A Study on User Recognition Using the Generated Synthetic Electrocardiogram Signal
Abstract
:1. Introduction
2. Related Works
2.1. Generated Synthetic Data Based on Deep Learning
2.2. Deep Learning Networks Design Using ECG Signals
3. Proposed User Recognition Using Synthetic ECG Signal
3.1. Synthetic ECG Generation of GAN Using Auxiliary Classifier
3.2. Ensemble Networks Design of Parallel Structure
4. Experimental Results
- Lying down: acquiring signals for 1-min in the lying posture after rest state
- Standing: acquiring signals for 1-min in the standing posture after rest state
- Before exercise: acquiring signals for 1-min in the sitting position before exercise after rest state
- After exercise: acquiring signals for 1-min while maintaining the heart rate above 120 through the stepper exercise equipment after exercise
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Simoons, M.; Hugenholtz, P. Gradual changes of ECG waveform during and after exercise in normal subjects. Circulation 1975, 52, 570–577. [Google Scholar] [CrossRef]
- Kim, J.S.; Kim, S.H.; Pan, S.B. Personal recognition using convolutional neural network with ECG coupling image. J. Ambient. Intell. Humaniz. Comput. 2019, 11, 1923–1932. [Google Scholar] [CrossRef]
- Tantawi, M.M.; Revett, K.; Salem, A.; Tolba, M.F. Fiducial feature reduction analysis for electrocardiogram (ECG) based biometric recognition. J. Intell. Inf. Syst. 2012, 40, 17–39. [Google Scholar] [CrossRef]
- Odinaka, I.; Lai, P.H.; Kaplan, A.D.; O’Sullivan, J.A.; Sirevaag, E.J.; Rohrbaugh, J.W. ECG biometric recognition: A compara-tive analysis. IEEE Trans. Inf. Forensics Secur. 2012, 7, 1812–1824. [Google Scholar] [CrossRef]
- Ogiela, M.R.; Ogiela, L. On Using Cognitive Models in Cryptography. In Proceedings of the 2016 IEEE 30th International Conference on Advanced Information Networking and Applications, Caserta, Italy, 15–17 April 2016; pp. 1055–1058. [Google Scholar]
- Goodfellow, I.J.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Bengio, Y. Generative adversarial nets. In Advances in Neural Information Processing Systems; MIT Press: Cambridge, MA, USA, 2014; pp. 2672–2680. [Google Scholar]
- Srivastava, A.; Valkov, L.; Russell, C.; Gutmann, M.U.; Sutton, C. Veegan: Reducing mode collapse in gans using implicit variational learning. In Advances in Neural Information Processing Systems; MIT Press: Cambridge, MA, USA, 2017; pp. 3308–3318. [Google Scholar]
- Odena, A.; Olah, C.; Shlens, J. Conditional image synthesis with auxiliary classifier GANs. In Proceedings of the International Conference on Machine Learning, Sydney, Australia, 6–11 August 2017; pp. 2642–2651. [Google Scholar]
- McSharry, P.E.; Clifford, G.D.; Tarassenko, L.; Smith, L.A. A dynamical model for generating synthetic electrocardiogram signals. IEEE Trans. Biomed. Eng. 2003, 50, 289–294. [Google Scholar] [CrossRef]
- Christian, L.; Lucas, T.; Ferenc, H.; Jose, C.; Andrew, C.; Alejandro, A.; Andrew, A.; Alykhan, T.; Johannes, T.; Zehan, W.; et al. Photo-realistic single image super-resolution using a generative adversarial network. Computer Vision and Pattern Recognition. arXiv 2016, arXiv:1609.04802. [Google Scholar]
- Hartmann, K.G.; Schirrmeister, R.T.; Ball, T. EEG-GAN: Generative adversarial networks for electroencephalograhic brain signals. arXiv 2018, arXiv:1806.01875. [Google Scholar]
- Golany, T.; Radinsky, K. PGANs: Personalized Generative Adversarial Networks for ECG Synthesis to Improve Patient-Specific Deep ECG Classification. Proc. Conf. AAAI Artif. Intell. 2019, 33, 557–564. [Google Scholar] [CrossRef]
- Zhai, X.; Tin, C. Automated ECG Classification Using Dual Heartbeat Coupling Based on Convolutional Neural Network. IEEE Access 2018, 6, 27465–27472. [Google Scholar] [CrossRef]
- Übeyli, E.D. Combining recurrent neural networks with eigenvector methods for classification of ECG beats. Digit. Signal Process. 2009, 19, 320–329. [Google Scholar] [CrossRef]
- Zubair, M.; Kim, J.; Yoon, C. An Automated ECG Beat Classification System Using Convolutional Neural Networks. In Proceedings of the 2016 6th International Conference on IT Convergence and Security (ICITCS), Prague, Czech Republic, 26 September 2016; pp. 1–5. [Google Scholar]
- Acharya, U.R.; Fujita, H.; Oh, S.L.; Hagiwara, Y.; Tan, J.H.; Adam, M. Application of deep convolutional neural network for automated detection of myocardial infarction using ECG signals. Inf. Sci. 2017, 415–416, 190–198. [Google Scholar] [CrossRef]
- Kiranyaz, S.; Ince, T.; Gabbouj, M. Real-Time Patient-Specific ECG Classification by 1-D Convolutional Neural Networks. IEEE Trans. Biomed. Eng. 2015, 63, 664–675. [Google Scholar] [CrossRef]
- Rajpurkar, P.; Hannun, A.Y.; Haghpanahi, M.; Bourn, C.; Ng, A.Y. Cardiologist-level arrhythmia detection with convolutional neural networks. In Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017. [Google Scholar]
- Fan, X.; Yao, Q.; Cai, Y.; Miao, F.; Sun, F.; Li, Y. Multiscaled Fusion of Deep Convolutional Neural Networks for Screening Atrial Fibrillation from Single Lead Short ECG Recordings. IEEE J. Biomed. Health Inform. 2018, 22, 1744–1753. [Google Scholar] [CrossRef] [PubMed]
- Liu, W.; Wang, F.; Huang, Q.; Chang, S.; Wang, H.; He, J. MFB-CBRNN: A Hybrid Network for MI Detection Using 12-Lead ECGs. IEEE J. Biomed. Health Inform. 2020, 24, 503–514. [Google Scholar] [CrossRef] [PubMed]
- Oh, S.L.; Eddie, Y.K.; Tan, R.S.; Acharya, U.R. Automated diagnosis of arrhythmia using combination of CNN and LSTM techniques with variable length heart beats. Comput. Biol. Med. 2018, 102. [Google Scholar] [CrossRef] [PubMed]
- Kim, M.G. A Study on User Recognition System Based on Ensemble Convolutional Neural Networks Using Synthetic Electrocar-Diogram Generation. Doctoral Dissertation, Department of Control and Instrumentation Engineering, Chosun University, Gwangju, Republic of Korea, 2019. [Google Scholar]
- Xiao, Y.; Wu, J.; Lin, Z.; Zhao, X. A deep learning-based multi-model ensemble method for cancer prediction. Comput. Methods Programs Biomed. 2018, 153, 1–9. [Google Scholar] [CrossRef] [PubMed]
- Kim, J.-H.; Lee, S.; Park, K.-H. Stepwise Detection of the QRS Complex in the ECG Signal. J. Korean Inst. Commun. Inf. Sci. 2016, 41, 244–253. [Google Scholar] [CrossRef]
- Slama, A.B.; Lentka, L.; Mouelhi, A.; Diouani, M.F.; Sayadi, M.; Smulko, J. Application of statistical features and multilayer neural network to automatic diagnosis of arrhythmia by ECG signals. Metrol. Meas. Syst. 2018, 25. [Google Scholar] [CrossRef]
- Kim, M.-G.; Pan, S.B. Deep Learning Based on 1-D Ensemble Networks Using ECG for Real-Time User Recognition. IEEE Trans. Ind. Inform. 2019, 15, 5656–5663. [Google Scholar] [CrossRef]
- Apaydin, T.; Ferhatosmanoglu, H. Access Structures for Angular Similarity Queries. IEEE Trans. Knowl. Data Eng. 2006, 18, 1512–1525. [Google Scholar] [CrossRef]
- Yoo, J.-C.; Han, T.H. Fast Normalized Cross-Correlation. Circuits Syst. Signal Process. 2009, 28, 819–843. [Google Scholar] [CrossRef]
- Jun, T.J.; Nguyen, H.M.; Kang, D.Y.; Kim, D.H.; Kim, D.Y.; Kim, Y.H. ECG arrhythmia classification using a 2-D convolutional neural network. In Proceedings of the IEEE Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 18–23 June 2018. [Google Scholar]
- Abdeldayem, S.S.; Bourlai, T. ECG-based Human Authentication using High-level Spectro-temporal Signal Features. In Proceedings of the 2018 IEEE International Conference on Big Data, Seattle, WA, USA, 10–13 December 2018; pp. 4984–4993. [Google Scholar]
- Zhang, Q.; Zhou, D.; Zeng, X. HeartID: A Multiresolution Convolutional Neural Network for ECG-Based Biometric Human Identification in Smart Health Applications. IEEE Access 2017, 5, 11805–11816. [Google Scholar] [CrossRef]
- Sidek, K.A.; Khalil, I.; Jelinek, H.F. ECG Biometric with Abnormal Cardiac Conditions in Remote Monitoring System. IEEE Trans. Syst. Man Cybern. Syst. 2014, 44, 1498–1509. [Google Scholar] [CrossRef]
Generator | Critic | ||||
---|---|---|---|---|---|
Layers | Act/Norm | Output Shape | Layers | Act/Norm | Output Shape |
Latent | - | Input | - | ||
Linear | LReLU | Conv 1 | LReLU | ||
Upsample | - | Conv 9 | LReLU | ||
Conv 9 | LReLU/PN | Conv 9 | LReLU | ||
Conv 9 | LReLU/PN | Downsample | - | ||
Upsample | - | Conv 9 | LReLU | ||
Conv 9 | LReLU/PN | Conv 9 | LReLU | ||
Conv 9 | LReLU/PN | Downsample | - | 92 | |
Upsample | - | Conv 9 | LReLU | 92 | |
Conv 9 | LReLU/PN | Conv 9 | LReLU | 92 | |
Conv 9 | LReLU/PN | Downsample | - | ||
Upsample | - | Conv 9 | LReLU | 96 | |
Conv 9 | LReLU/PN | Conv 9 | LReLU | ||
Conv 9 | LReLU/PN | 192 | Downsample | - | |
Upsample | - | Conv 9 | LReLU | ||
Conv 9 | LReLU/PN | Conv 9 | LReLU | ||
Conv 9 | LReLU/PN | 384 | Downsample | - | |
Upsample | - | Conv 9 | - | ||
Conv 9 | LReLU/PN | Conv 9 | LReLU | ||
Conv 9 | LReLU/PN | Downsample | - | 2 | |
Conv 1 | - | Linear | - |
Class | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
Result | 0.998 | 0.996 | 0.996 | 0.995 | 0.996 | 0.991 | 0.99 | 0.997 | 0.994 |
Class | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 |
Result | 0.99 | 0.989 | 0.988 | 0.991 | 0.979 | 0.991 | 0.974 | 0.995 | 0.994 |
Class | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 |
Result | 0.993 | 0.991 | 0.992 | 0.99 | 0.989 | 0.994 | 0.992 | 0.994 | 0.98 |
Class | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 |
Result | 0.995 | 0.996 | 0.993 | 0.99 | 0.979 | 0.996 | 0.989 | 0.996 | 0.995 |
Class | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 |
Result | 0.993 | 0.987 | 0.99 | 0.989 | 0.996 | 0.995 | 0.995 | 0.975 | 0.987 |
Class | 46 | 47 | 48 | 49 | 50 | 51 | 52 | 53 | 54 |
Result | 0.988 | 0.984 | 0.93 | 0.994 | 0.994 | 0.994 | 0.985 | 0.986 | 0.997 |
Class | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 |
Result | 0.995 | 0.998 | 0.978 | 0.998 | 0.996 | 0.98 | 0.996 | 0.995 | 0.993 |
Class | 64 | 65 | 66 | 67 | 68 | 69 | 70 | 71 | 72 |
Result | 0.982 | 0.996 | 0.996 | 0.992 | 0.995 | 0.993 | 0.988 | 0.995 | 0.995 |
Class | 73 | 74 | 75 | 76 | 77 | 78 | 79 | 80 | 81 |
Result | 0.992 | 0.995 | 0.994 | 0.99 | 0.992 | 0.987 | 0.994 | 0.996 | 0.989 |
Class | 82 | 83 | 84 | 85 | 86 | 87 | 88 | 89 | AVG |
Result | 0.994 | 0.996 | 0.996 | 0.989 | 0.995 | 0.992 | 0.995 | 0.998 | 0.991 |
Class | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
Result | 0.264 | 0.352 | 0.166 | 0.247 | 0.212 | 0.31 | 0.178 | 0.242 | 0.36 |
Class | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 |
Result | 0.136 | 0.254 | 0.223 | 0.176 | 0.198 | 0.261 | 0.36 | 0.163 | 0.227 |
Class | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 |
Result | 0.325 | 0.298 | 0.271 | 0.35 | 0.221 | 0.259 | 0.197 | 0.362 | 0.296 |
Class | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 |
Result | 0.27 | 0.344 | 0.168 | 0.19 | 0.224 | 0.314 | 0.356 | 0.172 | 0.34 |
Class | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 |
Result | 0.198 | 0.21 | 0.332 | 0.281 | 0.26 | 0.314 | 0.171 | 0.364 | 0.218 |
Class | 46 | 47 | 48 | 49 | 50 | 51 | 52 | 53 | 54 |
Result | 0.323 | 0.167 | 0.229 | 0.284 | 0.341 | 0.217 | 0.29 | 0.192 | 0.23 |
Class | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 |
Result | 0.276 | 0.139 | 0.224 | 0.261 | 0.25 | 0.317 | 0.163 | 0.183 | 0.266 |
Class | 64 | 65 | 66 | 67 | 68 | 69 | 70 | 71 | 72 |
Result | 0.189 | 0.16 | 0.335 | 0.238 | 0.162 | 0.281 | 0.318 | 0.29 | 0.314 |
Class | 73 | 74 | 75 | 76 | 77 | 78 | 79 | 80 | 81 |
Result | 0.195 | 0.263 | 0.225 | 0.242 | 0.22 | 0.196 | 0.329 | 0.214 | 0.31 |
Class | 82 | 83 | 84 | 85 | 86 | 87 | 88 | 89 | AVG |
Result | 0.324 | 0.161 | 0.243 | 0.324 | 0.212 | 0.194 | 0.22 | 0.139 | 0.25 |
Test Data Set | |||||
---|---|---|---|---|---|
Real1~5 | Real1 | Real2 | Real3 | Real4 | Real5 |
Real1~4+Synthetic1 | Real1 | Real2 | Real3 | Real4 | Synthetic1 |
Real1~4+Real4 | Real1 | Real2 | Real3 | Real4 | Real4 |
Real1~3+Synthetic1~2 | Real1 | Real2 | Real3 | Synthetic1 | Synthetic2 |
Real1~3+Real3~3 | Real1 | Real2 | Real3 | Real3 | Real3 |
Real1~2+Synthetic1~3 | Real1 | Real2 | Synthetic1 | Synthetic2 | Synthetic3 |
Real1~2+Real2~2 | Real1 | Real2 | Real2 | Real2 | Real2 |
Classifier | Work | Database | Test Set | Accuracy | Specificity | Sensitivity |
---|---|---|---|---|---|---|
1D Ensemble Networks | Kim [22] | MIT-BIH database | 1692 | 99.6% | 0.99 | 0.99 |
2D CNN | Jun et al. [29] | 100,000 | 99% | 0.99 | 0.97 | |
Abdeldayem et al. [30] | 250 | 98.8% | - | - | ||
1D CNN | Zhang et al. [31] | 250 | 91.1% | - | - | |
MLP | Sidek et al. [32] | - | 94.4% | 0.99 | 0.94 | |
RBF | 96.2% | 0.99 | 0.96 | |||
KNN | 97.9% | 0.99 | 0.97 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Kim, M.-G.; Pan, S.B. A Study on User Recognition Using the Generated Synthetic Electrocardiogram Signal. Sensors 2021, 21, 1887. https://doi.org/10.3390/s21051887
Kim M-G, Pan SB. A Study on User Recognition Using the Generated Synthetic Electrocardiogram Signal. Sensors. 2021; 21(5):1887. https://doi.org/10.3390/s21051887
Chicago/Turabian StyleKim, Min-Gu, and Sung Bum Pan. 2021. "A Study on User Recognition Using the Generated Synthetic Electrocardiogram Signal" Sensors 21, no. 5: 1887. https://doi.org/10.3390/s21051887
APA StyleKim, M.-G., & Pan, S. B. (2021). A Study on User Recognition Using the Generated Synthetic Electrocardiogram Signal. Sensors, 21(5), 1887. https://doi.org/10.3390/s21051887