摘要
研究了无袖带方式的血压测量。利用脉搏信号的时序数据,融合脉搏传导时间特征,构建了基于多生理参数的LSTM异常血压预测模型。实验结果表明,本文模型在准确率、召回率和F1值等评估指标上优于其他非时序模型,可以为多生理参数的连续血压测量提供数据支持。
In this paper,cuffless blood pressure measurement is investigated.Using the time series data of pulse signal and integrating the characteristics of pulse transmission time,the LSTM abnormal blood pressure prediction model based on multiple physiological parameters is constructed.The experimental results show that this model is better than other models in accuracy,recall and F1 value,and can provide support for continuous blood pressure measurement with multiple physiological parameters.
作者
宋伟泽
Song Weize(Department of Information,Women’s Hospital School of Medicine Zhejiang University,Hangzhou,Zhejiang 310006,China)
出处
《计算机时代》
2023年第7期70-73,共4页
Computer Era
关键词
无袖带式血压测量
多生理参数
脉搏波
长短期记忆网络(LSTM)
cuffless blood pressure measurement
multiple physiological parameters
pulse wave
long short-term memory(LSTM)
