摘要
目的通过采集人体脉搏波信号和动脉血压值,建立基于BP神经网络的无创连续血压测量模型。方法将脉搏波信号进行预处理,计算其特征参数并筛选出相关性较高的特征参数,利用BP神经网络建立无创血压测量模型。结果收缩压、舒张压的仿真值和测得值的均方根误差分别为5.92、6.11 mmHg(1 mmHg=0.133 kPa)。结论该模型的仿真值对临床上血压连续测量具有一定的参考意义。
Objective A noninvasive continuous blood pressure measurement model based on BP neural network was established through collecting pulse wave signals and arterial blood pressure values.Methods The pulse wave signals were preprocessed,the characteristic parameters were calculated and then the parameters of high correlation were screened out.The noninvasive continuous blood pressure measurement model was established based on BP neural network.Results The root mean square errors of simulated and measured values of systolic and diastolic blood pressure were 5.92,6.11 mmHg(1 mmHg=0.133 kPa),respectively.Conclusion The simulated results proved that the model could provide references for continuous blood pressure measurement in clinic to a certain extent.
作者
成刚
查晓俊
Cheng Gang;Zha Xiaojun(The Drum Tower Hospital Affiliated to Medical School of Nanjing University,Nanjing Jiangsu 210008,China)
出处
《医疗装备》
2021年第11期3-5,共3页
Medical Equipment
关键词
脉搏波
动脉血压
BP神经网络
无创血压
仿真值
Pulse wave
Arterial blood pressure
BP neural network
Noninvasive blood measure
Simulated result
