摘要
正压通气治疗机是临床中重要的呼吸支持设备,如何提高通气中呼吸异常预警的实时性和通气的顺应性是当前面临的关键问题。本研究结合小波算法的多变率分解技术和Kalman滤波器组的同时最优估计方法,设计了基于小波卡尔曼滤波的呼吸信号预测方案。并将其与常用预测方法的预测结果作对比分析。结果表明,该方案提高了呼吸压力、流量信号预测的实时性和准确性,并且该方案的算法结构简单,易于移植到可穿戴设备,为进一步提升正压通气治疗机的自动化水平提供了研究基础。
Positive pressure ventilator is an important respiratory support equipment in clinical practice.How to improve the real-time warning of abnormal breathing and ventilation compliance is the key problem in the current development trend.Combined with the polytropic decomposition technology of wavelet algorithm and the simultaneous optimal estimation method of Kalman filter bank,we designed a respiratory signal prediction scheme based on wavelet Kalman filter.Compared with the characteristics of common prediction methods,this scheme improved the real-time performance and accuracy of the prediction of respiratory pressure and flow signal.Moreover,due to the simple algorithm structure of this scheme,it was easy to be transplanted to wearable devices.It will provide a research basis for further improving the automation level of positive pressure ventilation therapy machines.
作者
周婷
车波
邓林红
ZHOU Ting;CHE Bo;DENG Linhong(Institute of Biomedical Engineering and Health Science,Changzhou University,Changzhou 213164,China;School of Computer Science and Artificial Intelligence,Changzhou University,Changzhou 213164)
出处
《生物医学工程研究》
2021年第4期348-353,共6页
Journal Of Biomedical Engineering Research
基金
国家自然科学基金资助项目(12072048)。
关键词
正压通气
呼吸波形
卡尔曼滤波
小波算法
并行分解
实时预测
Positive pressure ventilate
Respiratory waveforms
Kalman filter
Wavelet algorithm
Parallel decomposition
Real-time prediction
