体表胃电的特征提取方法研究

目的提取表征性强、稳定性高且能够为临床应用提供诊断参考的体表胃电图(EGG)特征参数。方法采用经验模态分解方法对EGG信号进行预处理,提取EGG信号的时域、频域以及非线性多维特征参数,并通过统计学方法筛选出最优特征参数组成特征向量,对功能性消化不良(FD)患者的EGG信号进行识别。结果基于时-频-非多维特征向量的FD患者的EGG信号识别率显著优于基于传统标准所构建的特征向量,其识别率最高可达90%以上。结论所提出的多维特征提取方法能够有效识别FD等胃肠疾病患者的EGG特征,可以为胃电的相关研究提供一种可靠的分析工具。

Study on Feature Extraction Method of Body Surface Electrogastrogram

Objective To extract feature parameters of body surface electrogastrogram(EGG)with strong characterization and high stability so as to provide diagnostic reference for clinical applications.Methods EGG signal was preprocessed by EMD method and the time-domain,frequency-domain and non-linear multi-dimensional feature parameters were extracted.The optimal features were selected by statistical method to form the feature vector,and the EGG signal of functional dyspepsia(FD)patients was identified.Results The recognition rate of EGG of FD patients based on time-frequency-non-multidimensional eigenvector was significantly better than that based on traditional examination and evaluation criteria of EGG,and the maximum recognition rate could reach more than 90%.Conclusion The new multi-dimensional feature extraction method proposed in this paper could effectively identify the EGG signal of FD patients and provided a new analysis tool for the research of EGG.