以手機實現無線心電圖儀之設計

Abstract

因應高齡化社會的來臨，遠距醫療及健康照護已成為先進國家的新興服務產業。心電圖紀錄心臟搏動相關的電位變化資訊，常用於心血管疾病的診斷及治療。本論文開發一基於分散式訊源編碼架構的無線心電圖儀。分散式訊源編碼器的壓縮率及重建品質取決於訊源相關模型及校驗子生成機制，我們首先針對MIT-BIH心電圖資料褲，使用向量量化來建立與其匹配的訊源相關模型，並提出一基於索引層級的軟性輸出解碼演算法。系統模擬結果證實新的演算法大幅降低了資料量，並保留了心電圖於臨床上的重要資訊。在即時實作方面，我們結合心電圖感測器及執行訊號壓縮的嵌入式系統開發板，並藉由現有的行動手機網路將壓縮資料傳至遠端的伺服器以供訊號重建及監測。實作結果亦證實新的方法適用於無線健康照護系統的未來應用。

In recent years, mobile telemedicine has been one of the emerging research topics. This thesis proposes using a mobile telephone to realize an electrocardiogram (ECG) signal monitoring system. In this system, we use a sensor to measure ECG signals and use an embedded system development board as the front-end device to compress the ECG signals. The proposed ECG signal compression technique is based on distributed source coding (DSC). The compressed ECG data are transmitted from the board to the back-end server via mobile cellular networks. At the server, we use the modified BCJR algorithm which integrates the symbol-level a priori information into the soft-output decoding algorithm. Through off-line simulation and real-time implementation, we test and verify the functions of each module of the ECG system. The results show that our proposed system is very feasible for future wireless health-care applications.