Seizure Prediction Based on Transformer Using Scalp Electroencephalogram
Source abstract
Abstract
Epilepsy is a chronic and recurrent brain dysfunction disease. An acute epileptic attack will interfere with a patient’s normal behavior and consciousness, having a great impact on their life. The purpose of this study was to design a seizure prediction model to improve the quality of patients’ lives and assist doctors in making diagnostic decisions. This paper presents a transformer-based seizure prediction model. Firstly, the time-frequency characteristics of electroencephalogram (EEG) signals were extracted by short-time Fourier transform (STFT). Secondly, a three transformer tower model was used to fuse and classify the features of the EEG signals. Finally, when combined with the attention mechanism of transformer networks, the EEG signal was processed as a whole, which solves the problem of length limitations in deep learning models. Experiments were conducted with a Children’s Hospital Boston and the Massachusetts Institute of Technology database to evaluate the performance of the model. The experimental results show that, compared with previous EEG classification models, our model can enhance the ability to use time, frequency, and channel information from EEG signals to improve the accuracy of seizure prediction.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: EEG and Brain-Computer Interfaces · ECG Monitoring and Analysis · Functional Brain Connectivity Studies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China
References (49)
Show source-provided references
1. WHO (2022, February 10). WHO Report. Available online: http://www.who.int/mediacentre/factsheets/fs999/en/.
2. Inspection of EEG signals for efficient seizure prediction10.1016/j.apacoust.2020.107327 ↗
3. Automatic recognition of eplieptic seizures in the EEG10.1016/0013-4694(82)90038-4 ↗
4. Artificial neural network based epileptic detection using time-domain and frequency-domain features10.1007/s10916-005-6133-1 ↗
5. The seizure prediction characteristic: A general framework to assess and compare seizure prediction methods10.1016/S1525-5050(03)00105-7 ↗
6. Classification of ictal and seizure-free EEG signals using fractional linear prediction10.1016/j.bspc.2013.08.006 ↗
7. Automatic seizure detection by modified line length and Mahalanobis distance function10.1016/j.bspc.2018.05.004 ↗
8. Jahankhani, P., Revett, K., and Kodogiannis, V. (April, January 1). Data mining an EEG dataset with an emphasis on dimensionality reduction. Proceedings of the 2007 IEEE Symposium on Computational Intelligence and Data Mining, Honolulu, HI, USA.10.1109/CIDM.2007.368903 ↗
9. Patient-specific seizure detection in long-term EEG using wavelet decomposition10.1016/j.bspc.2018.07.006 ↗
10. Pérez-Elvira, R., Oltra-Cucarella, J., Carrobles, J.A., Teodoru, M., Bacila, C., and Neamtu, B. (2021). Individual alpha peak frequency, an important biomarker for live z-score training neurofeedback in adolescents with learning disabilities. Brain Sci., 11.10.3390/brainsci11020167 ↗
11. Spectral information of EEG signals with respect to epilepsy classification10.1186/s13634-019-0606-8 ↗
12. Convolutional neural networks for seizure prediction using intracranial and scalp electroencephalogram10.1016/j.neunet.2018.04.018 ↗
13. Mixed-band wavelet-chaos-neural network methodology for epilepsy and epileptic seizure detection10.1109/TBME.2007.891945 ↗
14. Real-time epileptic seizure detection using EEG10.1109/TNSRE.2017.2697920 ↗
15. Patient non-specific algorithm for seizures detection in scalp EEG10.1016/j.compbiomed.2016.02.016 ↗
16. Detection of seizure event and its onset/offset using orthonormal triadic wavelet based features10.1016/j.irbm.2018.12.002 ↗
17. Distinguishing Epileptiform Discharges From Normal Electroencephalograms Using Scale-Dependent Lyapunov Exponent10.3389/fbioe.2020.01006 ↗
18. A novel Machine Learning approach for epilepsy diagnosis using EEG signals based on Correlation Dimension10.1016/j.ifacol.2021.11.018 ↗
19. Classification of epileptiform EEG using a hybrid system based on decision tree classifier and fast Fourier transform
20. Automatic detection of epileptic seizures using permutation entropy, Tsallis entropy and Kolmogorov complexity10.1166/jmihi.2016.1710 ↗
21. Detection of epileptic electroencephalogram based on permutation entropy and support vector machines10.1016/j.eswa.2011.07.008 ↗
22. Li, K., Zhang, X., and Du, Y. (November, January 30). A SVM based classification of EEG for predicting the movement intent of human body. Proceedings of the 2013 10th International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Jeju, Korea.10.1109/URAI.2013.6677297 ↗
23. Online seizure prediction using an adaptive learning approach10.1109/TKDE.2013.151 ↗
24. Wang, G., Deng, Z., and Choi, K.-S. (2015). Detection of epileptic seizures in EEG signals with rule-based interpretation by random forest approach. International Conference on Intelligent Computing, Springer.10.1007/978-3-319-22053-6_78 ↗
25. Optimized deep neural network architecture for robust detection of epileptic seizures using EEG signals10.1016/j.clinph.2018.10.010 ↗
26. Scalp EEG classification using deep Bi-LSTM network for seizure detection10.1016/j.compbiomed.2020.103919 ↗
27. Automatic epileptic EEG detection using convolutional neural network with improvements in time-domain10.1016/j.bspc.2019.04.028 ↗
28. Epileptic seizure forecasting with generative adversarial networks10.1109/ACCESS.2019.2944691 ↗
29. A long short-term memory deep learning network for the prediction of epileptic seizures using EEG signals10.1016/j.compbiomed.2018.05.019 ↗
30. Ra, J.S., Li, T., and Li, Y. (2021). A Novel Permutation Entropy-Based EEG Channel Selection for Improving Epileptic Seizure Prediction. Sensors, 21.10.3390/s21237972 ↗
31. Prediction of epileptic seizures based on multivariate multiscale modified-distribution entropy10.7717/peerj-cs.744 ↗
32. Attention is all you need
33. Tao, Y., Sun, T., Muhamed, A., Genc, S., Jackson, D., Arsanjani, A., Yaddanapudi, S., Li, L., and Kumar, P. (2021, January 1–5). Gated Transformer for Decoding Human Brain EEG Signals. Proceedings of the 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Guadalajara, Mexico.10.1109/EMBC46164.2021.9630210 ↗
34. Shoeb, A.H. (2009). Application of Machine Learning to Epileptic Seizure Onset Detection and Treatment, Massachusetts Institute of Technology.
35. Kovacs, P., Samiee, K., and Gabbouj, M. (2014, January 4–9). On application of rational discrete short time fourier transform in epileptic seizure classification. Proceedings of the 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Florence, Italy.10.1109/ICASSP.2014.6854723 ↗
36. A novel STFT-ranking feature of multi-channel EMG for motion pattern recognition10.1016/j.eswa.2014.11.044 ↗
37. Chen, R., and Parhi, K.K. (2021, January 1–5). Seizure Prediction using Convolutional Neural Networks and Sequence Transformer Networks. Proceedings of the 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Guadalajara, Mexico.10.1109/EMBC46164.2021.9629732 ↗
38. Seizure prediction in scalp EEG based channel attention dual-input convolutional neural network10.1016/j.physa.2021.126376 ↗
39. (2022, April 05). Fourth International Workshop on Seizure Prediction. Available online: http://www.iwsp4.org/patients.htm.
40. Real-time mining of epileptic seizure precursors via nonlinear mapping and dissimilarity features10.1049/iet-spr.2013.0297 ↗
41. Power efficient refined seizure prediction algorithm based on an enhanced benchmarking10.1038/s41598-021-02798-8 ↗
42. Deep Learning of Simultaneous Intracranial and Scalp EEG for Prediction, Detection, and Lateralization of Mesial Temporal Lobe Seizures10.3389/fneur.2021.705119 ↗
43. On the proper selection of preictal period for seizure prediction10.1016/j.yebeh.2015.03.010 ↗
44. Liu, M., Ren, S., Ma, S., Jiao, J., Chen, Y., Wang, Z., and Song, W. (2021). Gated Transformer Networks for Multivariate Time Series Classification. arXiv.
45. Patient-specific epileptic seizure prediction in long-term scalp EEG signal using multivariate statistical process control10.1016/j.irbm.2019.08.004 ↗
46. Xu, Y., Yang, J., Zhao, S., Wu, H., and Sawan, M. (September, January 31). An end-to-end deep learning approach for epileptic seizure prediction. Proceedings of the 2020 2nd IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS), Genova, Italy.10.1109/AICAS48895.2020.9073988 ↗
47. Seizure Prediction Using Multi-View Features and Improved Convolutional Gated Recurrent Network10.1109/ACCESS.2020.3024580 ↗
48. Zhao, S., Yang, J., Xu, Y., and Sawan, M. (2020, January 12–14). Binary single-dimensional convolutional neural network for seizure prediction. Proceedings of the 2020 IEEE International Symposium on Circuits and Systems (ISCAS), Seville, Spain.10.1109/ISCAS45731.2020.9180430 ↗
49. A lightweight solution to epileptic seizure prediction based on EEG synchronization measurement10.1007/s11227-020-03426-4 ↗