癲癇是一種由大腦內神經元異常放電活動導致的神經系統疾病，由於部分患者無法透過現有藥物或手術療法獲得充分控制，提前預測癲癇發作以利及早採取預防或治療措施顯得尤為關鍵。本研究發展基於腦電圖的癲癇發作預測方法，以協助患者改善病情與生活品質。既有文獻在處理腦波訊號時，多採用共同空間型樣法擷取對數變異數特徵並結合線性分類器進行預測。然而，此類方法存在可改善的空間，首先，共同空間型樣法易受雜訊及小資料集影響而導致過擬合，且變異數特徵難以精準呈現受試者腦波頻率的細微變化；其次，傳統研究多採用固定數量或全數保留的方式來挑選空間濾波器，難以適用於不同受試者。為解決上述課題，本文提出一種結合Tikhonov正則化共同空間型樣法與擠壓激勵塊一維卷積神經網路的預測模型。在特徵擷取階段，首先利用Tikhonov正則化共同空間型樣法轉換腦電圖訊號，以最大化發作間期與前發作期的差異；接著針對轉換後的主成分訊號，計算具生理意義之固定頻帶(如Gamma、Beta和Alpha等)的相對頻譜功率，藉此保留更豐富的頻率資訊並構成二維特徵圖。為使模型能夠學習在時間序列上的變化，將連續時間窗的二維特徵圖整合為三維特徵，並輸入至分類器進行預測。該網路透過擠壓激勵塊，能根據通道重要性自動分配權重強化關鍵特徵並抑制無效或冗餘通道，藉此間接解決了濾波器數量的挑選難題。經Kaggle與CHB-MIT公開資料集驗證，TRCSP-SECNN的平均測試AUC達到0.77；顯著優於傳統共同空間型樣法結合線性判別分析(AUC = 0.6, p = 0.02)及結合卷積神經網路(AUC = 0.66, p = 0.03)的方法。總結而言，本研究驗證結果支持Tikhonov正則化共同空間型樣法在處理小資料集腦電訊號時具備較傳統共同空間型樣法更佳的穩定性。同時，擠壓激勵塊的引入實現了特徵通道的自適應校準，不僅間接克服了以往空間濾波器數量選取不易的限制，更顯著提升了整體預測模型的泛化能力。

Early prediction of epileptic seizures from electroencephalogram (EEG) signals is critical for timely intervention, yet existing common spatial pattern (CSP)-based methods suffer from overfitting on small datasets and rigid spatial filter selection. To address these limitations, this paper presents a seizure prediction framework integrating Tikhonov regularized CSP (TRCSP) with a squeeze-and-excitation 1D convolutional neural network (SE-1DCNN). The TRCSP first transforms EEG signals to maximize the discriminability between interictal and preictal states. From the transformed principal component signals, the relative spectral power is extracted to capture essential frequency domain information. Through the squeeze-and-excitation (SE) block, the network automatically allocates weights based on channel importance, reinforcing key features while suppressing invalid or redundant channels. This adaptively calibrated channel mechanism indirectly resolves the traditional dilemma of manually selecting the number of spatial filters. Extensive evaluation on the Kaggle and CHB-MIT benchmark datasets demonstrates that the proposed TRCSP-SECNN achieves a mean test Area under the curve (AUC) of 0.77, significantly outperforming CSP- Linear discriminant analysis (LDA) (AUC = 0.60, p = 0.02) and CSP- Convolutional neural network (CNN) (AUC = 0.66, p = 0.03). These results underscore the robustness of TRCSP and the enhanced generalization enabled by adaptive feature calibration.

[1]“Epilepsy.” Accessed: Jun. 02, 2026. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/epilepsy
[2]R. S. Fisher et al., “Operational classification of seizure types by the International League Against Epilepsy: Position Paper of the ILAE Commission for Classification and Terminology,” Epilepsia, vol. 58, no. 4, pp. 522–530, 2017.
[3]A. Brodovskaya and J. Kapur, “Circuits generating secondarily generalized seizures,” Epilepsy & Behavior, vol. 101, p. 106474, Dec. 2019.
[4]B. Litt and K. Lehnertz, “Seizure prediction and the preseizure period,” Current Opinion in Neurology, vol. 15, no. 2, p. 173, Apr. 2002.
[5]K. Lehnertz et al., “Seizure prediction by nonlinear EEG analysis,” IEEE Engineering in Medicine and Biology Magazine, vol. 22, no. 1, pp. 57–63, Jan. 2003.
[6]L. Kuhlmann, K. Lehnertz, M. P. Richardson, B. Schelter, and H. P. Zaveri, “Seizure prediction — ready for a new era,” Nat Rev Neurol, vol. 14, no. 10, pp. 618–630, Oct. 2018.
[7]P. N. Banerjee, D. Filippi, and W. Allen Hauser, “The descriptive epidemiology of epilepsy—A review,” Epilepsy Research, vol. 85, no. 1, pp. 31–45, Jul. 2009.
[8]S. B. Dumanis, J. A. French, C. Bernard, G. A. Worrell, and B. E. Fureman, “Seizure Forecasting from Idea to Reality. Outcomes of the My Seizure Gauge Epilepsy Innovation Institute Workshop,” eNeuro, vol. 4, no. 6, Nov. 2017.
[9]S. M. Usman, S. Khalid, R. Akhtar, Z. Bortolotto, Z. Bashir, and H. Qiu, “Using scalp EEG and intracranial EEG signals for predicting epileptic seizures: Review of available methodologies,” Seizure, vol. 71, pp. 258–269, Oct. 2019.
[10]J. Gotman and P. Gloor, “Automatic recognition and quantification of interictal epileptic activity in the human scalp EEG,” Electroencephalography and Clinical Neurophysiology, vol. 41, no. 5, pp. 513–529, Nov. 1976.
[11]T. Ball, M. Kern, I. Mutschler, A. Aertsen, and A. Schulze-Bonhage, “Signal quality of simultaneously recorded invasive and non-invasive EEG,” NeuroImage, vol. 46, no. 3, pp. 708–716, Jul. 2009.
[12]K. Lehnertz, “Non-linear time series analysis of intracranial EEG recordings in patients with epilepsy — an overview,” International Journal of Psychophysiology, vol. 34, no. 1, pp. 45–52, Oct. 1999.
[13]J. Parvizi and S. Kastner, “Human Intracranial EEG: Promises and Limitations,” Nat Neurosci, vol. 21, no. 4, pp. 474–483, Apr. 2018.
[14]A. K. Shah and S. Mittal, “Invasive electroencephalography monitoring: Indications and presurgical planning,” Annals of Indian Academy of Neurology, vol. 17, no. Suppl 1, p. S89, Mar. 2014.
[15]B. Maimaiti et al., “An Overview of EEG-based Machine Learning Methods in Seizure Prediction and Opportunities for Neurologists in this Field,” Neuroscience, vol. 481, pp. 197–218, Jan. 2022.
[16]P. R. Carney, S. Myers, and J. D. Geyer, “Seizure prediction: Methods,” Epilepsy & Behavior, vol. 22, pp. S94–S101, Dec. 2011.
[17]M. M. Ahsan, S. A. Luna, and Z. Siddique, “Machine-Learning-Based Disease Diagnosis: A Comprehensive Review,” Healthcare, vol. 10, no. 3, Art. no. 3, Mar. 2022.
[18]Z. J. Koles, M. S. Lazar, and S. Z. Zhou, “Spatial patterns underlying population differences in the background EEG,” Brain Topogr, vol. 2, no. 4, pp. 275–284, Jun. 1990.
[19]Z. J. Koles, “The quantitative extraction and topographic mapping of the abnormal components in the clinical EEG,” Electroencephalography and Clinical Neurophysiology, vol. 79, no. 6, pp. 440–447, Dec. 1991.
[20]Z. J. Koles, J. C. Lind, and P. Flor-Henry, “Spatial patterns in the background EEG underlying mental disease in man,” Electroencephalography and Clinical Neurophysiology, vol. 91, no. 5, pp. 319–328, Nov. 1994.
[21]F. Lotte and C. Guan, “Regularizing Common Spatial Patterns to Improve BCI Designs: Unified Theory and New Algorithms,” IEEE Transactions on Biomedical Engineering, vol. 58, no. 2, pp. 355–362, Feb. 2011.
[22]K. K. Ang, Z. Y. Chin, H. Zhang, and C. Guan, “Filter Bank Common Spatial Pattern (FBCSP) in Brain-Computer Interface,” in 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), Jun. 2008, pp. 2390–2397.
[23]T. N. Alotaiby, S. A. Alshebeili, F. M. Alotaibi, and S. R. Alrshoud, “Epileptic Seizure Prediction Using CSP and LDA for Scalp EEG Signals,” Computational Intelligence and Neuroscience, vol. 2017, p. e1240323, Oct. 2017.
[24]Y. Zhang, Y. Guo, P. Yang, W. Chen, and B. Lo, “Epilepsy Seizure Prediction on EEG Using Common Spatial Pattern and Convolutional Neural Network,” IEEE Journal of Biomedical and Health Informatics, vol. 24, no. 2, pp. 465–474, Feb. 2020.
[25]G. Zheng et al., “Seizure prediction model based on method of common spatial patterns and support vector machine,” in 2012 IEEE International Conference on Information Science and Technology, Mar. 2012, pp. 29–34.
[26]郭劭頎, “共同空間型樣法與延伸之數據驅動頻帶應用於癲癇預測,” 碩士論文, 國立成功大學機械工程學系, 台灣台南市, 2024.
[27]H. Lu, H.-L. Eng, C. Guan, K. N. Plataniotis, and A. N. Venetsanopoulos, “Regularized Common Spatial Pattern With Aggregation for EEG Classification in Small-Sample Setting,” IEEE Transactions on Biomedical Engineering, vol. 57, no. 12, pp. 2936–2946, Feb. 2010.
[28]H. Kang, Y. Nam, and S. Choi, “Composite Common Spatial Pattern for Subject-to-Subject Transfer,” IEEE Signal Processing Letters, vol. 16, no. 8, pp. 683–686, Aug. 2009.
[29]J. Zhang, X. Wang, B. Xu, Y. Wu, X. Lou, and X. Shen, “An overview of methods of left and right foot motor imagery based on Tikhonov regularisation common spatial pattern,” Medical and Biological Engineering and Computing, vol. 61, no. 5, pp. 1047–1056, May 2023.
[30]K. Rasheed et al., “Machine Learning for Predicting Epileptic Seizures Using EEG Signals: A Review,” IEEE Reviews in Biomedical Engineering, vol. 14, pp. 139–155, 2021.
[31]P. Mirowski, D. Madhavan, Y. LeCun, and R. Kuzniecky, “Classification of patterns of EEG synchronization for seizure prediction,” Clinical Neurophysiology, vol. 120, no. 11, pp. 1927–1940, Nov. 2009.
[32]蔡明倫, “定向轉移函數基於卷積神經網路與 L1 正則化線性迴歸之癲癇預測,” 碩士論文, 國立成功大學機械工程學系, 台灣台南市, 2023.
[33]B. Blankertz, R. Tomioka, S. Lemm, M. Kawanabe, and K. Muller, “Optimizing Spatial filters for Robust EEG Single-Trial Analysis,” IEEE Signal Processing Magazine, vol. 25, no. 1, pp. 41–56, 2008.
[34]J. Hu, L. Shen, and G. Sun, “Squeeze-and-Excitation Networks,” presented at the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 7132–7141. Accessed: Jan. 18, 2026. [Online]. Available: https://openaccess.thecvf.com/content_cvpr_2018/html/Hu_Squeeze-and-Excitation_Networks_CVPR_2018_paper.html
[35]B. H. Brinkmann et al., “Crowdsourcing reproducible seizure forecasting in human and canine epilepsy,” Brain, vol. 139, no. 6, pp. 1713–1722, Jun. 2016.
[36]S. Wong et al., “EEG datasets for seizure detection and prediction— A review,” Epilepsia Open, vol. 8, no. 2, pp. 252–267, 2023.
[37]A. H. Shoeb, “Application of machine learning to epileptic seizure onset detection and treatment,” Thesis, Massachusetts Institute of Technology, 2009. Accessed: May 29, 2025. [Online]. Available: https://dspace.mit.edu/handle/1721.1/54669
[38]S.-H. Park, D. Lee, and S.-G. Lee, “Filter Bank Regularized Common Spatial Pattern Ensemble for Small Sample Motor Imagery Classification,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 26, no. 2, pp. 498–505, Feb. 2018.
[39]P.-N. Yu, C. Y. Liu, C. N. Heck, T. W. Berger, and D. Song, “A sparse multiscale nonlinear autoregressive model for seizure prediction,” J. Neural Eng., vol. 18, no. 2, p. 026012, Feb. 2021.
[40]F. Lotte, “Signal Processing Approaches to Minimize or Suppress Calibration Time in Oscillatory Activity-Based Brain–Computer Interfaces,” Proceedings of the IEEE, vol. 103, no. 6, pp. 871–890, Jun. 2015.
[41]K. Fukunaga, “Introduction to Statistical Pattern Recognition Second Edition”.
[42]H. Ramoser, J. Muller-Gerking, and G. Pfurtscheller, “Optimal spatial filtering of single trial EEG during imagined hand movement,” IEEE Trans. Rehab. Eng., vol. 8, no. 4, pp. 441–446, Dec. 2000.
[43]H. Wang, Q. Tang, and W. Zheng, “L1-Norm-Based Common Spatial Patterns,” IEEE Transactions on Biomedical Engineering, vol. 59, no. 3, pp. 653–662, Mar. 2012.
[44]M. Bandarabadi, C. A. Teixeira, J. Rasekhi, and A. Dourado, “Epileptic seizure prediction using relative spectral power features,” Clinical Neurophysiology, vol. 126, no. 2, pp. 237–248, Feb. 2015.
[45]F. Sohil, M. U. Sohali, and J. Shabbir, “An introduction to statistical learning with applications in R: by Gareth James, Daniela Witten, Trevor Hastie, and Robert Tibshirani, New York, Springer Science and Business Media, 2013, $41.98, eISBN: 978-1-4614-7137-7,” Statistical Theory and Related Fields, vol. 6, no. 1, pp. 87–87, Jan. 2022.
[46]Z. Ma, K. Wang, M. Xu, W. Yi, F. Xu, and D. Ming, “Transformed common spatial pattern for motor imagery-based brain-computer interfaces,” Front. Neurosci., vol. 17, Mar. 2023.
[47]G. Dharani. Y, N. G. Nair, P. Satpathy, and J. Christopher, “Covariate Shift: A Review and Analysis on Classifiers,” in 2019 Global Conference for Advancement in Technology (GCAT), Oct. 2019, pp. 1–6.
[48]Q. Novi, C. Guan, T. H. Dat, and P. Xue, “Sub-band Common Spatial Pattern (SBCSP) for Brain-Computer Interface,” in 2007 3rd International IEEE/EMBS Conference on Neural Engineering, May 2007, pp. 204–207.
[49]K. K. Ang, Z. Y. Chin, C. Wang, C. Guan, and H. Zhang, “Filter Bank Common Spatial Pattern Algorithm on BCI Competition IV Datasets 2a and 2b,” Front. Neurosci., vol. 6, Mar. 2012.
[50]X. Zhang, Q. She, Y. Chen, W. Kong, and C. Mei, “Sub-band target alignment common spatial pattern in brain-computer interface,” Computer Methods and Programs in Biomedicine, vol. 207, p. 106150, Aug. 2021.