生物医学工程学杂志

生物医学工程学杂志

基于深度残差卷积神经网络的心电信号心律不齐识别

查看全文

心电图(ECG)信号在采集过程中容易受内部和外部噪声干扰,而且不同患者的 ECG 信号形态特征差异较大,即使同一患者在不同时间和环境下其 ECG 信号也会有差异,因此 ECG 信号特征检测与识别在心脏病远程实时监测与智能诊断中具有一定难度。基于此,本研究提出将小波自适应阈值去噪和深度残差卷积神经网络算法用于多种心律不齐的信号识别过程中。其中,使用小波自适应阈值技术完成 ECG 信号滤波,并设计了包含多个残差块(residual block)结构的 20 层卷积神经网络(CNN),即深度残差卷积神经网络(DR-CNN),对 5 大类心律不齐 ECG 信号进行了识别。然后,本文采用残差块局部神经网络结构单元构建 DR-CNN,缓解了深层网络的收敛难、调优难等问题,克服了 CNN 随着网络层数增加而导致的退化问题;进一步引入批标准化(batch normalization)技术,保证了网络的平滑收敛。按照美国医疗器械促进协会(AAMI)的心搏分类标准,使用麻省理工学院和波士顿贝丝以色列医院(MIT-BIH)心律不齐数据库中 94 091 个 ECG 心搏信号(2 个导联),完成了心律不齐多分类、室性异位搏动(Veb)和室上性异位搏动(Sveb)等分类识别实验。实验结果表明,本文所提出的方法在 ECG 信号多分类、Veb 和 Sveb 识别中的准确率分别达到了 99.034 9%、99.498 0% 和 99.334 7%。在相同的数据集和实验平台下,DR-CNN 在分类准确率、特异性和灵敏度上均优于相同结构复杂度的 CNN、深度多层感知机等传统算法。DR-CNN 算法提高了心律不齐智能诊断的精度,该方法与可穿戴设备、物联网和无线通信技术相结合,可以将心脏病的预防、监测和诊断延伸到家庭、养老院等院外场景,从而提高心脏病患者的救治率,并且有效地节约医疗资源。

Electrocardiogram (ECG) signals are easily disturbed by internal and external noise, and its morphological characteristics show significant variations for different patients. Even for the same patient, its characteristics are variable under different temporal and physical conditions. Therefore, ECG signal detection and recognition for the heart disease real-time monitoring and diagnosis are still difficult. Based on this, a wavelet self-adaptive threshold denoising combined with deep residual convolutional neural network algorithm was proposed for multiclass arrhythmias recognition. ECG signal filtering was implemented using wavelet adaptive threshold technology. A 20-layer convolutional neural network (CNN) containing multiple residual blocks, namely deep residual convolutional neural network (DR-CNN), was designed for recognition of five types of arrhythmia signals. The DR-CNN constructed by residual block local neural network units, alleviated the difficulty of deep network convergence, the difficulty in tuning and so on. It also overcame the degradation problem of the traditional CNN when the network depth was increasing. Furthermore, the batch normalization of each convolution layer improved its convergence. Following the recommendations of the Association for the Advancements of Medical Instrumentation (AAMI), experimental results based on 94 091 2-lead heart beats from the MIT-BIH arrhythmia benchmark database demonstrated that our proposed method achieved the average detection accuracy of 99.034 9%, 99.498 0% and 99.334 7% for multiclass classification, ventricular ectopic beat (Veb) and supra-Veb (Sveb) recognition, respectively. Using the same platform and database, experimental results showed that under the comparable network complexity, our proposed method significantly improved the recognition accuracy, sensitivity and specificity compared to the traditional deep learning networks, such as deep Multilayer Perceptron (MLP), CNN, etc. The DR-CNN algorithm improves the accuracy of the arrhythmia intelligent diagnosis. If it is combined with wearable equipment, internet of things and wireless communication technology, the prevention, monitoring and diagnosis of heart disease can be extended to out-of-hospital scenarios, such as families and nursing homes. Therefore, it will improve the cure rate, and effectively save the medical resources.

关键词: 心电图; 小波自适应滤波; 深度残差卷积神经网络; 心律不齐分类; 美国医疗器械促进协会

Key words: electrocardiogram; wavelet adaptive filtering; deep residual convolutional neural network; arrhythmia classification; the Association for the Advancements of Medical Instrumentation

登录后 ,请手动点击刷新查看全文内容。 没有账号,
登录后 ,请手动点击刷新查看图表内容。 没有账号,
1. Huikuri H V, Castellanos A, Myerburg R J. Sudden death due to cardiac arrhythmias. N Engl J Med, 2001, 345(12): 1473-1482.
2. Homaeinezhad M R, Atyabi S A, Tavakkoli E, et al. ECG arrhythmia recognition via a neuro-SVM-KNN hybrid classifier with virtual QRS image-based geometrical features. Expert Syst Appl, 2012, 39(2): 2047-2058.
3. Guo Shuli, Han Lina, Liu Hongwei, et al. The future of remote ECG monitoring systems. J Geriatr Cardiol, 2016, 13(6): 528-530.
4. Yin Wenfeng, Yang Xiuzhu, Zhang Lin, et al. ECG monitoring system integrated with IR-UWB radar based on CNN. IEEE Access, 2016, 4: 6344-6351.
5. Clifford G, Tarassenko L, Townsend N. One-pass training of optimal architecture auto-associative neural network for detecting ectopic beats. Electron Lett, 2001, 37: 1126-1127.
6. Moein S. An MLP neural network for ECG noise removal based on Kalman filter. Adv Exp Med Biol, 2010, 680: 109-116.
7. Inan O T, Giovangrandi L, Kovacs G T. Robust neural-network-based classification of premature ventricular contractions using wavelet transform and timing interval features. IEEE Trans Biomed Eng, 2006, 53(12): 2507-2515.
8. Yu S N, Chen Y H. Electrocardiogram beat classification based on wavelet transformation and probabilistic neural network. Pattern Recognit Lett, 2007, 28(10): 1142-1150.
9. Ebrahimzadeh A, Khazaee A. Detection of premature ventricular contractions using MLP neural networks: a comparative study. Measurement, 2010, 43(1): 103-112.
10. Li Duan, Zhang Hongxin, Zhang Mingming. Wavelet de-noising and genetic algorithm-based least squares twin SVM for classification of arrhythmias. Circuits Systems and Signal Processing, 2017, 36(7): 2828-2846.
11. 金林鹏, 董军. 面向临床心电图分析的深层学习算法. 中国科学:信息科学, 2015, 45(3): 398-416.
12. Rajpurkar P, Hannun A Y, Haghpanahi M, et al. Cardiologist-level arrhythmia detection with convolutional neural networks. Cornell University arXiv. (2017-07-06). http://arxiv.org/abs/1707.01836v1.
13. Kiranyaz S, Ince T, Gabbouj M. Real-time patient-specific ECG classification by 1-D convolutional neural networks. IEEE Trans Biomed Eng, 2016, 63(3): 664-675.
14. Wu Z Y, Ding X Q, Zhang G R. A novel method for classification of ECG arrhythmias using deep belief networks. International Journal of Computational Intelligence and Applications, 2016, 15(4): 1650021.
15. 陈诗慧, 刘维湘, 秦璟, 等. 基于深度学习和医学图像的癌症计算机辅助诊断研究进展. 生物医学工程学杂志, 2017, 34(2): 314-319.
16. 张贤达. 现代信号处理, 译. 第2版. 北京: 清华大学出版社, 2002.
17. Milchevski A, Gusev M. Improved pipelined wavelet implementation for filtering ECG signals. Pattern Recognit Lett, 2017, 95: 85-90.
18. Sahoo S, Kanungo B, Behera S, et al. Multiresolution wavelet transform based feature extraction and ECG classification to detect cardiac abnormalities. Measurement, 2017, 108: 55-66.
19. 王媛媛, 周涛, 陆惠玲, 等. 基于集成卷积神经网络的肺部肿瘤计算机辅助诊断模型. 生物医学工程学杂志, 2017, 34(4): 543-551.
20. He Kaiming, Zhang Xiangyu, Ren Shaoqing, et al. Deep residual learning for image recognition//2016 IEEE Conference on Computer Vision and Pattern Recognition (CPVR). Las Vegas, USA, 2016: 770-778.
21. Yu Lequan, Chen Hao, Dou Qi, et al. Automated melanoma recognition in dermoscopy images via very deep residual networks. IEEE Trans Med Imaging, 2017, 36(4): 994-1004.
22. Zagoruyko S, Komodakis N. Wide residual networks. (2017-07-14). http://arxiv.org/abs/1605.07146.
23. Bahoura M, Hassani M, Hubin M. DSP implementation of wavelet transform for real time ECG wave forms detection and heart rate analysis. Comput Meth Prog Bio, 1997, 52(1): 35-44.