Abstract:

This paper proposes a continuous blood pressure prediction method based on improved SENet convolutional neural network and a self-learning parameter filter. The experimental results show that the improved SENet can effectively increase the prediction ability of simple convolutional neural networks for time series data. When the number of convolutional layers is two, three and four layers, the prediction accuracy is improved by 34.8%, 23.5 % and 36.0%. On this basis, the self-learning parameter filter is used to eliminate the burr in the predicted blood pressure waveform, and finally a smooth continuous blood pressure prediction result is obtained.

Key words: convolutional neural network, Sequeeze-and-Excitation Network（SENet）, blood pressure prediction, pulse information

摘要：

提出了基于改进SENet卷积神经网络和自学习参数滤波器的连续血压预测方法。实验结果表明，改进SENet可以有效增加简单卷积神经网络对时序数据的预测能力，在卷积层数为二层、三层和四层时比简单卷积神经网络预测精度提升了34.8%、23.5%和36.0%，在此基础上利用自学习参数滤波器消除血压预测波形中的毛刺，最终得到平滑的连续血压预测结果。

关键词: 卷积神经网络, SENet, 血压预测, 脉搏信息