Automatic Identifcation of Heart Block Precise Location Based on Sparse Connection Residual Network
-
摘要: 目的 实现对右束支阻滞、左束支阻滞和正常心电信号进行自动分类。方法 以MTI-BIH数据库为实验数据来源,从中提取训练集和测试集数据用于训练和测试网络模型,基于卷积神经网络提出核心算法:稀疏连接残差网络。将稀疏连接残差网络与已有的经典网络模型进行对比,以评估模型的识别效果。结果 稀疏连接残差网络的测试集准确率为95.2%,识别结果优于经典网络模型。结论 该文提出的算法能够辅助医生进行心脏传导阻滞类疾病的诊断,有一定的临床应用价值。Abstract: Objective To classify Right Bundle Branch Block (RBBB),Left Bundle Branch Block (LBBB) and normal ECG signals automatically.Methods The MIT-BIH database was used as experimental data sources.The training set and test set were extracted for training and testing network models.Based on convolutional neural network,this paper proposed the core algorithm:sparse connection residual network.Compared the sparse connected residual network with classic network models,then evaluated the recognition effect of the model.Results The accuracy of the test set the MIT-BIH database was 95.2%,the result is better than classic network models.Conclusion The algorithm proposed in this paper can assist doctors in the diagnosis of heart block related disease and place a high value on clinical application.
-
Keywords:
- heart block /
- deep learning /
- CNN
-
-
[1] 钟杭美. 临床12导联同步静态心电图[M]. 北京:中国医药科技出版社, 2013. [2] 王龙. 心脏房室传导阻滞的分类与心电图表现[J]. 中华全科医师杂志, 2009, 8(6):429-430. [3] 邓勇, 施文康, 刘琪. 小波变换的信号分形分析及其在心电信号处理中的应用研究[J]. 物理学报, 2002, 51(4):759-762. [4] KIRANYAZ S, INCE T, GABBOUJ M. Real-Time Patient-Specific ECG Classification by 1D Convolutional Neural Networks.[J]. IEEE Trans Biomed Eng, 2016, 63(3):664.
[5] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proc IEEE, 1998, 86(11):2278-2324.
[6] RAJPURKAR P, HANNUN A Y, HAGHPANAHI M, et al. Cardiologist-Level Arrhythmia Detection with Convolutional Neural Networks[J]. arXiv preprint arXiv:1707.01836, 2017.
[7] ISIN A, OZDALILI S. Cardiac arrhythmia detection using deep learning[J]. Proc Comput Sci, 2017, 120:268-275.
[8] ROBINSON A J. An application of recurrent nets to phone probability estimation[J]. IEEE T Neural Network, 1994, 5(2):298-305.
[9] HINTON G E, OSINDERO S, TEH Y W. A fast learning algorithm for deep belief nets[J]. Neur Comput, 2006, 18(7):1527-1554.
[10] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Comput, 1997, 9(8):1735-1780.
[11] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[C]. International Conference on Neural Information Processing Systems. MIT Press, 2014:2672-2680.
[12] NAIR V, HINTON G E. Rectified linear units improve restricted Boltzmann machines[C].Proceedings of the 27th international conference on machine learning (ICML-10). Haifa, 2010:807-814.
[13] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014:1-9.
[14] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[J]. Comput Sci, 2014.
[15] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]. Computer Vision and Pattern Recognition. IEEE, 2015:770-778.
[16] Krizhevsky A, Sutskever I, Hinton G. Imagenet classification with deep convolutional neural networks[C]. Proceedings of Advances in Neural Information Processing Systems, Lake Tahoe, USA, 2012:1097-1105.
[17] IOFFE S, SZEGEDY C. Batch normalization:accelerating deep network training by reducing internal covariate shift[C]. Proc 32nd Int Conf Mach Learn, 2015, 37:448-456.
计量
- 文章访问数: 13
- HTML全文浏览量: 0
- PDF下载量: 1