerging topic in recent years. It can save time for patients, improve patients life, and reduce the abuse cost for medical treatment. The goal of this research is to develop a real-time portable Electrocardiography (ECG) monitoring device, and fall detection that is capable of discriminating between Activities of Daily Life (ADL) and falls. There are two main parts which discussed in this thesis. One is ECG real-time monitoring. Another fall detection is using fuzzy logic.
The system contains wireless sensor node that capturing the bio-signal of the body and a mobile hub that wireless sensor node can send information to the mobile hub. The wireless sensor node consists an analog front-end amplifier, an MCU that control the inside analog digital converter (ADC), and a bluetooth module with 4.0 Bluetooth low energy (BLE) version. We even use Android platform as a hub to transfer data to remote server. In fall detection part, we use 3-axis accelerometer to develop an effective fall detection algorithm based on the characteristics of falls. To increase the error tolerance and increase accuracy rate , we propose a behavior estimation method which consists of the change of acceleration (COA) and fuzzy rule based system to estimate the subject’s behavior. Then, we use MATLAB Fuzzy Logic Toolbox to simulate and estimate the behavior. Results show that falls can be distinguished from ADL with a sensitivity over 95% and a specificity of 97.5%, for a total set of 160 movements.

並列關鍵字

ECG ； Fall detection ； Fuzzy Theorem ； MSP430 ； Arduino ； BLE

參考文獻

[7] R. Fensli, E. Gunnarson, and O. Hejlesen, “A wireless ECG system for continuous event recording and communication to a clinical alarm station,” in IEEE International Confer- ence of the Engineering in Medicine and Biology Society (IEMBS), vol. 1, San Francisco, CA, Sept. 2004, pp. 2208–2211.

[8] Y. Y. ChenWu, H. P. Ma, C. Biswas, and D. Markovic, “Universal architecture prototype for patient-centric medical environment,” in IEEE International Symposium on VLSI Design, Automation, and Test (VLSI-DAT), Hsinchu, Taiwan, April 2012, pp. 1–4.

[12] K.Ho, K.Yamamoto, N.Tsuchiya, H.Nakajima, K.Kuramoto, S.Kobashi, and Y.Hata, “Multi sensor approach to detection of heartbeat and respiratory rate aided by fuzzy logic,” in IEEE International Conference on Fuzzy Systems (FUZZ), Barcelona, Spain, July 2010, pp. 1–6.

[15] F. Sposaro and G. Tyson, “iFall: An android application for fall monitoring and re- sponse,” in IEEE International Conference Engineering in Medicine and Biology Society (EMBC), Minnesota, USA, Sept. 2009, pp. 6119–6122.

[17] A. Bourke, K. J. O’Donovan, J. Nelson, and G. M. OLaighin, “Fall-detection through vertical velocity thresholding using a tri-axial accelerometer characterized using an optical motion-capture system,” in IEEE International Conference on Engineering in Medicine and Biology Society (EMBS), Vancouver, Canada, Aug. 2008, pp. 2832–2835.