摘要
目的研究开发一种便携式无创血红蛋白检测系统,实现人体总血红蛋白浓度的连续、实时监测。方法系统采用8个波长发光二极管(LED)作为光电容积脉搏波描记法(PPG)信号光源,联合约束独立成分分析和自适应滤波新方法消除PPG源信号的运动伪迹,应用主成分分析(PCA)和后向反馈人工神经网络(BP-ANN)相结合的新方法对PPG信号特征进行提取并利用十折交叉验证方法建立血红蛋白浓度的预测模型,无创系统样机和传统有创血液分析仪同步检测106例志愿者血红蛋白浓度。结果无创检测结果(SpHb_BP)和传统有创检测结果(Lab_tHb)的均方根误差为1.02g/dL,相关系数为0.73(P<0.001);Bland-Altman图显示SpHb_BP与Lab_tHb平均偏倚绝对值为0.05g/dL,99.1%(105/106)的点在95%一致性界限内。结论基于8波长PPG信号实现了无创血红蛋白检测系统研制,系统与临床有创血红蛋白浓度检测结果具有较好的一致性。
Objective To develop aportable,noninvasive hemoglobin concentration detection system capable of continuous and real-time monitoring of human total hemoglobin concentration.Methods The 8-wavelength broadband LED light source was adopted by the system as the PPG signal light source.A new method combining temporally independent component analysis(cICA)and adaptive filter was proposed for the removal of motion artifact in PPG signal.Another new method combining principal component analysis(PCA)and back propagation artificial neural network(BP-ANN)was used to extract the characteristic information of the PPG signals.In addition,a 10-fold cross-validation method was used to establish the prediction model for the total hemoglobin concentration.Simultaneous detection of hemoglobin concentration was conducted in 106 volunteers with the prototype of the non-invasive system and the traditional invasive blood analyzer.Results The root mean square error between non-invasive detection result(SpHb_BP)and traditional invasive detection result(Lab_tHb)was 1.02 g/dL and correlation coefficient was 0.73(P<0.001).The Bland-Altman plot showed that the mean absolute bias was 0.05 g/dL,and 99.1%(105/106)points were within 95%confidence interval of the agreement limits.Conclusion A noninvasive hemoglobin detection system was developed based on 8-wavelength PPG signal.The detection results of the new system are in good agreement with those of the clinical invasive detection.
作者
艾海明
彭福来
刘洪运
王国静
胡敏露
王卫东
Ai Haiming;Peng Fulai;Liu Hongyun;Wang Guojing;Hu Minlu;Wang Weidong(不详;Center of Medical Device R&D and clinical evaluation,Chinese PLA General Hospital,Beijing 100853,China)
出处
《航天医学与医学工程》
CAS
CSCD
北大核心
2019年第6期508-516,共9页
Space Medicine & Medical Engineering
基金
国家科技部重点研发计划(2013BAI03B03,2015BAI01B08)
国家自然科学基金(61372047,61701540)
北京市教委科技项目(KM201951160001)
