摘要
通过低氧实验提出一种快速识别人体低氧状态的方法.通过搭建深层神经网络训练实验数据识别氧气体积分数(16%~21%)与人体可耐受极端低氧气体积分数(15. 5%~16%)条件下光电容积脉搏波(photoplethysmography,PPG)信号,获得人体生理状态的模式识别网络.经测试该网络的识别正确率可达92. 8%.利用混淆矩阵及接受者操作性能(receiver operating characteristic,ROC)曲线分析,混淆矩阵的训练集、验证集、测试集、全集识别正确率分别达到97. 9%、94. 8%、92. 8%和96. 3%,AUC (area under curve)值接近1,认为该网络分类性能优良,并且可在4 s内完成整个识别过程.
Due to the development of industrialization,low-oxygen environment has become common in the confined spaces of construction industries,chemical industries,military,urban underground spaces,and poorly ventilated crowed areas and caused a large number of hypoxic injuries. The traditional method of preventing hypoxic injuries is to monitor the oxygen concentration in the environment without considering the difference in oxygen tolerance limits when the human body is in different physiological states. Photoplethysmography(PPG) can comprehensively reflect physiological information,including heart rate,blood pressure,blood oxygen saturation,cardiovascular blood flow parameters,and respiratory rate. When the human body enters a hypoxic environment,the physiological parameters change rapidly,resulting in a change in the PPG signal. By measuring the PPG signal of the human body,the physiological state is considered to determine whether the human body reaches the oxygen tolerance limit. This study proposed a method for quickly identifying the hypoxic state of the human body using hypoxia experiment. According to the latest research on aviation medicine,mountain medicine and naval submarine medicine,15. 5% oxygen volume fraction can guarantee the basic life safety of personnel. Through the training experimental data of a constructed deep neural network,the PPG signal of a human in normal oxygen volume fraction(16%-21%) and extremely low-oxygen volume fraction( 15. 5%-16%) was determined to obtain the pattern recognition network of human physiological state. After testing,the recognition accuracy of the network could reach 92. 8%. Using the confusion matrix and receiver operating characteristic curve analysis,the accuracy rate of training set,verification set,test set,and ensemble recognition of the confusion matrix reached 97. 9%,94. 8%,92. 8%,and 96. 3%,respectively. The area under the curve value is close to 1,the network classification performance is excellent,and the entire identification process could be completed within 4 s.
引文
[1]Burlet-Vienney D,Chinniah Y,Bahloul A,et al.Occupational safety during interventions in confined spaces.Safety Sci,2015,79:19
[2]Zang T Z,Zhang L J,Zhang L,et al.Causes and countermeasures of casualty accident induced by unexpected factors in limited job space.J Nanjing Univ Technol Nat Sci Ed,2015,27(3):103(臧铁柱,张礼敬,张丽,等.有限空间作业意外伤亡事故的成因及其对策.南京工业大学学报(自然科学版),2005,27(3):103)
[3]Sun Y J,Shen H G.Industrial Ventilation.4th Ed.Beijing:China Architecture&Building Press,2010(孙一坚,沈恒根.工业通风.4版.北京:中国建筑工业出版社,2010)
[4]Zhou X F,Yang F,Guo L,et al.Analysis and preventive recommendation of national confined space accidents due to asphyxiation and poisoning from 2014 to 2015.J Eniviron Occup Med,2018,35(8):735(周兴藩,杨凤,郭玲,等.2014-2015年全国有限空间作业中毒与窒息事故分析及预防建议.环境与职业医学,2018,35(8):735)
[5]Mejías C,Jiménez D,Mu1oz A,et al.Clinical response of 20people in a mining refuge:study and analysis of functional parameters.Safety Sci,2014,63:204
[6]Selman J,Spickett J,Jansz J,et al.An investigation into the rate and mechanism of incident of work-related confined space fatalities.Safety Sci,2018,109:333
[7]Li G J.Research on Human Heat Tolerance under Extreme Hot,Humid and Low-oxygen Environment[Dissertaion].Tianjin:Tianjin University,2008(李国建.高温高湿低氧环境下人体热耐受性研究[学位论文].天津:天津大学,2008)
[8]Kamshilin A A,Nippolainen E,Sidorov I S,et al.A new look at the essence of the imaging photoplethysmography.Sci Rep,2015,5:10494
[9]Shin H,Min S D.Feasibility study for the non-invasive blood pressure estimation based on ppg morphology:normotensive subject study.Biomed Eng Online,2017,16(1):10
[10]Zhou Q,Tu H,Zuo J X.Blood oxygen saturation real-time monitoring system research based on PPG.Inform Res,2017,43(3):75(张强,涂浩,左佳鑫.基于PPG的血氧饱和度实时监测系统研究.信息化研究,2017,43(3):75)
[11]Njoum H,Kyriacou P A.Photoplethysmography for the assessment of haemorheology.Sci Rep,2017,7(1):1406
[12]Ma J L,Wang C,Li Z J,et al.Study of measuring heart rate and respiration rate based on PPG.Opt Tech,2011,37(3):309(马俊领,王成,李章俊,等.基于PPG的心率和呼吸频率的测量研究.光学技术,2011,37(3):309)
[13]Tamura T,Maeda Y,Sekine M,et al.Wearable photoplethysmographic sensors-past and present.Electronics,2014,3(2):282
[14]Khadse C B,Chaudhari M A,Borghate V B.Electromagnetic compatibility estimator using scaled conjugate gradient back-propagation based artificial neural network.IEEE Trans Ind Inform,2017,13(3):1036
[15]Akhtar N,Mian A.Threat of adversarial attacks on deep learning in computer vision:a survey.IEEE Access,2018,6:14410
[16]Razzak M I,Naz S,Zaib A.Deep learning for medical image processing:overview,challenges and the future//Classification in Bioapps:Automation of Decision Making.Springer,2018
[17]Wang H,Zhao Y,Xu Y M,et al.Cross-language speech attribute detection and phone recognition for Tibetan using deep learning//The 9th International Symposium on Chinese Spoken Language Processing.Singapore,2014:474
[18]Hoffman J,Pathak D,Tzeng E,et al.Large scale visual recognition through adaptation using joint representation and multiple instance learning.J Mach Learn Res,2016,17(1):1
[19]Le Cun Y,Bengio Y,Hinton G.Deep learning.Nature,2015,521(7553):436
[20]Bronstein M M,Bruna J,Le Cun Y,et al.Geometric deep learning:going beyond euclidean data.IEEE Signal Process Mag,2017,34(4):18
[21]Craig A,Tran Y,Wijesuriya N,et al.A controlled investigation into the psychological determinants of fatigue.Biol Psychol,2006,72(1):78
[22]Wang Q.Multi Objective Learning and Ensembling for Deep Neura Network Based Speech Enhancement[Dissertation].Hefei:University of Science and Technology China,2018(王青.基于深层神经网络的多目标学习和融合的语言增强研究[学位论文].合肥:中国科学技术大学,2018)