心臟電生理訊號分析及應用

Physiological Signal Analysis and Applications in Cardiology

周以琪(I-Chi Chou)

指導教授 ： 李仁貴

國立臺北科技大學/機電學院/電腦與通訊研究所/博士(2009年)

摘要

本論文之主要目標在提供適用於心電紀錄儀之資料壓縮技術與安全防護機制。鑑於健康照護服務之傳輸應用已成重要趨勢，越來越多的資料需傳輸至遠端照護中心，以提供更多元之應用服務。其中，心電紀錄儀因可長時間紀錄心電訊號而被廣泛使用，卻也造成大量資料傳輸與處理的需求。然而傳輸之頻寬卻是有限的。此外，新的心電紀錄儀為了可攜式需求，多採用電池供電並搭配有限運算能力之中央處理單元。因此，在有限的運算能力與傳輸頻寬下，心電紀錄儀需能提供有效且簡易之資料減量技術。
為此，本文心電紀錄儀端提供病症分類技術以辨識異常心電訊，再藉由無線傳輸技術傳送至家用閘道器。由於傳輸頻寬之限制，家用閘道器利用資料壓縮技術，將異常訊號進行壓縮處理，爾後則利用無線或有線方式送至遠端。藉此克服大量資料處理與傳輸問題。本文所採用之壓縮技術不僅能藉由能量關係來控制重建誤差，且無須透過小波反轉換進行驗證，因而減少運算複雜度。
從安全觀點來看，因傳輸需求而衍生之資訊安全問題，亦需納入考量。本文採用馴轉換數位簽章技術(TTS)，進行使用者訊號加密處理。屬於多變量公鑰系統之TTS，因具備有快速金鑰生成及低運算量之優點，故適用於心電紀錄儀。另一個安全疑慮則是來自心電紀錄儀。心電紀錄儀在使用中是否有潛在風險，內部電子元件是否有故障疑慮。為確保安全使用，風險管控實為必要。本文採用故障模式與效應分析法(FMEA)來評估心電紀錄儀之潛在風險。藉由系統化分析，釐清風險與效應並提出防護系統以提升使用安全。
如上所述，本文提出以病症辨識為基礎之壓縮技術及心電紀錄儀之安全防護機制，並採用MIT-BIH心率不整資料庫進行正確率及效能測試。根據實驗結果，病症辨識部分之平均正確率為98.38%。資料壓縮則因具有可控制之重建誤差特性，而可彈性選擇壓縮率。至於安全機制之資料加密技術TTS，以高速及低運算量特性為主，僅需0.0187ms即可完成簽章功能。在私鑰與公鑰生成時間則分別需要0.0515ms及0.0547ms。最後，為確保心電紀錄儀之安全，本文在完成潛在風險分析後，提出適當防護策略，經由驗證確實能降低心電紀錄儀之風險等級。

關鍵字

心電訊號 ； 壓縮 ； 小波轉換 ； 數位簽章 ； 故障模式與效應分析

並列摘要

The primary goal of this dissertation is to propose an effective ECG signal compression algorithm and the safety strategy in a novel Holter. As the growth of health-care applications the most remarkable trend in the transmission field, more and more data need to be transmit to the remote care center for various services. Generally, the Holter monitor is usually utilized to record user’s ECG signal with long-tern monitoring which caused a huge amount of data to be processed and transmitted; however, the transmission bandwidth is limited. In addition, the novel Holter is designed of battery powered and CPU with limited computation ability for portable. Hence, the requirement of Holter is to apply an effective and simple algorithm for reducing the data amount under the limitation of computation ability and communication bandwidth.
In this dissertation, the portable Holter is utilized to identify the ECG signal by the proposed arrhythmia recognition algorithm and send the abnormal beat to the home gateway via wireless technology. Due to the bandwidth limitation of communication technology, the home gateway can compress the received data from Holter and transmit to the remote care center through wired/wireless transmission medium. The adopted compression algorithm not only helps to control the reconstruct error according to the relationship of energy but also reduces computation efforts without inverse wavelet transform.
From a different viewpoint, the information security problem is a major issue for data transmission, thus, this study also researched the topic in data delivery of Holter. In order to transmit user’s information through communication channel, the Tame Transformation Signature (TTS) technology is applied to encrypt private data. TTS is one of the multivariate public key cryptosystems which has the prominent advantages of fast key generation and low computation to fit the conditions of Holter.
The other safety issue is in relation to Holter device while user using it. Holter is consisted with numerous electronic components and operated by user; nevertheless, there are some potential hazards need to be controlled. For this reason, the Failure Mode and Effects Analysis (FMEA) method is utilized to assess the risk of Holter. By systematic analysis, this study can identify the potential risks and provide applicable protection system for improving safety in Holter.
As above mentioned, the compression algorithm with symptom recognition and the safety strategies are used in the Holter and tested the performance and accuracy according to the MIT-BIH database. In the experiment result, the average accuracy of arrhythmia recognition is 98.38% for MIT-BIH database, and the compression rate is flexible with a controllable percent root-mean-square difference (PRD). TTS technology provides high speed and low calculations, in which the average time to write a signature is 0.0187ms. In addition, the average time to generate private key and public key is 0.0515ms and 0.0547ms respectively. Finally, the FMEA analysis is utilized to discover the potential risks from component or user’s operation. According to the analytic results, two protection system are provides to decrease the risks of Holter.

並列關鍵字

Physiological ； Compression ； Wavelet ； Digital Signature ； FMEA

參考文獻

[1] J. Bai, Y. Zhang, D. Shen, L. Wen, C. Ding, Z. Cui, F. Tian, B. Yu, B. Dai, and J. Zhang, “A portable ECG and blood pressure telemonitoring system,” IEEE Engineering in Medicine and Biology Magazine, vol. 18, 1999, pp. 63–70.

[2] J. J. Segura-Juarez, D. Cuesta-Frau, and L. Samblas-Pena, “A microcontroller-based portable electrocardiograph system,” Proceedings of the 10th IEEE International Conference on Electronics Circuits and Systems, vol. 2, 2003, pp. 922 – 925.

[3] K. Hung, Y.T. Zhang, and B. Tai, “Wearable medical devices for telehome healthcare,” Proceedings of 226th Annual International Conference of the engineering in Medicine and Biology Society, 2004, pp. 5384-5387.

[4] J. Kim, J. M. Kim, S. H. Noh, S. L. Min, and Y. Cho, “A space-efficient flash translation layer for compact flash systems,” IEEE Transactions on Consumer Electronics, vol. 48, issue 2, 2002, pp. 366–375.

[5] T. Deniz, and A. Yilmaz, “Design and implementation of a digital ambulatory ECG recorder based on flash MultiMediaCard memory,” Proceedings of the IEEE International Symposium on Micro-NanoMechatronics and Human Science, Vol. 1, 2003, pp. 368-371.