采用基于密度加权和偏好信息的K均值聚类的胸阻抗信号自动检测算法

发布时间：2018-03-05 14:54

本文选题：自动识别　切入点：胸阻抗　出处：《电子与信息学报》2015年04期 　论文类型：期刊论文

【摘要】：为了自动识别胸阻抗(Trans Thoracic Impedance,TTI)信号中的按压和通气波形,完成相关重要参数的计算,从而实现对心肺复苏质量的监测评估,该文提出一种基于密度加权与偏好信息的胸阻抗信号自动检测算法。该方法针对实验采集的猪的电诱导心脏骤停模型TTI信号,通过预处理和多分辨率窗口搜索法完成潜在按压和通气波形的标记;接着,提取每个标记波形的宽度、幅值以及相邻波形特征差作为特征,并按标记波形宽度对信号进行分段;之后,再对信号进行小波分解,提取其小波系数每段的能量与原始波形幅值之比作为特征;最后采用基于密度加权与偏好信息的K均值聚类分析法对标记的波形进行分类识别。实验结果表明,该算法对TTI信号中按压波形和波形分析识别的正确率和敏感度均较高,鲁棒性好,且运行时间(0.43 s±0.07 s)满足实时性要求。
[Abstract]:In order to automatically identify the compression and ventilation waveforms in the chest impedance trans Thoracic Impedance (TTI) signal, and complete the calculation of relevant important parameters, the quality of cardiopulmonary resuscitation can be monitored and evaluated, so as to monitor and evaluate the quality of cardiopulmonary resuscitation (CPR). This paper presents an automatic detection algorithm for thoracic impedance signals based on density weighting and preference information. This method is aimed at the TTI signal of electroinduced cardiac arrest model in pigs. The potential compression and ventilation waveforms are marked by pre-processing and multi-resolution window search, and then the width, amplitude and adjacent waveform characteristic difference of each marker waveform are extracted as features, and the signals are segmented according to the marking waveform width. After that, the signal is decomposed by wavelet, and the ratio of the energy of each segment of the wavelet coefficient to the amplitude of the original waveform is extracted as the feature. Finally, the K-means clustering method based on density weighting and preference information is used to classify and recognize the marked waveforms. The experimental results show that the algorithm has a high accuracy and sensitivity to the recognition of pressure-pressing waveforms and waveform analysis in TTI signals. The robustness is good, and the running time is 0.43 s 卤0.07 s).
【作者单位】：重庆大学通信工程学院;第三军医大学生物医学工程与医学影像学院;
【基金】：国家自然科学基金(61108086) 重庆市自然科学基金(CSTC2011BB5066,CSTC2012jj A0612) 重庆市科技攻关计划项目(CSTC2012gg-yyjs0572) 中央高校基金(CDJZR10160003,CDJZR13160008) 军队博士后基金重庆市博士后基金资助课题
【分类号】：TN911.23

【共引文献】