Please wait a minute...
浙江大学学报(工学版)
Journal of ZheJiang University (Engineering Science)  2020, Vol. 54 Issue (11): 2258-2265    DOI: 10.3785/j.issn.1008-973X.2020.11.021
    
Seizure prediction based on pre-ictal period selection of EEG signal
Ya-jing WANG1(),Qun WANG1,*(),Bo-wen LI1,Zhi-wen LIU1,Yuan-yuan PIAO2,Tao YU2
1. School of Information and Electronics, Beijing Institute of Technology, Beijing 100081, China
2. Xuanwu Hospital, Capital Medical University, Beijing 100053, China
Download: HTML     PDF(1202KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

A novel algorithm for seizure prediction based on patient specific manner was proposed to improve the accuracy of epileptic prediction, including feature extraction, pre-ictal period selection, feature selection and channel selection. Time-frequency features and spatial features were extracted from each channel by 2 s windows with 1 s overlap. A continuous 10 min data was selected as a valid positive sample of the pre-ictal period from segment before seizure onset, which achieved the maximum linear separability compared with the inter-ictal period. The effective features were selected by elastic net, then the selected effective features were used to select effective channels in greedy manner. The effective features of effective channels were input into classifier. The algorithm was tested on the scalp electroencephalogram (sEEG) from the MIT Physio database and the database collected in Xuanwu Hospital. The algorithm achieved a recall of 95.76% and a false positive rate of 0.1073 h?1 in MIT database, and a recall of 97.80% and a false positive rate of 0.0453 h?1 in Xuanwu Hospital database. Results show that the algorithm has high sensitivity and low false positive rate.

Key words： seizure prediction      scalp electroencephalogram (sEEG)      patient-specific      feature selection      pre-ictal period selection     
Received: 25 November 2019      Published: 15 December 2020
CLC:  TP 29  
Corresponding Authors: Qun WANG     E-mail: 3220180493@bit.edu.cn;wq99102@bit.edu.cn
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Ya-jing WANG
Qun WANG
Bo-wen LI
Zhi-wen LIU
Yuan-yuan PIAO
Tao YU
Cite this article:

Ya-jing WANG,Qun WANG,Bo-wen LI,Zhi-wen LIU,Yuan-yuan PIAO,Tao YU. Seizure prediction based on pre-ictal period selection of EEG signal. Journal of ZheJiang University (Engineering Science), 2020, 54(11): 2258-2265.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2020.11.021     OR     http://www.zjujournals.com/eng/Y2020/V54/I11/2258


基于脑电信号预发作数据段选取的癫痫发作预测

为了提高癫痫发作预测的准确性，提出针对病患个体进行癫痫发作预测的算法，包括特征提取、预发作数据段选取、特征挑选与导联挑选. 算法采用半重叠的2 s窗对每个导联分别提取时频特征和空域特征. 从发作前期选择与发作间期相比具有最大线性可分性的连续10 min数据作为预发作数据段的有效正样本. 算法通过弹性网进行特征挑选，再基于所选特征通过贪婪算法选择有效导联，将有效导联的有效特征输入到分类器中. 将该算法在MIT公共头皮脑电数据库和宣武医院收集的数据集上进行测试，测试结果为：在MIT数据库上的击中率为95.76%，假阳性率为0.1073 h?1；在宣武医院数据集上的击中率为97.80%，假阳性率为0.0453 h?1. 结果表明，该算法具有较高的击中率和较低的假阳性率.


关键词： 癫痫发作预测,  头皮脑电图（sEEG）,  患者特异性,  特征挑选,  预发作数据段选取 
Fig.1 Data processing flow chart
Fig.2 Matrixes of brain function connectivity network
编号 Ne nh R /% FPR /h?1
1 6 5.6 94.44 0.0430
5 4 4.0 100.00 0.1311
6 10 9.5 95.00 0.1741
7 3 3.0 100.00 0.1717
8 5 4.4 88.00 0.0419
9 4 3.5 87.50 0.1177
10 7 6.5 93.88 0.0884
11 3 3.0 100.00 0.1387
14 6 6.0 100.00 0.1195
16 5 4.4 88.00 0.1814
18 4 3.7 93.75 0.0916
19 2 2.0 100.00 0.0056
20 6 6.0 100.00 0.0487
22 3 3.0 100.00 0.1483
平均 ? ? 95.76 0.1073
Tab.1 Prediction results based on MIT dataset
编号 T Ne nh R /% FPR /h?1
1 20 6 5.5 91.67 0.0412
2 22 6 4.8 80.56 0.0302
3 8 2 2.0 100.00 0.0412
4 16 3 3.0 100.00 0.1161
5 8 2 2.0 100.00 0.0527
6 11 4 4.0 100.00 0.0475
7 18 10 9.7 97.00 0.0280
8 8 2 2.0 100.00 0.0379
9 22 9 9.0 100.00 0.0726
10 19 6 6.0 100.00 0.0632
11 19 5 5.0 100.00 0.0485
12 17 4 4.0 100.00 0.0410
13 7 4 4.0 100.00 0.0465
14 13 4 4.0 100.00 0.0334
15 8 4 4.0 100.00 0.0489
16 16 7 7.0 100.00 0.0105
17 11 5 5.0 100.00 0.0324
18 27 9 8.0 88.89 0.0395
19 8 2 2.0 100.00 0.0289
平均 ? ? ? 97.80 0.0453
Tab.2 Prediction results based on dataset collected from Xuanwu Hospital
Fig.3 Comparison of features between pre-ictal period and 10 minutes before onset
Fig.4 Performance comparison of PPS and no-PPS on Recall
Fig.5 Performance comparison of PPS and no-PPS on FPR
Fig.6 Performance comparison of FCN and no-FCN on Recall
Fig.7 Performance comparison of FCN and no-FCN on FPR
时间 方法 数据来源 发作次数 R /% FPR /h?1
2016 文献[32] MIT10位患者 31 77.00 0.1700
2013 文献[33] MIT10位患者 51 88.20 ?
2017 文献[34] MIT13位患者 45 86.67 0.3670
2017 文献[35] MIT24位患者 170 89.00 0.3900
2017 文献[36] MIT13位患者 64 81.20 0.1600
本研究 MIT17位患者 75 95.76 0.1073
Tab.3 Comparison of different algorithms based on MIT database
[1]   BOER H M D, MULA M, SANDER J W The global burden and stigma of epilepsy[J]. Epilepsy and Behavior, 2008, 12 (4): 540- 546
doi: 10.1016/j.yebeh.2007.12.019
[2]   LEONARDI M, USTUN T B The global burden of epilepsy[J]. Epilepsia, 2010, 43 (6): 21- 25
[3]   AARABI A, FAZEL-REZAI R, AGHAKHANI Y. EEG seizure prediction: measures and challenges [C]// International Conference of the IEEE Engineering in Medicine and Biology Society. 2009: 1864-1867.
[4]   SAAB M E, GOTMAN J A system to detect the onset of epileptic seizures in scalp EEG[J]. Clinical Neurophysiology, 2005, 116 (2): 427- 442
doi: 10.1016/j.clinph.2004.08.004
[5]   TEIXEIRA C A, DIREITO B, BANDARABADI M, et al Epileptic seizure predictors based on computational intelligence techniques: acomparative study with 278 patients[J]. Computer Methods and Programs in Biomedicine, 2014, 114 (3): 324- 336
doi: 10.1016/j.cmpb.2014.02.007
[6]   BANDARABADI M, TEIXEIRA C A, RASEKHI J, et al Epileptic seizure prediction using relative spectral power features[J]. Clinical Neurophysiology Official Journal of the International Federation of Clinical Neurophysiology, 2015, 126 (2): 237- 248
doi: 10.1016/j.clinph.2014.05.022
[7]   YUN P, LAN L, PARHI KESHAB K, et al Seizure prediction with spectral power of EEG using cost-sensitive support vector machines[J]. Epilepsia, 2011, 52 (10): 1761- 1770
doi: 10.1111/j.1528-1167.2011.03138.x
[8]   ALARCON G, BINNIE C D, ELWES R D C, et al Power spectrum and intracranial EEG patterns at seizure onset in partial epilepsy[J]. Electroencephalography and Clinical Neurophysiology, 1995, 94 (5): 326- 337
doi: 10.1016/0013-4694(94)00286-T
[9]   SLAWSKI M Fisher's linear discriminant analysis[J]. Department of Computer Science, 2015, 16 (94): 237- 280
[10]   PIERO P, FRANCOIS D, JEAN G Intracranial electroencephalographic seizure-onset patterns: effect of underlying pathology[J]. Brain, 2014, 137 (1): 183- 196
doi: 10.1093/brain/awt299
[11]   FKREUZ M On the predictability of epileptic seizures[J]. Dkgest of the World Latest Medical Information, 2005, 116 (3): 569- 587
[12]   FLORIAN M, ANDRZEJAK R G, ELGER C E, et al Seizure prediction: the long and winding road[J]. Brain, 2007, 130 (2): 314- 333
[13]   LITT B, ECHAUZ J Prediction of epileptic seizures[J]. Lancet Neurology, 2002, 1 (1): 22- 30
doi: 10.1016/S1474-4422(02)00003-0
[14]   JACOBS J, STABA R, ASANO E, et al High-frequency oscillations (HFOs) in clinical epilepsy[J]. Progress in Neurobiology, 2012, 98 (3): 302- 315
doi: 10.1016/j.pneurobio.2012.03.001
[15]   OUYANG G, LI X, LI Y, et al Application of wavelet-based similarity analysis to epileptic seizures prediction[J]. Computers in Biology and Medicine, 2007, 37 (4): 430- 437
doi: 10.1016/j.compbiomed.2006.08.010
[16]   LUIGI C, ANTONIO M, GUIDO P, et al Real-time epileptic seizure prediction using AR models and support vector machines[J]. IEEE Transactions on Biomedical Engineering, 2010, 57 (5): 1124- 1132
doi: 10.1109/TBME.2009.2038990
[17]   IASEMIDIS L D, SACKELLARES J C, ZAVERI H P, et al Phase space topography and the Lyapunov exponent of electrocorticograms in partial seizures[J]. Brain Topography, 1990, 2 (3): 187- 201
doi: 10.1007/BF01140588
[18]   IASEMIDIS L D Epileptic seizure prediction and control[J]. IEEE Transactions on Biomedical Engineering, 2003, 50 (5): 549- 558
doi: 10.1109/TBME.2003.810705
[19]   SILVA F H L D, WOUTER B, KALITZIN STILIYAN N, et al Dynamical diseases of brain systems: different routes to epileptic seizures[J]. IEEE Transactions on Bio-medical Engineering, 2003, 50 (5): 540- 8
doi: 10.1109/TBME.2003.810703
[20]   NOACHTAR S, RéMI J The role of EEG in epilepsy: acritical review[J]. Epilepsy and Behavior, 2009, 15 (1): 22- 33
doi: 10.1016/j.yebeh.2009.02.035
[21]   UHLHAAS P J, SINGER W Neural synchrony in brain disorders: relevance for cognitive dysfunctions and pathophysiology[J]. Neuron, 2006, 52 (1): 155- 168
doi: 10.1016/j.neuron.2006.09.020
[22]   FLORIAN M, ANDRZEJAK RALPH G, THOMAS K, et al Automated detection of a preseizure state based on a decrease in synchronizationin intracranial electroencephalogram recordings from epilepsy patients[J]. Physical Review E Statistical Nonlinear and Soft Matter Physics, 2003, 67 (1): 021912
[23]   KASPAR S, HOWAN L, ELGER CHRISTIAN E, et al Assessing seizure dynamics by analysing the correlation structure of multichannel intracranial EEG[J]. Brain, 2007, 130 (1): 65- 77
[24]   WILLIAMSON J R, BLISS D W, BROWNE D W, et al Seizure prediction using EEG spatiotemporal correlation structure[J]. Epilepsy and Behavior, 2012, 25 (2): 230- 238
doi: 10.1016/j.yebeh.2012.07.007
[25]   ASSI E B, SAWAN M, NGUYEN D K, et al. A hybrid mRMR-genetic based selection method for the prediction of epileptic seizures [C]// 2015 IEEE Biomedical Circuits and Systems Conference. Atlanta: [s.n.], 2015: 1−4.
[26]   DIREITO B, VENTURA F, TEIXEIRA C, et al. Optimized feature subsets for epileptic seizure prediction studies [C]// International Conference of the IEEE Engineering in Medicine and Biology Society. Boston: [s.n.], 2011: 1636−1639.
[27]   ZHANG Z, PARHI K K Low-complexity seizure prediction from iEEG/sEEG using spectral power and ratios of spectral power[J]. IEEE Transactions on Biomedical Circuits and Systems, 2016, 10 (3): 693- 706
doi: 10.1109/TBCAS.2015.2477264
[28]   BIRJANDTALAB J, POUYAN M B, COGAN D, et al Automated seizure detection using limited-channel EEG and non-linear dimension reduction[J]. Computers in Biology and Medicine, 2017, 82 (C): 49- 58
[29]   GOLDBERGER A L, AMARAL L A, GLASS L, et al PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals[J]. Circulation, 2000, 101 (23): 215- 220
[30]   WENDLING F, CHAUVEL P, BIRABEN A, et al From intracerebral EEG signals to brain connectivity: identification of epileptogenic networks in partial epilepsy[J]. Frontiers in Systems Neuroscience, 2010, 4: 154
[31]   WANG Q, TIAN X, LIU Z. Seizure prediction in eeg records based on spatial-frequency features and preictal period selection [C]// 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). Honolulu: [s.n.], 2018: 5354–5357.
[32]   MYERS M H, PADMANABHA A, HOSSAIN G, et al Seizure prediction and detection via phase and amplitude lock values[J]. Frontiers in Human Neuroscience, 2016, 10: 80
[33]   CONSUL S, MORSHED B I, KOZMA R. Hardware efficient seizure prediction algorithm [J]. SpieNanosensors, Biosensors, and Info-Tech Sensors and Systems, 2013, 8691: 1–10.
[34]   CHU H, CHUNG C K, JEONG W, et al Predicting epileptic seizures from scalp EEG based on attractor state analysis[J]. Computer Methods and Programs in Biomedicine, 2017, 143: 75- 87
[35]   ALOTAIBY T N, ALSHEBEILI S A, ALOTAIBI F M, et al Epileptic seizure prediction using CSP and LDA for scalp EEG signals[J]. Computational Intelligence and Neuroscience, 2017, (6): 1- 11
[1] You-wei WANG,Li-zhou FENG. Improved AdaBoost algorithm using group degree and membership degree based noise detection and dynamic feature selection[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(2): 367-376.
[2] Yun-qing HU,Qing-ying QIU,Xiu YU,Jian-wei WU. Semi-supervised patent text classification method based on improved Tri-training algorithm[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(2): 331-339.
[3] ZHU Xiao-en, HAO Xin, XIA Shun-ren. Feature selection algorithm based on Levy flight[J]. Journal of ZheJiang University (Engineering Science), 2013, 47(4): 638-643.
[4] LI Wei-tao, ZHOU Xiao-jie, CHAI Tian-you. Gabor filter and latent semantic analysi based
burning state recognition[J]. Journal of ZheJiang University (Engineering Science), 2011, 45(12): 2120-2126.
[5] ZHANG Yu-hong, HU Xue-gang, YANG Qiu-jie. A feature selection approach suitable for
data stream classification[J]. Journal of ZheJiang University (Engineering Science), 2011, 45(12): 2247-2251.
[6] SHANG Jian, DIAO Li-Jie, YUE Heng, CHAI Tian-You. Soft sensor for ball mill load based on multisource data
feature fusion[J]. Journal of ZheJiang University (Engineering Science), 2010, 44(7): 1406-1413.
[7] XIE Jian-Fang, BO Xiao-Hong, WANG Zheng-Xiao, et al. Scheduling feature selection based on immune binary partial swarm optimization[J]. Journal of ZheJiang University (Engineering Science), 2009, 43(12): 2203-2207.