基于信号分析技术的呼吸机工作状态自动化识别研究  

Research on automatic identification of ventilator working state based on signal analysis technology

作　　者：张宝慧 陈艳林[1] 訾聪娜[1] 韩伟 滕金亮[1,2] ZHANG Baohui;CHEN Yanlin;ZI Congna;HAN Wei;TENG Jinliang(The First Affiliated Hospital of Hebei North University,Zhangjiakou 075000,China;Hebei Medical University,Shijiazhuang 050004,China)

机构地区：[1]河北北方学院附属第一医院,河北张家口075000 [2]河北医科大学,河北石家庄050004

出　　处：《电子设计工程》2024年第1期55-58,63,共5页Electronic Design Engineering

基　　金：张家口市重点研发计划项目(2322171D)。

摘　　要：呼吸机作为重症监护室内必不可少的设备,对重症患者起着重要的生命维持和抢救的作用。若呼吸机使用时出现了管路积液的问题且未被发觉,容易造成呼吸机无法正常工作。针对这一问题,以小波变换信号分析技术为基础,设计了呼吸机工作状态自动识别的方案。通过对呼吸波进行小波变换,对时-频信号进行有效提取,分析呼吸机管路是否积液,并通过自动识别保证医护人员对呼吸机工作状态能够实时掌握。实验结果表明,S1重构方式在临床数据以及模拟数据中均能实现较高的准确率,达到96.6%以上,因此该方法切实可行,具有实际应用价值。As an essential equipment in the intensive care unit,ventilators play an important role in maintaining and rescuing critically ill patients.However,there is a problem of fluid accumulation in the pipeline during the use of the ventilator,which is difficult to detect and can easily cause the ventilator to malfunction.Based on wavelet transform signal analysis technology,a scheme for automatic recognition of the working state of a ventilator was designed to address this issue.By performing wavelet transform on respiratory waves,time⁃frequency signals are effectively extracted,analyzing whether there is fluid accumulation in the ventilator pipeline,and ensuring that medical staff can grasp the working status of the ventilator in real time through automatic recognition.The experimental results show that the S1 reconstruction method can achieve a high accuracy of over 96.6%in both clinical and simulated data.Therefore,this method is feasible and has practical application value.

关 键 词：呼吸机 小波变换 管路积液 感受性曲线 小波重构 

分 类 号：TN98[电子电信—信息与通信工程]