基于深度学习的临床心电图分类算法

计算机与现代化 ›› 2021, Vol. 0 ›› Issue (08): 52-57.

基于深度学习的临床心电图分类算法

（1.中国科学院大学，北京100049；2.中国科学院微电子研究所，北京100029;
3.新一代通信射频芯片技术北京市重点实验室，北京100029）

出版日期:2021-08-19 发布日期:2021-08-19
作者简介:刘守华(1992—),男,安徽六安人,硕士研究生,研究方向:数据挖掘,E-mail: liushouhua@ime.ac.cn; 王小松(1981—),男,副研究员,博士,研究方向:高性能混合信号集成电路,生物电传感技术,物联网技术; 刘昱(1975—),男,研究员,博士,研究方向:高性能模拟,射频CMOS集成电路,硅基毫米波集成电路,超低功耗短距离无线通信系统,物联网相关技术及医疗电子系统集成技术。
基金资助:
天津市院市合作专项(18YFYSZC00130)

Clinical Electrocardiogram Classification Algorithm Based on Deep Learning

(1. University of Chinese Academy of Sciences, Beijing 100049, China;
2. Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China;
3. Beijing Key Laboratory of Radio Frequency IC Technology for Next Generation Communications, Beijing 100029, China)

Online:2021-08-19 Published:2021-08-19

摘要/Abstract

摘要： 心电图反映了人体心脏健康状况，是临床诊断心血管类疾病的重要依据。随着心电图数量的快速增长，计算机辅助心电图分析的需求愈加迫切，心电图自动分类作为实现计算机辅助心电图分析不可或缺的技术手段，具有重要的医学价值。由于心电信号非常微弱、抗干扰性差，传统心电图分类算法存在测试集上效果好，实际临床应用效果欠佳的问题。为此，本文研究一种基于多导联二维结构的一维卷积ResNet网络结构，通过平移起始点、“加噪”等数据增强手段增加训练样本多样性，并采用Focal Loss损失函数优化病人个体的心电图分类模型。该模型利用2万条完整的8导联心电图数据，共计34类心电异常事件进行分类实验，取得了0.91的F1值、93.96%的准确率和87.89%的召回率的分类性能。实验结果表明，该心电图分类算法模型具有较优的深层特征挖掘与分类能力，验证了其在心电异常自动分类上的有效性。

关键词: 深度学习, 残差网络, 卷积神经网络, 心电图, 数据分布, 损失函数

Abstract: Electrocardiogram (ECG) which can reflect the health state of human heart is widely used in clinical examination on heart diseases as an important basis. With the increasing number of ECG data, the demand of the computer-assisted electrocardiogram analysis has become urgent. Electrocardiogram automatic classification as an indispensable technical means of computer aided electrocardiogram analysis has important medical value. However, because of the weakness and low anti-interference of ECG signal, the traditional ECG classification algorithms have the problems of good effect on test set and poor effect in clinical application. So, this paper introduces a ResNet network structure of one-dimensional convolution based on multi-lead two-dimensional structure, increases the diversity of training samples by means of data enhancement such as translation starting point and adding noise, and uses Focal Loss function to optimize the ECG classification model of individual patients. The model uses 20000 complete 8-lead ECG data and a total of 34 types of abnormal ECG events for classification experiments. The results obtained are: F1 score 0.91, accuracy 93.96%, recall rate 87.89%. Experiment results show the proposed algorithm has better ability of deep feature mining and classification, which verifies its effectiveness in arrhythmia classification.

Key words: deep learning, residual network, convolutional neural network(CNN), ECG, data distribution, loss function

刘守华, 王小松, 刘昱, . 基于深度学习的临床心电图分类算法[J]. 计算机与现代化, 2021, 0(08): 52-57.

LIU Shou-hua , WANG Xiao-song , , LIU Yu , . Clinical Electrocardiogram Classification Algorithm Based on Deep Learning[J]. Computer and Modernization, 2021, 0(08): 52-57.

参考文献［27］

［1］	郭继鸿. 中国心脏性猝死现状与防治［J］. 中国循环杂志， 2013,28(5):323-326．
［2］	鲁力,史若飞. 心脏骤停心脏性猝死和现场心肺复苏［J］. 中国急救医学, 2012,32(11):1059-1062．
［3］	DOTSINSKY I. Clifford Gari D, Azuaje Francisco, McSharry Patrick E, Eds: Advanced methods and tools for ECG analysis［J］. Biomedical Engineering Online, 2007,6(1): Article Number 18.
［4］	YE C, KUMAR B V K V, COIMBRA M T. Heartbeat classification using morphological and dynamic features of ECG signals［J］. IEEE Transactions on Biomedical Engineering, 2012,59(10):2930-2941.
［5］	KUTLU Y, KUNTALP D. A multi-stage automatic arrhythmia recognition and classification system［J］. Computers in Biology & Medicine, 2011,41(1):37-45.
［6］	UBEYLI E D. Combining recurrent neural networks with eigenvector methods for classification of ECG beats［J］. Digital Signal Processing, 2009,19(2):320-329.
［7］	UBEYLI E D. Recurrent neural networks with composite features for detection of electrocardiographic changes in partial epileptic patients［J］. Computers in Biology and Medicine, 2008,38(1):401-410.
［8］	LEI W K, LI B N, DONG M C, et al. An application of morphological feature extraction and support vector machines in computerized ECG interpretation［C］// 2007 6th Mexican International Conference on Artificial Intelligence, Special Session. 2007:82-90.
［9］	LI X H, SHU L, HU H L. Kernel-based nonlinear dimensionality reduction for electrocardiogram recognition［J］. Neural Computing and Applications, 2009,18(8): Article Number 1013.
［10］	DOQUIRE G, DE LANNOY L, FRANCOIS D, et al. Feature selection for interpatient supervised heart beat classification［J］. Computational Intelligence and Neuroscience, 2011: Article ID 643816.
［11］	陈佩. 主成分分析法研究及其在特征提取中的应用［D］. 西安.陕西师范大学, 2014.
［12］	卞水荣,顾媛媛,赵强. PCA-SVM模式分类方法在心电信号分析中的应用［J］. 电子设计工程, 2018,28(20):37-41.
［13］	张泾周,李陈,李婷,等. 基于神经网络的心电信号分类方法研究［J］. 中国医疗器械杂志, 2008,32(3):183-185.
［14］	王润,贺兵兵. 基于神经网络的心电信号分类识别［J］. 现代计算机(专业版), 2017(23):37-40.
［15］	RAJPURKAR P, HANNUN A Y, HAGHPANAHI M, et al. Cardiologist-level arrhythmia detection with convolutional neural networks［J］. arXiv preprint arXiv:1707.01836, 2017.
［16］	HANNUN A Y, RAJPURKAR P, HAGHPANAHI M, et al. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network［J］. Nature Medicine, 2019,25(1):65-69.
［17］	JUN T J, NGUYEN H M, KANG D, et al. ECG arrhythmia classification using a 2-D convolutional neural network［J］. arXiv preprint arXiv:1804.06812, 2018.
［18］	金林鹏,董军. 面向临床心电图分析的深层学习算法［J］. 中国科学:信息科学, 2015,45(3):398-416.
［19］	ACHARYA U R, OH S L, HAGIWARA Y, et al. A deep convolutional neural network model to classify heartbeats［J］. Computers in Biology and Medicine, 2017,89:389-396.
［20］	CHAZAL P D, O’DWYER M , 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.
［21］	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition［C］// 2016 IEEE Conference on Computer Vision & Pattern Recognition. 2016:770-778.
［22］	SANDLER M, HOWARD A, ZHU M, et al. MobileNetV2: Inverted residuals and linear bottlenecks［C］// 2018 IEEE/CVF Conference on Computer Vision and Pattern. 2018:4510-4520.
［23］	MATEUSZ B, ATSUTO M, MAZUROWSKI M A. A systematic study of the class imbalance problem in convolutional neural networks［J］. Neural Networks, 2018,106:249-259.
［24］	LIN T Y, GOYAL P, GIRSHICK R, et al. Focal Loss for dense object detection［C］// 2017 IEEE International Conference on Computer Vision. 2017:2999-3007.
［25］	SHRIVASTAVA A, GUPTA A, GIRSHICK R. Training region-based object detectors with online hard example mining［C］// 2016 IEEE Conference on Computer Vision and Pattern Recognition. 2016:761-769.
［26］	LI T Y, ZHOU M. ECG classification using wavelet packet entropy and random forests［J］. Entropy, 2016,18(8):285-301．
［27］	KACHUEE M, FAZELI S, SARRAFZADEH M. ECG heartbeat classification: A deep transferable representation［C］// 2018 IEEE International Conference on Healthcare Informatics. 2018:443-444．

[1]	李健, 张克亮, 唐亮, 夏榕璟, 任静静. 面向中文命名实体识别任务的数据增强[J]. 计算机与现代化, 2022, 0(04): 1-6.
[2]	陈云翔, 王巍, 宁娟, 陈怡丹, 赵永新, 周庆华. PSWGAN-GP:改进梯度惩罚的生成对抗网络[J]. 计算机与现代化, 2022, 0(04): 21-26.
[3]	贺璇, 刘怡欣, 何小海, 卿粼波, 陈洪刚. 基于改进残差网络的联合损失步态特征识别[J]. 计算机与现代化, 2022, 0(04): 27-32.
[4]	梁正友, 耿经邦, 孙宇. 基于改进残差网络的交通标志识别算法[J]. 计算机与现代化, 2022, 0(04): 52-57.
[5]	谢辉, 师后勤, 齐宇霄, 陈瑞, 童莹. 基于注意力机制子网络的时空跌倒检测算法[J]. 计算机与现代化, 2022, 0(03): 70-75.
[6]	徐鑫强, 何鹏, . 基于图过滤框架对图卷积滤波器灵活性的研究[J]. 计算机与现代化, 2022, 0(03): 103-110.
[7]	胡焱, 卓书龙, 司成可. 数据驱动的ADS-B干扰源信号类型识别[J]. 计算机与现代化, 2022, 0(02): 19-25.
[8]	张晓航, 李政, 朱晓明, 张海锋, 赵博宇. 基于RBF神经网络的可信加密流量分类方法[J]. 计算机与现代化, 2022, 0(02): 45-51.
[9]	唐洁, 文元美. 基于3CNN-BiGRU的睡眠自动分期研究[J]. 计算机与现代化, 2022, 0(02): 120-126.
[10]	徐小亮, 赵英. 基于BiLSTM和ResCNN的实体关系抽取方法[J]. 计算机与现代化, 2022, 0(01): 10-16.
[11]	黄子平, 黄继风, 周小平. 基于YOLOv3的改进仪表检测算法[J]. 计算机与现代化, 2022, 0(01): 77-84.
[12]	冯敬翔. 基于迁移学习的卷积神经网络通道剪枝[J]. 计算机与现代化, 2021, 0(12): 13-18.
[13]	陈勋豪, 杨莹, 黄俊茹, 孙玉宝. 基于多尺度融合网络的视频快照压缩感知重建[J]. 计算机与现代化, 2021, 0(12): 58-64.
[14]	仇誉, 韩俊英, 封成智, 陈永卫. 基于卷积神经网络的苹果栽培品种识别[J]. 计算机与现代化, 2021, 0(12): 65-71.
[15]	汪帆, 魏宪, 郭杰龙, 梁培栋. 基于多通道分离整合的多尺度单幅图像去雨算法[J]. 计算机与现代化, 2021, 0(12): 72-78.