QIN Leiliang,HE Peiyu,FANG Ancheng,et al.Research on Five-phase Classification Method of Korotkoff Sounds based on CNN-LSTM[J].Journal of Chengdu University of Information Technology,2022,37(02):125-130.[doi:10.16836/j.cnki.jcuit.2022.02.002]
基于CNN-LSTM的柯氏音五时相分类方法研究
- Title:
- Research on Five-phase Classification Method of Korotkoff Sounds based on CNN-LSTM
- 文章编号:
- 2096-1618(2022)02-0125-06
- Keywords:
- Korotkoff sounds; CNN; LSTM; 10-fold cross-validation
- 分类号:
- TP301.6
- 文献标志码:
- A
- 摘要:
- 采用柯氏音进行血压测量的人工听诊法是间接测量血压的方法之一。柯氏音有5个时相,分别为弹响音、杂音、拍击音、捂音、消失音。在用人工听诊法时,弹响音开始的第一声对应的袖带压力为舒张压,消失音出现时对应的袖带压力为收缩压。由于孕妇及儿童测量舒张压需以捂音为判断标准,因此,对于不同人群的血压测量而言,识别柯氏音的时相具有重要意义。为精确识别柯氏音五时相,采用CNN加LSTM的分类模型进行柯氏音五时相的分类。同时为解决过拟合的问题,采用10折交叉验证。结果表明采用CNN和LSTM的融合模型对柯氏音的5个时相分类的平均准确率为88.69%,相比于单独的CNN准确率85.42%和单独的LSTM准确率81.39%分别提高了3.27%和7.30%。
- Abstract:
- The manual auscultation method of blood pressure measurement using Ko-sound is one of the methods of indirect blood pressure measurement. Offset sound of Ko-sound has five phases, such as namely, snapping sound, noise, slapping sound, cover sound, and disappearing sound. In the manual auscultation method, the cuff pressure corresponding to the first sound of the buzzing sound is the systolic pressure, and the cuff pressure corresponding to the disappearing sound is the systolic pressure. As for the measure diastolic blood pressure of pregnant women and children, it is necessary to use the muffled sound as the criterion. Therefore, it is of great significance to identify the phase of Korotkoff sounds for blood pressure measurement of different groups of people. In order to accurately identify the five-tempo of the offset, the classification model of CNN with LSTM is used to classify the five-tempo of the offset. At the same time, to solve the problem of over-fitting, 10-fold cross-validation is used. The results show that the fusion model of CNN and LSTM has an average accuracy of 88.69% for the five-phase classification of skew sounds, which is 3.27% higher than the accuracy of CNN alone of 85.42% and 7.30% the accuracy of LSTM alone of 81.39%.
参考文献/References:
[1] 王文,张维忠,孙宁玲,等.中国血压测量指南[J].中华高血压杂志,2011,19(12):1101-1115.
[2] 黄青霞,曹云云,程晓萍,等.二次改良柯氏音法在心血管疾病病人血压测量及管理中的应用研究[J].护理研究,2019,33(19):3333-3337.
[3] J Allen,A Murray.Time-frequencyanalysisKorotkoffof sounds[C].IEE Colloquium on Time-Frequency Analysis of Biomedical Signals 1997.
[4] 王锡山.未来医学时代——人工智能诊疗[J].中华结直肠疾病电子杂志,2017,6(4):349-352.
[5] Argha,Celler,Lovell.A Novel Automated Blood Pressure Estimation Algorithm Using Sequences of Korotkoff Sounds[J].IEEE Journal of Biomedical and Health Informatics,2021,25(4):1257-1264.
[6] B G Celler,P N Le,A Argha.GMM-HMM-Based Blood Pressure Estimation Using Time-Domain Features[J].IEEE Transactions on Instrumentation and Measurement,2020,69(6):3631-3641.
[7] Yin Wenfeng,Yang Xiuzhu,Lin Zhang,et al. ECG monitoringsystem integrated with IR-UWB radar based on CNN [J].IEEE Access,2016,4(99):6344-6351.
[8] Latif,Usman,Rana,et al.Phonocardiographic sensing using deep learning for Abnormal heartbeat detectio[J].IEEE Sensors Journal,2018,18(22):9393-9400.
[9] 胡欣宇,王昕波,赵召龙,等.基于柯氏音法与示波法结合的新型血压测量系统[J].软件,2017,38(3):78-81.
[10] 李刚,王宏,林凌.基于血压形成原理的血压测量方法及应用[J].计量技术,2007(10):16-19.
[11] 张政波,吴太虎.无创血压测量技术与进展[J].中国医疗器械杂志,2003,27(3):196-199.
[12] K Tara,A K Sarkar,M A G Khan,et al.Detection of cardiac disorder using MATLAB based graphical user interface(GUI)[C].2017 IEEE Region 10 Humanitarian Technology Conference(R10-HTC),2017:440-443.
[13] S Y Wong,S Yong Lim.Towards a Livelier Electromagnetic Education with an Interactive MATLAB-based GUI[C].2019 IEEE Asia-Pacific Conference on Applied Electromagnetics(APACE),2019:1-4.
[14] 李响,李国正,邓明君,等.基于语音频谱图像特征的人体疲劳检测方法[J].仪器仪表学报,2021,42(2):123-132.
[15] H Yanagisawa,T Yamashita,H Watanabe.A study on object detection method from manga images using CNN[C].2018 International Workshop on Advanced Image Technology(IWAIT),2018:1-4.
[16] 熊一橙,徐炜,张锐,等.基于LSTM网络的长江上游流域径流模拟研究[J].水电能源科学,2021,39(9):22-24.
[17] ZhangY,DaisS,Song W,et al.Exposing speech resampling manipulation by local texture analysis onspectrogramimages [J].Electronics,2020,9(1):1-16.
相似文献/References:
[1]杨 芮,文 武,徐 虹.基于PCC-CNN-GRU的短期风电功率预测[J].成都信息工程大学学报,2022,37(02):165.[doi:10.16836/j.cnki.jcuit.2022.02.009]
YANG Rui,WEN Wu,XU Hong.Short-term Wind Power Prediction based on PCC-CNN-GRU[J].Journal of Chengdu University of Information Technology,2022,37(02):165.[doi:10.16836/j.cnki.jcuit.2022.02.009]
[2]郭楠馨,林宏刚,张运理,等.基于元学习的僵尸网络检测研究[J].成都信息工程大学学报,2022,37(06):615.[doi:10.16836/j.cnki.jcuit.2022.06.001]
GUO Nanxin,LIN Honggang,ZHANG Yunli,et al.Botnet Detection Method based on Meta-Learning Network[J].Journal of Chengdu University of Information Technology,2022,37(02):615.[doi:10.16836/j.cnki.jcuit.2022.06.001]
[3]戴宇睿,安俊秀,李焯炜.基于信号分解降噪的CNN-BiLSTM金融市场趋势预测[J].成都信息工程大学学报,2023,38(01):28.[doi:10.16836/j.cnki.jcuit.2023.01.005]
DAI Yurui,AN Junxiu,LI Zhuowei.CNN-BiLSTM Financial Market Trend Prediction based on Signal Decomposition and Noise Reduction[J].Journal of Chengdu University of Information Technology,2023,38(02):28.[doi:10.16836/j.cnki.jcuit.2023.01.005]
备注/Memo
收稿日期:2021-10-11
基金项目:四川省科技计划资助项目(2020YJ0282)