XIONG Lei,HE Peiyu,FANG Ancheng,et al.Classification of Arrhythmia Beat Model based on Residual-Attention and LSTM[J].Journal of Chengdu University of Information Technology,2022,37(02):119-124.[doi:10.16836/j.cnki.jcuit.2022.02.001]
基于残差-注意力和LSTM的心律失常心拍分类方法研究
- Title:
- Classification of Arrhythmia Beat Model based on Residual-Attention and LSTM
- 文章编号:
- 2096-1618(2022)02-0119-06
- Keywords:
- residual; attention; long short term memory; arrhythmia; beat classification
- 分类号:
- TP301.6
- 文献标志码:
- A
- 摘要:
- 长时心电图是长时间连续记录心电状态的一种心电图,心律失常在心电图上的表现为心拍频率不规则或者波形异常,故高效精准地从长时心电图中识别心律失常具有重要的临床意义。针对从长时心电图中识别不同心律失常类型的问题,提出基于注意力机制的残差和LSTM网络的心律失常心拍分类模型。首先利用小波变换对原始长时心电信号进行滤波处理,然后利用QRS波检测算法对R波波峰进行定位,并以R波波峰为基准将原始信号切分成心拍图,最后放入Residual-Attention和LSTM网络进行特征提取并实现分类。提出的模型对正常心拍、室性早搏心拍及室上性早搏心拍的三分类准确率为96.09%,比传统的CNN网络模型提高了3.26%; 三类心拍的F1值分别提高了2.04%、2.56%和5.30%。对比实验表明,提出的基于Residual-Attention和LSTM的心律失常心拍分类模型,相比传统的CNN模型有着更好的分类准确率和F1值。
- Abstract:
- Long term ECG is a kind of ECG that continuously records the ECG state for a long time. Arrhythmia is characterized by irregular beat frequency or abnormal waveform. Therefore, it is of great clinical significance to identify arrhythmia from long-term ECG efficiently and accurately.In order to recognize the problem from arrhythmia beats from long-term ECG, an arrhythmia beat classification model based on Residual-Attention mechanism and LSTM network is proposed.Firstly, long-term ECG signal is filtered by wavelet transform, then QRS wave algorithm is used to find the location of R-wave, and the R-wave is segmented into two-dimensional cardiogram based on the location of the peak of R-wave. Finally, it is put into Residual-Attention and LSTM network for feature extraction and classification. The accuracy of Normal、PVC and SPBclassification is 96.09%, which is 3.26% higher than the traditional Convolutional Neural Networks model. The F1 values of three classes are increased by 2.04%,2.56%,5.30% respectively. The comparative experimental results show that the proposed arrhythmia beat classification model based on residual attention and LSTM has better classification accuracy and F1 value than the traditional CNN model.
参考文献/References:
[1] Hoekema R,Uijen G J H,Van Oosterom A.Geometrical aspects of the interindividual variability of multilead ECG recordings[J].IEEE Transactions on Biomedical Engineering,2001,48(5):551-559.
[2] Yan Z,Zhou J,Wong W F.Energy efficient ECG classification with spiking neural network[J].Biomedical Signal Processing and Control,2021,63:102170.
[3] Salem M,Taheri S,Yuan J S.ECG arrhythmia classification using transfer learning from 2-dimensional deep CNN features[C].2018 IEEE Biomedical Circuits and Systems Conference(BioCAS).IEEE,2018:1-4.
[4] Li F,Chen K,Ling J,et al.Automatic diagnosis of cardiac arrhythmia in electrocardiograms via multigranulation computing[J].Applied Soft Computing,2019,80:400-413.
[5] Prakash A J,Ari S.A system for automatic cardiac arrhythmia recognition using electrocardiogram signal[M].England:Woodhead Publishing,2019:891-911.
[6] Ye C,Coimbra M T,Kumar B V K V.Arrhythmia detection and classification using morphological and dynamic features of ECG signals[C].2010 Annual International Conference of the IEEE Engineering in Medicine and Biology. IEEE, 2010: 1918-1921.
[7] Essa E,Xie X. Multi-model Deep Learning Ensemble for ECG Heartbeat Arrhythmia Classification[C].2020 28th European Signal Processing Conference(EUSIPCO).IEEE,2021:1085-1089.
[8] Kanani P,Padole M.ECG Heartbeat Arrhythmia Classification Using Time-Series Augmented Signals and Deep Learning Approach[J].Procedia Computer Science,2020,171:524-531.
[9] Romdhane T F,Pr M A.Electrocardiogram heartbeat classification based on a deep convolutional neural network and focal loss[J].Computers in Biology and Medicine,2020,123:103866.
[10] Kachuee M,Fazeli S,Sarrafzadeh M.Ecg heartbeat classification:A deep transferable representation[C].2018 IEEE International Conference on Healthcare Informatics(ICHI).IEEE,2018:443-444.
[11] Guo L,Sim G,Matuszewski B.Inter-patient ECG classification with convolutional and recurrent neural networks[J].Biocybernetics and Biomedical Engineering,2019,39(3):868-879.
[12] Izci E,Ozdemir M A,Degirmenci M,et al.Cardiac arrhythmia detection from 2d ecg images by using deep learning technique[C].2019 Medical Technologies Congress(TIPTEKNO).IEEE,2019:1-4.
[13] 杨浩,黄茂林,蔡志鹏,等.融合CNN和BiLSTM的心律失常心拍分类模型[J].中国生物医学工程学报,2020,39(6):719-726.
[14] Ma C,Wei S,Chen T,et al.Integration of results from convolutional neural network in a support vector machine for the detection of atrial fibrillation[J].IEEE Transactions on Instrumentation and Measurement,2020,70:1-10.
[15] 谢佳静,魏守水,江兴娥,等.基于CNN和频率切片小波变换的T波形态分类[J].中国生物医学工程学报,2021,40(1):1-11.
[16] Mathunjwa B M,Lin Y T,Lin C H,et al.ECG arrhythmia classification by using a recurrence plot and convolutional neural network[J].Biomedical Signal Processing and Control,2021,64:102262.
相似文献/References:
[1]冯金慧,陶宏才.基于注意力的深度协同在线学习资源推荐模型[J].成都信息工程大学学报,2020,35(02):151.[doi:10.16836/j.cnki.jcuit.2020.02.005]
FENG Jinhui,TAO Hongcai.An Attention-based Deep Collaborative Filtering Model for Online Course Recommendation[J].Journal of Chengdu University of Information Technology,2020,35(02):151.[doi:10.16836/j.cnki.jcuit.2020.02.005]
[2]张 姝,陶宏才.基于改进DeepFM的车险索赔预测模型的研究[J].成都信息工程大学学报,2021,36(03):311.[doi:10.16836/j.cnki.jcuit.2021.03.012]
ZHANG Shu,TAO Hongcai.Research on Prediction Model of Auto Insurance Claim based on Improved DeepFM[J].Journal of Chengdu University of Information Technology,2021,36(02):311.[doi:10.16836/j.cnki.jcuit.2021.03.012]
[3]陶全桧,安俊秀,陈宏松.基于跨模态融合ERNIE的多模态情感分析研究[J].成都信息工程大学学报,2022,37(05):501.[doi:10.16836/j.cnki.jcuit.2022.05.003]
TAO Quanhui,AN Junxiu,CHEN Hongsong.Multi-modal Sentiment Analysis based on Cross-modal Fusion ERNIE[J].Journal of Chengdu University of Information Technology,2022,37(02):501.[doi:10.16836/j.cnki.jcuit.2022.05.003]
[4]林弟忠,邹书蓉,符 颖.基于注意力类特定编码的小样本目标检测[J].成都信息工程大学学报,2022,37(05):527.[doi:10.16836/j.cnki.jcuit.2022.05.007]
LIN Dizhong,ZOU Shurong,FU Ying.Attention-based Class Encoding for Few-Shot Object Detection[J].Journal of Chengdu University of Information Technology,2022,37(02):527.[doi:10.16836/j.cnki.jcuit.2022.05.007]
[5]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]
[6]高瑜曼,高 琳,何 晋.基于Dual-Hard-Net网络的结直肠息肉分割算法[J].成都信息工程大学学报,2023,38(03):291.[doi:10.16836/j.cnki.jcuit.2023.03.007]
GAO Yuman,GAO Lin,HE Jin.Colorectal Polyp Segmentation Algorithm Using Dual-Hard-Net Network[J].Journal of Chengdu University of Information Technology,2023,38(02):291.[doi:10.16836/j.cnki.jcuit.2023.03.007]
[7]郑晓旭,舒珊珊,文成玉.基于注意力多分支卷积和Transformer的手写文本识别[J].成都信息工程大学学报,2023,38(06):649.[doi:10.16836/j.cnki.jcuit.2023.06.005]
ZHENG Xiaoxu,SHU Shanshan,WEN Chengyu.Handwritten Text Recognition based on Attentional Multi-branch Convolution and Transformer[J].Journal of Chengdu University of Information Technology,2023,38(02):649.[doi:10.16836/j.cnki.jcuit.2023.06.005]
备注/Memo
收稿日期:2021-09-30
基金项目:国家自然科学基金资助项目(62066042)