中文摘要：

目的：ST段是心电图的重要组成部分，其起始于心电图QRS波群的结束点并结束于T波的开始点。ST段的压低、分上斜形和下垂形压低、水平压低和以及鱼钩样改变都可以反映各种心脏疾病，所以，精确提取心电图ST段具有重要意义。因此。本文提出了基于小波包变换的心电图ST段精确提取算法。方法：首先对心电图信号进行去噪处理，滤去工频干扰信号以及基线漂移信号；然后，提取单周期的心电图信号；最后，引入小波包变换算法提取QRS波群、T波的主频带，重构QRS波群、T波的波形并确定ST段的始末位置。结果：本文算法在时域心电图上实现了ST的精确定位，提取了心电图的ST段。通过在经典数据库中的验证，本文算法具有非常好的表现。结论：实验结果表明，相较传统的对心电图加时间窗的方式提取ST段，本文算法可以精确提取心电图ST段，这为心电图ST段的自动精确识别，以及用于自动医疗检测与便携式医疗设备提供了依据。

英文摘要：

Objective： ST-segment is an important part in ECGs, it starts at the end point of QRS wave group and ends at the beginning of T wave. ST-segment depression, upward diagonal and drop shaped depression, level down and hook like changes can reflect various heart diseases. Therefore, the accurate extraction of ECG ST-segment is very important. So, an extraction method of the ST-segment, which is based on the wavelet packet transform, is proposed in this paper. Methods： Firstly, the in- terference signal of ECGs is eliminated, the power-line interference and baseline drift signal are filtered. The single cycle of ECG signal is then extracted. Finally, the wavelet packet transform algorithm is introduced to extract the QRS wave group and the T wave. The QRS wave and T wave are reconstructed, and ST-segment is determined by the time location. Results： The lo- cation of ST-segment in the time domain is realized by the proposed algorithm, and the ST-segment is extracted from the ECG. The proposed algorithm has a good performance by validation from classical ECG database. Conclusions： The experimental re- sults show that, compared with the traditional ST extraction method which is based on time windows, ＇the proposed algorithm can precisely extract the ST-segment. This provides a solid basis for the ST-segment automatic identification, as well as the au- tomatic detection of medical treatment and the portable medical devices.