adb736ed-3833-481a-b434-3623d3f44e6920220107051041948naun:naunmdt@crossref.orgMDT DepositInternational Journal of Circuits, Systems and Signal Processing1998-446410.46300/9106http://www.naun.org/cms.action?id=30291720221720221610.46300/9106.2022.16https://npublications.com/journals/circuitssystemssignal/2022.phpZengHui(Tianjin Key Laboratory of Bioelectromagnetic Technology and Intelligent Health, Hebei University of Technology, Tianjin 300130, China)LiYing(State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology, Tianjin 300130, China)WangLingyue(State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology, Tianjin 300130, China)YinNing(State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology, Tianjin 300130, China)YangShuo(State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology, Tianjin 300130, China)Electroencephalography (EEG) inverse problem is a typical inverse problem, in which the electrical activity within the brain is reconstructed based on EEG data collected from the scalp electrodes. In this paper, the four-layer concentric head model is used for simulation firstly, four deep neural network models including a multilayer perceptron (MLP) model and three convolutional neural networks (CNNs) are adopted to solve EEG inverse problem based on equal current dipole (ECD) model. In the simulations, 100,000 samples are generated randomly, of which 60% are used for network training and 20% are used for cross-validation. Eventually, the generalization performance of the model using the optimal function is measured by the errors in the rest 20% testing set. The experimental results show that the absolute error, relative error, mean positioning error and standard deviation of the four models are extremely low. The CNN with 6 convolutional layers and 3 pooling layers (CNN-3) is the best model. Its absolute error is about 0.015, its relative error is about 0.005, and its dipole position error is 0.040±0.029 cm. Furthermore, we use CNN-3 for source localization of the real EEG data in Working Memory. The results are in accord with physiological experience. The deep neural network method in our study needs fewer calculation parameters, takes less time, and has better positioning results.17202217202213214017https://npublications.com/journals/circuitssystemssignal/2022/a342005-017(2022).pdf10.46300/9106.2022.16.17https://npublications.com/journals/circuitssystemssignal/2022/a342005-017(2022).pdfC.M. Michel, B. He, "EEG source localization," Handbook of Clinical Neurology, vol. 160, pp. 85-101 2019. 10.1016/j.bspc.2014.01.009M.A. Jatoi, N. Kamel, A.S. Malik, I. Faye, T. Begum, "A survey of methods used for source localization using EEG signals," Biomedical Signal Processing and Control, vol. 11, pp. 42-52 2014. 10.1155/2009/656092K. Wendel, O. Vaisanen, J. Malmivuo, N.G. Gencer, B. Vanrumste, P.J. Durka, et al., "EEG/MEG source imaging: methods, challenges, and open issues," Computational Intelligence and Neuroscience, vol. 2009, pp. 1-12 2009. 10.1109/iembs.2004.1403564Y. Li, H. Li, R. He, L. Rao, Q. Wu, G. Xu, et al., "EEG source localization using differential evolution method," International Conference of the IEEE Engineering in Medicine and Biology Society, vol. 1, pp. 1903-1906 2004. 10.1016/j.clinph.2004.06.001C.M. Michel, M.M. Murray, G. Lantz, S.L. Gonzalez, L. Spinelli, R.G. De Peralta, "EEG source imaging," Clinical Neurophysiology, vol. 115, pp. 2195-2222 2004. 10.1016/j.neucom.2013.04.010D. Escalonavargas, D. Gutierrez, I. Lopezarevalo, "Performance of different metaheuristics in EEG source localization compared to the Cramér–Rao bound," Neurocomputing, vol. 120, pp. 597-609 2013. L.C. Jiao, S.Y. Yang, F. Liu, S.G. Wang, Z.X. Feng, "Seventy years beyond neural networks: retrospect and prospect," Chinese Journal of Computers, vol. 39, pp. 1697-1716 2016. 10.1088/0031-9155/45/4/314G. Van Hoey, J. De Clercq, B. Vanrumste, R.V. De Walle, I. Lemahieu, M. Dhave, et al., "EEG dipole source localization using artificial neural networks," Physics in Medicine Biology, vol. 45, pp. 997-1011 2000. 10.1163/156855700750265468K. Kamijo, T. Kiyuna, Y. Takaki, A. Kenmochi, T. Tanigawa, T. Yamazaki, "Integrated approach of an artificial neural network and numerical analysis to multiple equivalent current dipole source localization," Frontiers of Medical Biological Engineering, vol. 10, pp. 285-301 2000. 10.1179/016164101101198848R.J. Sclabassi, M. Sonmez, M. Sun, "EEG source localization: a neural network approach," Neurological Research, vol. 23, pp. 457-464 2001. 10.1109/iembs.2006.259217L.S. Qing Wu, Youxi Wu, Guizhi Xu, Ying Li, Weili Yan, "Combining nonlinear dimensionality reduction with wavelet network to solve EEG inverse problem," International Conference of the IEEE Engineering in Medicine and Biology Society, vol. 2006, pp. 4245-4248 2006. 10.1109/icmlc.2009.5212441Y.-H.W. Jian-Wei Li, Gui-Long Zong, Qing Wu, "Support vector machine method using in EEG signals study of epileptic spike," International Conference on Machine Learning and Cybernetics vol. 2, pp. 1241-1245 2009. 10.1109/tnnls.2013.2293637M. Bianchini, F. Scarselli, "On the complexity of neural network classifiers: a comparison between shallow and deep architectures," IEEE Transactions on Neural Networks, vol. 25, pp. 1553-1565 2014. 10.1109/cvpr.2016.90K. He, X. Zhang, S. Ren, J. Sun, "Deep residual learning for image recognition," 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770-778 2016. I. Goodfellow, Y. Bengio, A. Courville, "Deep learning," MA: The MIT Press, pp. 2016. 10.20965/jrm.2017.p0037N. Yalta, K. Nakadai, T. Ogata, "Sound source localization using deep learning models," Journal of Robotics and Mechatronics, vol. 29, pp. 37-48 2017. 10.1109/tip.2017.2713099K.H. Jin, M.T. Mccann, E. Froustey, M. Unser, "Deep convolutional neural network for inverse problems in imaging," IEEE Transactions on Image Processing, vol. 26, pp. 4509-4522 2017. 10.1109/msp.2017.2760358A. Lucas, M. Iliadis, R. Molina, A.K. Katsaggelos, "Using deep neural networks for inverse problems in imaging: beyond analytical methods," IEEE Signal Processing Magazine, vol. 35, pp. 20-36 2018. 10.1088/1361-6420/aa9581J. Adler, O. Oktem, "Solving ill-posed inverse problems using iterative deep neural networks," Inverse Problems, vol. 33, pp. 124007 2017. 10.1109/tbme.2017.2756869T. Yang, L. Yu, Q. Jin, L. Wu, B. He, "Localization of origins of premature ventricular contraction by means of convolutional neural network from 12-lead ECG," IEEE Transactions on Biomedical Engineering, vol. 65, pp. 1662-1671 2018. 10.23919/jcc.2019.07.011S. Cui, L. Duan, B. Gong, Y. Qiao, F. Xu, J. Chen, et al., "EEG source localization using spatio-temporal neural network," China Communications, vol. 16, pp. 131-143 2019. 10.1007/s13534-020-00149-6S.K. Veeramalla, V.K.H.R. Talari, "Multiple dipole source localization of EEG measurements using particle filter with partial stratified resampling," Biomedical Engineering Letters, vol. 2020, pp. 1-11 2020. 10.1109/tsp.2019.2904921G. Wang, G.B. Giannakis, J. Chen, "Learning ReLU networks on linearly separable data: algorithm, optimality, and generalization," IEEE Transactions on Signal Processing, vol. 67, pp. 2357-2370 2019. D.P. Kingma, J. Ba, "Adam: a method for stochastic optimization," International Conference on Learning Representations 2015. 10.1109/tpami.2013.19C. Bo, P. G, S. G, "Deep learning with hierarchical convolutional factor analysis," IEEE Trans Pattern Anal Mach Intell, vol. 35, pp. 1887-1901 2013. 10.1109/10.649996M. Sun, "An efficient algorithm for computing multishell spherical volume conductor models in EEG dipole source localization," IEEE Transactions on Biomedical Engineering, vol. 44, pp. 1243-1252 1997. A. Baddeley, "Working memory: theories, models, and controversies," Annual Review of Psychology, vol. 63, pp. 1-29 2006. 10.1162/jocn.2008.21028M.E. Thomason, E. Race, B. Burrows, S. Whitfield-Gabrieli, G.H. Glover, J.D.E. Gabrieli, "Development of spatial and verbal working memory capacity in the human brain," Journal of Cognitive Neuroscience vol. 21, pp. 316-332 2009. 10.1162/jocn.2009.21210L. Wang, X. Liu, K.G. Guise, R.T. Knight, J.J.J.o.C.N. Fan, "Effective connectivity of the fronto-parietal network during attentional control," Journal of Cognitive Neuroscience, vol. 22, pp. 543-553 2009. 10.1016/j.neuroimage.2005.09.060C. Elfgren, D.V. Westen, U. Passant, E.M. Larsson, P. Mannfolk, P.J.N. Fransson, "fMRI activity in the medial temporal lobe during famous face processing," NeuroImage, vol. 30, pp. 609-616 2006.