Accurate and Fast ElectroCardioGram Classification Method Based on Adaptive Fast S-Transform and XGBoost
-
摘要: 针对心电信号(ECG)传统分类方法效率较低的问题，该文提出一种基于自适应快速S变换(AFST)和XGBoost的心电信号精确快速分类方法。该方法首先通过快速定位算法确定心电信号特征频率点，再根据特征频率点自适应调节S变换窗宽因子，增强S变换的时频分辨率的同时避免迭代计算，大大减少运行时间。其次，基于自适应快速S变换的时频矩阵提取12个特征量来表征5种心电信号的特征信息，特征向量维数低，识别能力强。最后，利用XGBoost算法对特征向量进行识别。MIT-BIH心律失常数据库和患者实测数据验证表明，该方法显著地缩短了分类时间，对5种心电信号的分类准确率分别为99.59%和97.32%，适用于实际检测系统中心律失常疾病的快速诊断。Abstract: Considering the low efficiency of traditional ElectroCardioGram(ECG) classification methods, an accurate and fast ElectroCardioGram classification method based on Adaptive Fast S-Transform (AFST) and XGBoost is proposed. Firstly, the main feature points of the ECG signals are determined through a fast positioning algorithm, and then the S-Transform window width factor is adjusted adaptively according to the main feature points to enhance the time-frequency resolution of the S-transform while avoiding iterative calculation and reducing the running time greatly; Secondly, based on the time-frequency matrix of AFST, 12 eigenvalues are extracted to represent the characteristic information of 5 kinds of ECG signals, with low eigenvector dimension and strong recognition ability. Finally, XGBoost is used to identify the eigenvectors. The experimental studies based on the MIT-BIH arrhythmia database and the verification of patient measurement data show that, with the proposed method, the classification time of ECG signals is significantly shortened and classification accuracy of 99.59%, 97.32% is obtained respectively, which is suitable for the rapid diagnosis of abnormal diseases in the center rate of the actual detection system.
-
表 1 不同算法运算量分析
算法 运算量 复数乘法 复数加法 时间复杂度 ST $ ({N^2} + {N^3})/2 $ $ ({N^3} - {N^2})/2 $ $ O({N^3}) $ FST $ (2{N^2} + {N^2}{\log _2}N)/4 $ $ ({N^2}{\log _2}N)/2 $ $ O({N^2}\log {}_2N) $ AFST $ q(2N + N{\log _2}N)/2 $ $ qN{\log _2}N $ $ O(qN\log {}_2N) $ 
下载: 导出CSV
表 2 分类结果(%)
心电类型 $ {\text{Acc}} $ $ {\text{Sen}} $ $ {\text{Spe}} $ $ {\text{PPV}} $ NORM 99.68 99.06 99.30 98.95 LBBB 99.43 98.83 99.58 99.64 RBBB 99.52 99.54 99.90 99.96 APC 99.70 99.36 98.86 99.43 PVC 99.61 99.48 99.71 99.29 总计 99.59 99.25 99.47 99.45
下载: 导出CSV
表 3 各算法性能对比
特征提取方法 分类器 特征提取
时间(s)分类器运行
时间(s)准确率
(%)ST 随机森林 389.10 6.42 84.14 GBDT 4.98 80.29 XGBoost 4.35 85.95 改进XGBoost 4.35 88.53 AFST 随机森林 152.07 6.47 98.42 GBDT 4.98 97.74 XGBoost 4.36 99.06 改进XGBoost 4.36 99.59
下载: 导出CSV
表 4 不同分类方法对比(%)
方法 分类器 ECG特征 Acc Sen Spe PPV Wavelet[24] E-SVM 小波系数、R-R间期、形态特征 94.50 94.70 93.90 – Wavelet[8] RNN-LSTM 小波系数、R-R间期、形态特征 97.10 85.70 94.28 98.35 TSFEL[16] RT+SVM 统计学特征、时频特征 98.21 84.21 95.95 92.81 XWT[8] SVM 小波系数、R-R间期、时域特征 98.50 99.69 98.80 – SST[25] SVM SST系数、R-R间期、形态特征 84.71 80.43 70.75 – ST[11] GA-SVM R-R间期、时频特征、形态特征 99.74 99.42 99.83 – AFST 改进XGBoost 时频特征 99.59 99.25 99.47 99.45
下载: 导出CSV
-
[1] KREATSOUS C and ANAND S S. The impact of social determinants on cardiovascular disease[J]. Canadian Journal of Cardiology, 2010, 26 Suppl C: 8C–13C. [2] World Health Organization. Cardiovascular disease (CVDs)[EB/OL].https://www.who.int/en/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds), 2021. [3] AMORIM P, MORAES T, FAZANARO D, et al. Shearlet and contourlet transforms for analysis of electrocardiogram signals[J]. Computer Methods and Programs in Biomedicine, 2018, 161: 125–132. doi: 10.1016/j.cmpb.2018.04.021 [4] 吴志勇, 丁香乾, 许晓伟, 等. 基于深度学习和模糊C均值的心电信号分类方法[J]. 自动化学报, 2018, 44(10): 1913–1920. doi: 10.16383/j.aas.2018.c170417WU Zhiyong, DING Xiangqian, XU Xiaowei, et al. A method for ECG classification using deep learning and fuzzy C-means[J]. Acta Automatica Sinica, 2018, 44(10): 1913–1920. doi: 10.16383/j.aas.2018.c170417 [5] DE CHAZAL P, O’DWYER M, and REILLY R B. Automatic classification of heartbeats using ECG morphology and heartbeat interval features[J]. IEEE Transactions on Biomedical Engineering, 2004, 51(7): 1196–1206. doi: 10.1109/TBME.2004.827359 [6] MAR T, ZAUNSEDER S, MARTÍNEZ J P, et al. Optimization of ECG classification by means of feature selection[J]. IEEE Transactions on Biomedical Engineering, 2011, 58(8): 2168–2177. doi: 10.1109/TBME.2011.2113395 [7] 王金海, 史梦颖, 张兴华. 基于EMD和ApEn特征提取的心律失常分类研究[J]. 仪器仪表学报, 2016, 37(S1): 168–173. doi: 10.19650/j.cnki.cjsi.2016.s1.028WANG Jinhai, SHI Mengying, and ZHANG Xinghua. Classification of arrhythmias based on EMD and ApEn feature extraction[J]. Chinese Journal of Scientific Instrument, 2016, 37(S1): 168–173. doi: 10.19650/j.cnki.cjsi.2016.s1.028 [8] JACOB N and JOSEPH L A. Classification of ECG beats using cross wavelet transform and support vector machines[C]. 2015 IEEE Recent Advances in Intelligent Computational Systems (RAICS), Trivandrum, India, 2015. [9] 尹柏强, 邓影, 王署东, 等. 时频广义S变换和VL-MOBP神经网络在人体动作识别中的应用[J]. 电子测量与仪器学报, 2020, 34(11): 1–9. doi: 10.13382/j.jemi.B2003032YIN Baiqiang, DENG Ying, WANG Shudong, et al. Application of time-frequency generalized S transform and VL-MOBP neural network in human motion recognition[J]. Journal of Electronic Measurement and Instrumentation, 2020, 34(11): 1–9. doi: 10.13382/j.jemi.B2003032 [10] DAS M K and ARI S. Electrocardiogram beat classification using S-transform based feature set[J]. Journal of Mechanics in Medicine and Biology, 2014, 14(5): 1450066. doi: 10.1142/S0219519414500663 [11] YANG Jianyong and YAN Ruqiang. A multidimensional feature extraction and selection method for ECG arrhythmias classification[J]. IEEE Sensors Journal, 2021, 21(13): 14180–14190. doi: 10.1109/JSEN.2020.3047962 [12] RAJ S and RAY K C. ECG signal analysis using DCT-Based DOST and PSO optimized SVM[J]. IEEE Transactions on Instrumentation and Measurement, 2017, 66(3): 470–478. doi: 10.1109/TIM.2016.2642758 [13] SINHA N and DAS A. Identification and localization of myocardial infarction based on analysis of ECG signal in cross spectral domain using boosted SVM classifier[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 4007409. doi: 10.1109/TIM.2021.3117663 [14] HOU Borui, YANG Jianyong, WANG Pu, et al. LSTM-based auto-encoder model for ECG arrhythmias classification[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(4): 1232–1240. doi: 10.1109/TIM.2019.2910342 [15] 柯丽, 王丹妮, 杜强, 等. 基于卷积长短时记忆网络的心律失常分类方法[J]. 电子与信息学报, 2020, 42(8): 1990–1998. doi: 10.11999/JEIT190712KE Li, WANG Danni, DU Qiang, et al. Arrhythmia classification based on convolutional long short term memory network[J]. Journal of Electronics &Information Technology, 2020, 42(8): 1990–1998. doi: 10.11999/JEIT190712 [16] BHATTACHARYYA S, MAJUMDER S, DEBNATH P, et al. Arrhythmic heartbeat classification using ensemble of random forest and support vector machine algorithm[J]. IEEE Transactions on Artificial Intelligence, 2021, 2(3): 260–268. doi: 10.1109/TAI.2021.3083689 [17] KUNG B H, HU P Y, HUANG C C, et al. An efficient ECG classification system using resource-saving architecture and random forest[J]. IEEE Journal of Biomedical and Health Informatics, 2021, 25(6): 1904–1914. doi: 10.1109/JBHI.2020.3035191 [18] 王帅, 赵钟瑶, 张翔宇, 等. 穿戴式心电信号质量的三分类评估方法[J]. 中国生物医学工程学报, 2020, 39(5): 550–556. doi: 10.3969/j.issn.0258-8021.2020.05.005WANG Shuai, ZHAO Zhongyao, ZHANG Xiangyu, et al. Three-type classification method for wearable ECG signal quality[J]. Chinese Journal of Biomedical Engineering, 2020, 39(5): 550–556. doi: 10.3969/j.issn.0258-8021.2020.05.005 [19] 喻玲, 符玲, 熊思宇, 等. 基于最优Kaiser窗的动态同步相量测量算法[J]. 电网技术, 2022, 46(3): 1100–1108. doi: 10.13335/j.1000-3673.pst.2021.0602YU Ling, FU Ling, XIONG Siyu, et al. A dynamic synchrophasor estimation algorithm based on optimal Kaiser window[J]. Power System Technology, 2022, 46(3): 1100–1108. doi: 10.13335/j.1000-3673.pst.2021.0602 [20] 尹柏强, 何怡刚, 朱彦卿. 一种广义S变换及模糊SOM网络的电能质量多扰动检测和识别方法[J]. 中国电机工程学报, 2015, 35(4): 866–872. doi: 10.13334/j.0258-8013.pcsee.2015.04.013YIN Baiqiang, HE Yigang, and ZHU Yanqing. Detection and classification of power quality multi-disturbances based on generalized s-transform and fuzzy SOM neural network[J]. Proceedings of the CSEE, 2015, 35(4): 866–872. doi: 10.13334/j.0258-8013.pcsee.2015.04.013 [21] 王奕森, 夏树涛. 集成学习之随机森林算法综述[J]. 信息通信技术, 2018, 12(1): 49–55. doi: 10.3969/j.issn.1674-1285.2018.01.009WANG Yisen and XIA Shutao. A survey of random forests algorithms[J]. Information and Communications Technologies, 2018, 12(1): 49–55. doi: 10.3969/j.issn.1674-1285.2018.01.009 [22] 赵洪山, 闫西慧, 王桂兰, 等. 应用深度自编码网络和XGBoost的风电机组发电机故障诊断[J]. 电力系统自动化, 2019, 43(1): 81–86. doi: 10.7500/AEPS20180708001ZHAO Hongshan, YAN Xihui, WANG Guilan, et al. Fault diagnosis of wind turbine generator based on deep autoencoder network and XGBoost[J]. Automation of Electric Power Systems, 2019, 43(1): 81–86. doi: 10.7500/AEPS20180708001 [23] PAN Jiapu and TOMPKINS W J. A real-time QRS detection algorithm[J]. IEEE Transactions on Biomedical Engineering, 1985, BME-32(3): 230–236. doi: 10.1109/TBME.1985.325532 [24] CHIA N G, HAU Y W, and JAMALUDIN M N. Robust arrhythmia classifier using wavelet transform and support vector machine classification[C]. The 13th International Colloquium on Signal Processing & its Applications (CSPA), Penang, Malaysia, 2017. [25] HERRY C L, FRASCH M, SEELY A J, et al. Heart beat classification from single-lead ECG using the synchrosqueezing transform[J]. Physiological Measurement, 2017, 38(2): 171–187. doi: 10.1088/1361-6579/aa5070 -
下载:
图(10) / 表(4)
计量
- 文章访问数: 298
- HTML全文浏览量: 415
- PDF下载量: 82
- 被引次数: 0
