Abstract：Nowadays, semi-supervised deep learning model has been successfully applied in epileptic seizure forecasting based on electroencephalogram (EEG), however, there is still room for improvement in EEG preprocessing method and stability of semi-supervised model. This paper proposed an improved solution which combined continuous wavelet transform (CWT) and Wasserstein generative adversarial network based gradient penalty (WGAN-GP). Firstly, CWT was performed on unlabeled EEG to acquire spectrograms, and WGAN-GP was trained using the EEG dataset of specific patients to get a high-performance feature extractor. Then the trained discriminator of WGAN-GP was used as a feature extractor and two fully connected layers were used as classifier. A small amount of spectrogram of CWT of labeled EEG were used to complete the training of classifier model. Finally, the discriminator of WGAN-GP and the fully-connected network constituted a semi-supervised deep learning prediction model, carrying out epileptic seizure forecasting. The proposed semi-supervised patient-specific seizure forecasting method was evaluated by CHB-MIT scalp EEG dataset and compare the performance with present semi-supervised method. The sensitivity, specificity, accuracy and AUC of the proposed method reached 82.69%, 67.48%, 82.08% and 84.03%, respectively, improving the original performance by 14.48%, 34.45%, 7.87% and 11.4%. Compared to the present semi-supervised method, the difference of the prediction performance of CWT-WGAN-GP and current methods is showing significance (P<0.05). The result showed that the combination of CWT and WGAN-GP effectively improved the prediction performance of semi-supervised deep learning model, and played an optimized role of unsupervised feature extraction in the epileptic seizure prediction.
[1] 大熊辉雄. 临床脑电图学:第5版[M]. 北京:清华大学出版社, 2005: 150-205. [2] Acharya UR, Hagiwara Y, Adeli H. Automated seizure prediction[J]. Epilepsy & Behavior, 2018, 88: 251-261. [3] Kuhlmann L, Lehnertz K, Richardson MP, et al. Seizure prediction — ready for a new era[J]. Nature Reviews Neurology, 2018, 14(10): 618-630. [4] Fisher RS, Cross JH, French JA, et al. Operational classification of seizure types by the International League Against Epilepsy: position paper of the ILAE Commission for Classification and Terminology[J]. Epilepsia, 2017, 58(4): 522-530. [5] Kuhlmann L, Karoly P, Freestone DR, et al. Epilepsyecosystem.org: crowd-sourcing reproducible seizure prediction with long-term human intracranial EEG[J]. Brain: A Journal of Neurology, 2018, 141(9): 2619-2630. [6] Rasheed K, Qayyum A, Qadir J, et al. Machine learning for predicting epileptic seizures using EEG signals: a review[J]. IEEE Reviews in Biomedical Engineering, 2021, 14: 139-155. [7] Assi EB, Nguyen DK, Rihana S, et al. Towards accurate prediction of epileptic seizures: a review[J]. Biomedical Signal Processing and Control, 2017, 34: 144-157. [8] Maiwald T, Winterhalder M, Aschenbrenner-Scheibe R, et al. Comparison of three nonlinear seizure prediction methods by means of the seizure prediction characteristic[J]. Physica D: Nonlinear Phenomena, 2004, 194(3-4): 357-368. [9] Winterhalder M, Schelter B, Maiwald T, et al. Spatio-temporal patient-individual assessment of synchronization changes for epileptic seizure prediction[J]. Clinical Neurophysiology, 2006, 117(11): 2399-2413. [10] Parvez MZ, Paul M. Seizure prediction using undulated global and local features[J]. IEEE Transactions on Biomedical Engineering, 2017, 64(1): 208-217. [11] 张瑞,宋江玲,胡文凤. 癫痫脑电的特征提取方法综述[J]. 西北大学学报:自然科学版, 2016, 46(6): 781-788. [12] Park P, Luo Lan, Parhi KK, et al. Seizure prediction with spectral power of EEG using cost-sensitive support vector machines[J]. Epilepsia, 2011, 52(10):1761-1770. [13] Wang Shouyi, Chaovalitwongse WA, Wong S. Online seizure prediction using an adaptive learning approach[J]. IEEE Transactions on Knowledge and Data Engineering, 2013, 25(12): 2854-2866. [14] Sharif B, Jafari AH. Prediction of epileptic seizures from EEG using analysis of ictal rules on Poincaré plane[J]. Computer Methods and Programs in Biomedicine, 2017, 145: 11-22. [15] Truong ND, Nguyen AD, Kuhlmann L, et al. Convolutional neural networks for seizure prediction using intracranial and scalp electroencephalogram[J]. Neural Networks, 2018, 105: 104-111. [16] Tsiouris KM, Pezoulas VC, Zervakis M, et al. A long short-term memory deep learning network for the prediction of epileptic seizures using EEG signals[J]. Computers in Biology and Medicine, 2018, 99: 24-37. [17] Wei Xiaoyan, Zhou Lin, Zhang Zhen, et al. Early prediction of epileptic seizures using a long-term recurrent convolutional network[J]. Journal of Neuroscience Methods, 2019, 327: 108395. [18] Makhzani A, Shlens J, Jaitly N, et al. Adversarial autoencoders[J]. Computer Science, 25 May, 2016[Epub ahead of print]. [19] Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks[J]. Computer Science, 7 Jan, 2016[Epub ahead of print]. [20] Hosseini MP, Pompili D, Elisevich K, et al. Optimized deep learning for EEG big data and seizure prediction BCI via internet of things[J]. IEEE Transactions on Big Data, 2017, 3(4): 392-404. [21] Daoud H, Bayoumi MA. Efficient epileptic seizure prediction based on deep learning[J]. IEEE Transactions on Biomedical Circuits and Systems, 2019, 13(5): 804-813. [22] Truong ND, Kuhlmann L, Bonyadi MR, et al. Epileptic seizure forecasting with generative adversarial networks[J]. IEEE Access, 2019, 7: 143999-144009. [23] 刘建伟,谢浩杰,罗雄麟. 生成对抗网络在各领域应用研究进展[J]. 自动化学报, 2020, 46(12): 2500-2536. [24] 谭宏卫,周林勇,王国栋,等. 生成式对抗网络的不稳定性分析及其处理技术[J].中国科学:信息科学, 2021, 51(4): 602-617. [25] Shoeb AH. Application of machine learning to epileptic seizure onset detection and treatment[D]. Cambridge: Massachusetts Institute of Technology, 2009. [26] Usman SM, Usman M, Fong S. Epileptic seizures prediction using machine learning methods[J]. Computational and Mathematical Methods in Medicine, 2017, 2017: 9074759. [27] Kitano LAS, Sousa MAA, Santos SD, et al. Epileptic seizure prediction from EEG signals using unsupervised learning and a polling-based decision process[C]//27th International Conference on Artificial Neural Networks. Island of Rhodes: ENNS, 2018: 117-126. [28] Khan H, Marcuse L, Fields M, et al. Focal onset seizure prediction using convolutional networks[J]. IEEE Transactions on Biomedical Engineering, 2017, 65(9): 2109-2118. [29] 陈多. 基于小波变换的癫痫脑电信号分析[D]. 南京:东南大学. 2017. [30] Goodfellow IJ, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139-144. [31] Gulrajani I, Ahmed F, Arjovsky M, et al. Improved training of Wasserstein GANs[J]. Advances in Neural Information Processing Systems, 2017, 30: 5769-5779. [32] Truong ND, Kuhlmann L, Bonyadi MR, et al. Supervised learning in automatic channel selection for epileptic seizure detection[J]. Expert Systems with Applications, 2017, 86: 199-207. [33] Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases[J]. Radiology, 1983, 148(3): 839-843. [34] Craik A, He Yongtian, Contreras-Vidal JL. Deep learning for electroencephalogram (EEG) classification tasks: a review[J]. Journal of Neural Engineering, 2019, 16(3): 28. [35] Tsiouris KM, Pezoulas VC, Zervakis M, et al. A long short-term memory deep learning network for the prediction of epileptic seizures using EEG signals[J]. Computers in Biology and Medicine, 2018, 99: 24-37.
