基于小波的呼吸机管路积液自动检测算法研究  

Research on automatic detection algorithm of pipeline hydrops in ventilator based on wavelet

作　　者：潘清[1] 马树 张华青[2] 徐志江[1] PAN Qing;MA Shu;ZHANG Huaqing;XU Zhijiang(College of Information Engineering,Zhejiang University of Technology,Hangzhou 310023,China;Department of Clinical Engineering,the Second Affiliated Hospital of Zhejiang University School of Medicine,Hangzhou 310009,China)

机构地区：[1]浙江工业大学信息工程学院,浙江杭州310023 [2]浙江大学医学院附属第二医院临床医学工程部,浙江杭州310009

出　　处：《浙江工业大学学报》2020年第1期47-54,共8页Journal of Zhejiang University of Technology

摘　　要：呼吸机是医院重症监护室中最重要的生命支持设备,使用过程中易发生管路积液、积痰等问题,医护人员难以及时发现该问题,因而会造成呼吸机难以对病人进行正常机械通气。针对该问题,设计了基于小波分析的呼吸机管路积液自动检测算法。该算法通过对呼吸机的气道压力波形进行小波分解和重构,提取管路积液情况下呼吸机压力波形特征,自动提示医护人员发生积液的情况。基于临床模拟数据集和实测病人数据集对上述算法进行了测试。测试结果表明:基于小波变换的呼吸机管路内积液波形判断最高能达到97.2%的准确率。Ventilator is one of the most important life support device in the intensive care unit(ICU). Pipeline hydrops are prone to occur during use. It is difficult for medical staff to find the problem in time. It will lead to abnormal ventilation for the patients. This study developed an algorithm to detect the hydrops automatically based on wavelet analysis. The algorithm performs wavelet decomposition and reconstruction on the airway pressure waveform of the ventilator, extracts the features of the ventilator pressure waveform under the condition of hydrops in the pipeline, and automatically prompts the medical staff about the hydrops situation. The performance of the algorithm was tested based on clinical simulation data set and measured patient data set. The results indicate that the accuracy of the detection can reach up to 97.2%.

关 键 词：呼吸机 小波变换 波形分析 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]