| Summary: | 碩士 === 國立臺北科技大學 === 電子工程系研究所 === 103 === The electrocardiogram (ECG) is one of the most important physiological signals for heart disease diagnosis. However, vast amount of data can be generated by electrocardiogram (ECG) monitoring device. Therefore, the storage capacity is an important issue in the design of an ECG device. For this, many ECG compression algorithms had been proposed in order that the data can be stored efficiently. When reviewing all of the ECG algorithms proposed, we find most of them are still with higher computational complexity due to nonlinear operation used. Therefore, we try to develop an efficient lossless ECG compression method that can be used for real-time applications.
We notice that the ECG waveform is quite similar among adjacent heartbeats. Thus, individual heartbeat is separated by using an R-wave detection method. After that the well-known Takagi–Sugeno Fuzzy Neural Network will be applied as the predictor so that the ECG signal can be predictively encoded. Finally, the proposed lossless compression algorithm will be evaluated by using MIT-BIH arrhythmia database. Experimental results show that an average compression ratio about 3.25 can be achieved by using the proposed approach which justifies the usefulness of the proposed predictive coding scheme.
|
|---|
