12導程心電圖是診斷心臟疾病的臨床標準之一，特別是對於致命性高的心肌塞梗。然而，標準12導程心電圖因體積龐大、導線眾多，始終無法普及於醫療機構外。可攜式單一導程心電圖目前已有多個微型化的商品，可讓病人隨身量測，但測得的訊號若有異常，多數還是得透過12導程心電圖確診，因此本研究預期應用已上市之單一導程心電圖記錄器，記錄於12導程之位置上不同時的心電圖，透過以小波轉換為基礎之演算法，重構出12導程心電圖，並進一步對於重構訊號以傅立葉轉換結合特徵值提取波形之特徵，利用多工接合稀疏表示技術進行分類。此研究與標準12導程心電圖相較而言，低成本及方便使用乃是它的優勢。若具有相當的準確度，可期許本研究之方法可應用於長照之醫療機構或是居家照護，輔助臨床人員提早診斷疾病。;The 12-lead electrocardiogram (ECG), which provides special projection of electrical activity of the heart in different orientation, is one of the standard procedures for clinical diagnosis of heart disease, especially the deadly myocardial infarction. However, the standard 12-lead ECG device is bulky and with lots of wires and therefore cannot be widely used outside the hospital and clinics. So far, there have been several miniaturized products for single-lead ECG recorders, allowing patients to measure ECG signals anytime. Once the measured signals seemed abnormal, a standard 12-lead ECG is still necessary to make a definite diagnosis. In this study, we aim to reconstruct the 12-lead ECG with a commercialized single lead ECG recorder. Using this device, we recorded ECG waveforms asynchronously on the positions of the traditional 12-lead ECG electrodes. Then, a wavelet-based algorithm was used to reconstruct the 12-lead signals. Furthermore, the reconstructed signal is classified using multitask joint sparse representation by the characteristics of Fourier transform combined with eigenvalue. The proposed method offers two advantages: low-cost and high-usability. Not only that, the result of experiment is considerably accurate comparing to standard 12-lead ECG device. On the other hand, this approach can be applied to both medical institutions and home care in the future, which can assist clinicians diagnose disease earlier.
