Nonlinear synchronization in EEG and whole-head MEG recordings of healthy subjects

国际自动化与计算杂志.英文版.1.Improved Exponential Stability Criteria for Uncertain Neutral System with Nonlinear Parameter PerturbationsFang Qiu，Ban-Tong Cui2.Robust Active Suspension Design Subject to Vehicle Inertial Parameter VariationsHai-Ping Du，Nong Zhang3.Delay-dependent Non-fragile H∞ Filtering for Uncertain Fuzzy Systems Based on Switching Fuzzy Model and Piecewise Lyapunov FunctionZhi-Le Xia，Jun-Min Li，Jiang-Rong Li4.Observer-based Adaptive Iterative Learning Control for Nonlinear Systems with Time-varying DelaysWei-Sheng Chen，Rui-Hong Li，Jing Li5.H∞ Output Feedback Control for Stochastic Systems with Mode-dependent Time-varying Delays and Markovian Jump ParametersXu-Dong Zhao，Qing-Shuang Zeng6.Delay and Its Time-derivative Dependent Robust Stability of Uncertain Neutral Systems with Saturating ActuatorsFatima El Haoussi，El Houssaine Tissir7.Parallel Fuzzy P+Fuzzy I+Fuzzy D Controller:Design and Performance EvaluationVineet Kumar，A.P.Mittal8.Observers for Descriptor Systems with Slope-restricted NonlinearitiesLin-Na Zhou，Chun-Yu Yang，Qing-Ling Zhang9.Parameterized Solution to a Class of Sylvester MatrixEquationsYu-Peng Qiao，Hong-Sheng Qi，Dai-Zhan Cheng10.Indirect Adaptive Fuzzy and Impulsive Control of Nonlinear SystemsHai-Bo Jiang11.Robust Fuzzy Tracking Control for Nonlinear Networked Control Systems with Integral Quadratic ConstraintsZhi-Sheng Chen，Yong He，Min Wu12.A Power-and Coverage-aware Clustering Scheme for Wireless Sensor NetworksLiang Xue，Xin-Ping Guan，Zhi-Xin Liu，Qing-Chao Zheng13.Guaranteed Cost Active Fault-tolerant Control of Networked Control System with Packet Dropout and Transmission DelayXiao-Yuan Luo，Mei-Jie Shang，Cai-Lian Chen，Xin-Ping Guanparison of Two Novel MRAS Based Strategies for Identifying Parameters in Permanent Magnet Synchronous MotorsKan Liu，Qiao Zhang，Zi-Qiang Zhu，Jing Zhang，An-Wen Shen，Paul Stewart15.Modeling and Analysis of Scheduling for Distributed Real-time Embedded SystemsHai-Tao Zhang，Gui-Fang Wu16.Passive Steganalysis Based on Higher Order Image Statistics of Curvelet TransformS.Geetha，Siva S.Sivatha Sindhu，N.Kamaraj17.Movement Invariants-based Algorithm for Medical Image Tilt CorrectionMei-Sen Pan，Jing-Tian Tang，Xiao-Li Yang18.Target Tracking and Obstacle Avoidance for Multi-agent SystemsJing Yan，Xin-Ping Guan，Fu-Xiao Tan19.Automatic Generation of Optimally Rigid Formations Using Decentralized MethodsRui Ren，Yu-Yan Zhang，Xiao-Yuan Luo，Shao-Bao Li20.Semi-blind Adaptive Beamforming for High-throughput Quadrature Amplitude Modulation SystemsSheng Chen，Wang Yao，Lajos Hanzo21.Throughput Analysis of IEEE 802.11 Multirate WLANs with Collision Aware Rate Adaptation AlgorithmDhanasekaran Senthilkumar，A. Krishnan22.Innovative Product Design Based on Customer Requirement Weight Calculation ModelChen-Guang Guo，Yong-Xian Liu，Shou-Ming Hou，Wei Wang23.A Service Composition Approach Based on Sequence Mining for Migrating E-learning Legacy System to SOAZhuo Zhang，Dong-Dai Zhou，Hong-Ji Yang，Shao-Chun Zhong24.Modeling of Agile Intelligent Manufacturing-oriented Production Scheduling SystemZhong-Qi Sheng，Chang-Ping Tang，Ci-Xing Lv25.Estimation of Reliability and Cost Relationship for Architecture-based SoftwareHui Guan，Wei-Ru Chen，Ning Huang，Hong-Ji Yang1.A Computer-aided Design System for Framed-mould in Autoclave ProcessingTian-Guo Jin，Feng-Yang Bi2.Wear State Recognition of Drills Based on K-means Cluster and Radial Basis Function Neural NetworkXu Yang3.The Knee Joint Design and Control of Above-knee Intelligent Bionic Leg Based on Magneto-rheological DamperHua-Long Xie，Ze-Zhong Liang，Fei Li，Li-Xin Guo4.Modeling of Pneumatic Muscle with Shape Memory Alloy and Braided SleeveBin-Rui Wang，Ying-Lian Jin，Dong Wei5.Extended Object Model for Product Configuration DesignZhi-Wei Xu，Ze-Zhong Liang，Zhong-Qi Sheng6.Analysis of Sheet Metal Extrusion Process Using Finite Element MethodXin-Cun Zhuang，Hua Xiang，Zhen Zhao7.Implementation of Enterprises' Interoperation Based on OntologyXiao-Feng Di，Yu-Shun Fan8.Path Planning Approach in Unknown EnvironmentTing-Kai Wang，Quan Dang，Pei-Yuan Pan9.Sliding Mode Variable Structure Control for Visual Servoing SystemFei Li，Hua-Long Xie10.Correlation of Direct Piezoelectric Effect on EAPap under Ambient FactorsLi-Jie Zhao，Chang-Ping Tang，Peng Gong11.XML-based Data Processing in Network Supported Collaborative DesignQi Wang，Zhong-Wei Ren，Zhong-Feng Guo12.Production Management Modelling Based on MASLi He，Zheng-Hao Wang，Ke-Long Zhang13.Experimental Tests of Autonomous Ground Vehicles with PreviewCunjia Liu，Wen-Hua Chen，John Andrews14.Modelling and Remote Control of an ExcavatorYang Liu，Mohammad Shahidul Hasan，Hong-Nian Yu15.TOPSIS with Belief Structure for Group Belief Multiple Criteria Decision MakingJiang Jiang，Ying-Wu Chen，Da-Wei Tang，Yu-Wang Chen16.Video Analysis Based on Volumetric Event DetectionJing Wang，Zhi-Jie Xu17.Improving Decision Tree Performance by Exception HandlingAppavu Alias Balamurugan Subramanian，S.Pramala，B.Rajalakshmi，Ramasamy Rajaram18.Robustness Analysis of Discrete-time Indirect Model Reference Adaptive Control with Normalized Adaptive LawsQing-Zheng Gao，Xue-Jun Xie19.A Novel Lifecycle Model for Web-based Application Development in Small and Medium EnterprisesWei Huang，Ru Li，Carsten Maple，Hong-Ji Yang，David Foskett，Vince Cleaver20.Design of a Two-dimensional Recursive Filter Using the Bees AlgorithmD. T. Pham，Ebubekir Ko(c)21.Designing Genetic Regulatory Networks Using Fuzzy Petri Nets ApproachRaed I. Hamed，Syed I. Ahson，Rafat Parveen1.State of the Art and Emerging Trends in Operations and Maintenance of Offshore Oil and Gas Production Facilities: Some Experiences and ObservationsJayantha P.Liyanage2.Statistical Safety Analysis of Maintenance Management Process of Excavator UnitsLjubisa Papic，Milorad Pantelic，Joseph Aronov，Ajit Kumar Verma3.Improving Energy and Power Efficiency Using NComputing and Approaches for Predicting Reliability of Complex Computing SystemsHoang Pham，Hoang Pham Jr.4.Running Temperature and Mechanical Stability of Grease as Maintenance Parameters of Railway BearingsJan Lundberg，Aditya Parida，Peter S(o)derholm5.Subsea Maintenance Service Delivery: Mapping Factors Influencing Scheduled Service DurationEfosa Emmanuel Uyiomendo，Tore Markeset6.A Systemic Approach to Integrated E-maintenance of Large Engineering PlantsAjit Kumar Verma，A.Srividya，P.G.Ramesh7.Authentication and Access Control in RFID Based Logistics-customs Clearance Service PlatformHui-Fang Deng，Wen Deng，Han Li，Hong-Ji Yang8.Evolutionary Trajectory Planning for an Industrial RobotR.Saravanan，S.Ramabalan，C.Balamurugan，A.Subash9.Improved Exponential Stability Criteria for Recurrent Neural Networks with Time-varying Discrete and Distributed DelaysYuan-Yuan Wu，Tao Li，Yu-Qiang Wu10.An Improved Approach to Delay-dependent Robust Stabilization for Uncertain Singular Time-delay SystemsXin Sun，Qing-Ling Zhang，Chun-Yu Yang，Zhan Su，Yong-Yun Shao11.Robust Stability of Nonlinear Plants with a Non-symmetric Prandtl-Ishlinskii Hysteresis ModelChang-An Jiang，Ming-Cong Deng，Akira Inoue12.Stability Analysis of Discrete-time Systems with Additive Time-varying DelaysXian-Ming Tang，Jin-Shou Yu13.Delay-dependent Stability Analysis for Markovian Jump Systems with Interval Time-varying-delaysXu-Dong Zhao，Qing-Shuang Zeng14.H∞ Synchronization of Chaotic Systems via Delayed Feedback ControlLi Sheng，Hui-Zhong Yang15.Adaptive Fuzzy Observer Backstepping Control for a Class of Uncertain Nonlinear Systems with Unknown Time-delayShao-Cheng Tong，Ning Sheng16.Simulation-based Optimal Design of α-β-γ-δ FilterChun-Mu Wu，Paul P.Lin，Zhen-Yu Han，Shu-Rong Li17.Independent Cycle Time Assignment for Min-max SystemsWen-De Chen，Yue-Gang Tao，Hong-Nian Yu1.An Assessment Tool for Land Reuse with Artificial Intelligence MethodDieter D. Genske，Dongbin Huang，Ariane Ruff2.Interpolation of Images Using Discrete Wavelet Transform to Simulate Image Resizing as in Human VisionRohini S. Asamwar，Kishor M. Bhurchandi，Abhay S. Gandhi3.Watermarking of Digital Images in Frequency DomainSami E. I. Baba，Lala Z. Krikor，Thawar Arif，Zyad Shaaban4.An Effective Image Retrieval Mechanism Using Family-based Spatial Consistency Filtration with Object RegionJing Sun，Ying-Jie Xing5.Robust Object Tracking under Appearance Change ConditionsQi-Cong Wang，Yuan-Hao Gong，Chen-Hui Yang，Cui-Hua Li6.A Visual Attention Model for Robot Object TrackingJin-Kui Chu，Rong-Hua Li，Qing-Ying Li，Hong-Qing Wang7.SVM-based Identification and Un-calibrated Visual Servoing for Micro-manipulationXin-Han Huang，Xiang-Jin Zeng，Min Wang8.Action Control of Soccer Robots Based on Simulated Human IntelligenceTie-Jun Li，Gui-Qiang Chen，Gui-Fang Shao9.Emotional Gait Generation for a Humanoid RobotLun Xie，Zhi-Liang Wang，Wei Wang，Guo-Chen Yu10.Cultural Algorithm for Minimization of Binary Decision Diagram and Its Application in Crosstalk Fault DetectionZhong-Liang Pan，Ling Chen，Guang-Zhao Zhang11.A Novel Fuzzy Direct Torque Control System for Three-level Inverter-fed Induction MachineShu-Xi Liu，Ming-Yu Wang，Yu-Guang Chen，Shan Li12.Statistic Learning-based Defect Detection for Twill FabricsLi-Wei Han，De Xu13.Nonsaturation Throughput Enhancement of IEEE 802.11b Distributed Coordination Function for Heterogeneous Traffic under Noisy EnvironmentDhanasekaran Senthilkumar，A. Krishnan14.Structure and Dynamics of Artificial Regulatory Networks Evolved by Segmental Duplication and Divergence ModelXiang-Hong Lin，Tian-Wen Zhang15.Random Fuzzy Chance-constrained Programming Based on Adaptive Chaos Quantum Honey Bee Algorithm and Robustness AnalysisHan Xue，Xun Li，Hong-Xu Ma16.A Bit-level Text Compression Scheme Based on the ACW AlgorithmHussein A1-Bahadili，Shakir M. Hussain17.A Note on an Economic Lot-sizing Problem with Perishable Inventory and Economies of Scale Costs:Approximation Solutions and Worst Case AnalysisQing-Guo Bai，Yu-Zhong Zhang，Guang-Long Dong1.Virtual Reality: A State-of-the-Art SurveyNing-Ning Zhou，Yu-Long Deng2.Real-time Virtual Environment Signal Extraction and DenoisingUsing Programmable Graphics HardwareYang Su，Zhi-Jie Xu，Xiang-Qian Jiang3.Effective Virtual Reality Based Building Navigation Using Dynamic Loading and Path OptimizationQing-Jin Peng，Xiu-Mei Kang，Ting-Ting Zhao4.The Skin Deformation of a 3D Virtual HumanXiao-Jing Zhou，Zheng-Xu Zhao5.Technology for Simulating Crowd Evacuation BehaviorsWen-Hu Qin，Guo-Hui Su，Xiao-Na Li6.Research on Modelling Digital Paper-cut PreservationXiao-Fen Wang，Ying-Rui Liu，Wen-Sheng Zhang7.On Problems of Multicomponent System Maintenance ModellingTomasz Nowakowski，Sylwia Werbinka8.Soft Sensing Modelling Based on Optimal Selection of Secondary Variables and Its ApplicationQi Li，Cheng Shao9.Adaptive Fuzzy Dynamic Surface Control for Uncertain Nonlinear SystemsXiao-Yuan Luo，Zhi-Hao Zhu，Xin-Ping Guan10.Output Feedback for Stochastic Nonlinear Systems with Unmeasurable Inverse DynamicsXin Yu，Na Duan11.Kalman Filtering with Partial Markovian Packet LossesBao-Feng Wang，Ge Guo12.A Modified Projection Method for Linear FeasibilityProblemsYi-Ju Wang，Hong-Yu Zhang13.A Neuro-genetic Based Short-term Forecasting Framework for Network Intrusion Prediction SystemSiva S. Sivatha Sindhu，S. Geetha，M. Marikannan，A. Kannan14.New Delay-dependent Global Asymptotic Stability Condition for Hopfield Neural Networks with Time-varying DelaysGuang-Deng Zong，Jia Liu hHTTp://15.Crosscumulants Based Approaches for the Structure Identification of Volterra ModelsHouda Mathlouthi，Kamel Abederrahim，Faouzi Msahli，Gerard Favier1.Coalition Formation in Weighted Simple-majority Games under Proportional Payoff Allocation RulesZhi-Gang Cao，Xiao-Guang Yang2.Stability Analysis for Recurrent Neural Networks with Time-varying DelayYuan-Yuan Wu，Yu-Qiang Wu3.A New Type of Solution Method for the Generalized Linear Complementarity Problem over a Polyhedral ConeHong-Chun Sun，Yan-Liang Dong4.An Improved Control Algorithm for High-order Nonlinear Systems with Unmodelled DynamicsNa Duan，Fu-Nian Hu，Xin Yu5.Controller Design of High Order Nonholonomic System with Nonlinear DriftsXiu-Yun Zheng，Yu-Qiang Wu6.Directional Filter for SAR Images Based on NonsubsampledContourlet Transform and Immune Clonal SelectionXiao-Hui Yang，Li-Cheng Jiao，Deng-Feng Li7.Text Extraction and Enhancement of Binary Images Using Cellular AutomataG. Sahoo，Tapas Kumar，B.L. Rains，C.M. Bhatia8.GH2 Control for Uncertain Discrete-time-delay Fuzzy Systems Based on a Switching Fuzzy Model and Piecewise Lyapunov FunctionZhi-Le Xia，Jun-Min Li9.A New Energy Optimal Control Scheme for a Separately Excited DC Motor Based Incremental Motion DriveMilan A.Sheta，Vivek Agarwal，Paluri S.V.Nataraj10.Nonlinear Backstepping Ship Course ControllerAnna Witkowska，Roman Smierzchalski11.A New Method of Embedded Fourth Order with Four Stages to Study Raster CNN SimulationR. Ponalagusamy，S. Senthilkumar12.A Minimum-energy Path-preserving Topology Control Algorithm for Wireless Sensor NetworksJin-Zhao Lin，Xian Zhou，Yun Li13.Synchronization and Exponential Estimates of Complex Networks with Mixed Time-varying Coupling DelaysYang Dai，YunZe Cai，Xiao-Ming Xu14.Step-coordination Algorithm of Traffic Control Based on Multi-agent SystemHai-Tao Zhang，Fang Yu，Wen Li15.A Research of the Employment Problem on Common Job-seekersand GraduatesBai-Da Qu。

第１４卷㊀第３期Ｖｏｌ．１４Ｎｏ．３㊀㊀智㊀能㊀计㊀算㊀机㊀与㊀应㊀用ＩｎｔｅｌｌｉｇｅｎｔＣｏｍｐｕｔｅｒａｎｄＡｐｐｌｉｃａｔｉｏｎｓ㊀㊀２０２４年３月㊀Ｍａｒ．２０２４㊀㊀㊀㊀㊀㊀文章编号：２０９５－２１６３（２０２４）０３－０１２８－０５中图分类号：ＴＰ１８，ＴＰ３９９文献标志码：Ａ基于平滑非线性能量算子划分的尖峰相关特征癫痫发作自动检测算法何雪兰，吴㊀江，蒋路茸（浙江理工大学信息科学与工程学院，杭州３１００１８）摘㊀要：针对癫痫发作自动检测算法多集中于时域㊁频域等传统特征，无法全面表征癫痫脑电信号的信息等问题，本文结合癫痫脑电图中异常波振幅和频率提高的现象，提出一种基于平滑非线性能量算子划分的尖峰相关特征癫痫发作自动检测算法㊂该算法使用传统的时域㊁频域特征，结合尖峰相关性特征对脑电信号进行刻画，使用有监督的机器学习分类器，测试癫痫发作自动检测的有效性和可靠性㊂本文将提出的方法在开源数据集ＣＨＢＭＩＴ上进行了评估，获得了９６．５２％的准确率㊁９５．６５％的敏感性和９７．０９％的特异性㊂实验结果表明，基于平滑非线性能量算子划分的尖峰相关特征，能够作为癫痫脑电信息的补充，提高癫痫发作检测的性能㊂关键词：癫痫发作检测；机器学习；尖峰相关性；平滑非线性能量算子Ａｕｔｏｍａｔｉｃｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎａｌｇｏｒｉｔｈｍｂａｓｅｄｏｎｓｐｉｋｅ－ｒｅｌａｔｅｄｆｅａｔｕｒｅｓｏｆｓｍｏｏｔｈｅｄｎｏｎｌｉｎｅａｒｅｎｅｒｇｙｏｐｅｒａｔｏｒｄｉｖｉｓｉｏｎＨＥＸｕｅｌａｎ，ＷＵＪｉａｎｇ，ＪＩＡＮＧＬｕｒｏｎｇ（ＳｃｈｏｏｌｏｆＩｎｆｏｒｍａｔｉｏｎＳｃｉｅｎｃｅａｎｄＥｎｇｉｎｅｅｒｉｎｇ，ＺｈｅｊｉａｎｇＳｃｉ－ＴｅｃｈＵｎｉｖｅｒｓｉｔｙ，Ｈａｎｇｚｈｏｕ３１００１８，Ｃｈｉｎａ）Ａｂｓｔｒａｃｔ：Ｍｏｓｔｃｕｒｒｅｎｔｓｅｉｚｕｒｅａｕｔｏｍａｔｉｃｄｅｔｅｃｔｉｏｎａｌｇｏｒｉｔｈｍｓｆｏｃｕｓｏｎｔｒａｄｉｔｉｏｎａｌｆｅａｔｕｒｅｓｓｕｃｈａｓｔｉｍｅｄｏｍａｉｎａｎｄｆｒｅｑｕｅｎｃｙｄｏｍａｉｎ，ｗｈｉｃｈｃａｎｎｏｔｆｕｌｌｙｃｈａｒａｃｔｅｒｉｚｅｔｈｅｉｎｆｏｒｍａｔｉｏｎｏｆｅｐｉｌｅｐｔｉｃＥＥＧｓｉｇｎａｌｓ．Ｔｈｉｓｐａｐｅｒｐｒｏｐｏｓｅｓａｎａｕｔｏｍａｔｉｃｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎａｌｇｏｒｉｔｈｍｂａｓｅｄｏｎｓｐｉｋｅｃｏｒｒｅｌａｔｉｏｎｆｅａｔｕｒｅｓｄｉｖｉｄｅｄｂｙａｓｍｏｏｔｈｎｏｎｌｉｎｅａｒｅｎｅｒｇｙｏｐｅｒａｔｏｒ，ｔａｋｉｎｇｉｎｔｏａｃｃｏｕｎｔｔｈｅｐｈｅｎｏｍｅｎｏｎｔｈａｔｔｈｅａｍｐｌｉｔｕｄｅａｎｄｆｒｅｑｕｅｎｃｙｏｆａｂｎｏｒｍａｌｗａｖｅｓｉｎｅｐｉｌｅｐｔｉｃＥＥＧｗｉｌｌｉｎｃｒｅａｓｅ．Ｔｈｅａｌｇｏｒｉｔｈｍｕｓｅｓｔｒａｄｉｔｉｏｎａｌｔｉｍｅ－ｄｏｍａｉｎａｎｄｆｒｅｑｕｅｎｃｙ－ｄｏｍａｉｎｆｅａｔｕｒｅｓ，ｃｏｍｂｉｎｅｄｗｉｔｈｓｐｉｋｅｃｏｒｒｅｌａｔｉｏｎｆｅａｔｕｒｅｓｔｏｃｈａｒａｃｔｅｒｉｚｅｔｈｅＥＥＧｓｉｇｎａｌ，ａｎｄｕｓｅｓｓｕｐｅｒｖｉｓｅｄｍａｃｈｉｎｅｌｅａｒｎｉｎｇｃｌａｓｓｉｆｉｅｒｓｔｏｔｅｓｔｉｔｓｅｆｆｅｃｔｉｖｅｎｅｓｓａｎｄｒｅｌｉａｂｉｌｉｔｙｆｏｒａｕｔｏｍａｔｉｃｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎ．ＴｈｅｒｅｓｅａｒｃｈｅｖａｌｕａｔｅｓｔｈｅｐｒｏｐｏｓｅｄｍｅｔｈｏｄｏｎｔｈｅｏｐｅｎｓｏｕｒｃｅｄａｔａｓｅｔＣＨＢＭＩＴａｎｄｏｂｔａｉｎｓ９６．５２％ｏｎａｃｃｕｒａｃｙ，９５．６５％ｏｎｓｅｎｓｉｔｉｖｉｔｙａｎｄ９７．０９％ｏｎｓｐｅｃｉｆｉｃｉｔｙ．Ｔｈｅｅｘｐｅｒｉｍｅｎｔａｌｒｅｓｕｌｔｓｓｈｏｗｔｈａｔｔｈｅｐｒｏｐｏｓｅｄｓｐｉｋｅ－ｒｅｌａｔｅｄｆｅａｔｕｒｅｓｂａｓｅｄｏｎｔｈｅｓｍｏｏｔｈｅｄｎｏｎｌｉｎｅａｒｅｎｅｒｇｙｏｐｅｒａｔｏｒｓｅｇｍｅｎｔａｔｉｏｎｃａｎｂｅｕｓｅｄａｓａｃｏｍｐｌｅｍｅｎｔｔｏｔｈｅｅｐｉｌｅｐｔｉｃＥＥＧｉｎｆｏｒｍａｔｉｏｎｔｏｉｍｐｒｏｖｅｔｈｅｐｅｒｆｏｒｍａｎｃｅｏｆｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎ．Ｋｅｙｗｏｒｄｓ：ｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎ；ｍａｃｈｉｎｅｌｅａｒｎｉｎｇ；ｓｐｉｋｅｃｏｒｒｅｌａｔｉｏｎ；ｓｍｏｏｔｈｅｄｎｏｎｌｉｎｅａｒｅｎｅｒｇｙｏｐｅｒａｔｏｒ基金项目：浙江省基础公益项目（ＬＧＦ１９Ｆ０１０００８）；北京邮电大学泛网无线通信教育部重点实验室（ＢＵＰＴ）（ＫＦＫＴ－２０１８１０１）；浙江省重点研发计（２０２２Ｃ０３１３６）；国家自然科学基金（６１６０２４１７）㊂作者简介：何雪兰（１９９９－），女，硕士研究生，主要研究方向：癫痫检测；吴㊀江（１９７８－），男，博士，高级工程师，主要研究方向：无线通信技术，工业物联网㊂通讯作者：蒋路茸（１９８２－），男，博士，教授，主要研究方向：生理电信号处理㊁复杂网络和无线传感器网络㊂Ｅｍａｉｌ：ｊｉａｎｇｌｕｒｏｎｇ＠ｚｓｔｕ．ｅｄｕ．ｃｎ收稿日期：２０２３－０３－１５０㊀引㊀言癫痫是一种神经系统疾病，由大脑神经元异常放电引起［１］，常常表现为突发性㊁反复性和复发性等特点㊂癫痫发作的临床症状复杂多样，如阵发性痉挛㊁意识丧失㊁认知功能障碍等［２］㊂这些发作事件对患者的认知水平及正常生活都产生了明显影响㊂因此，癫痫的诊断和治疗对于预防癫痫发作和改善生活质量至关重要㊂头皮脑电图是一种用于临床记录脑活动的无创信号采集方法［３］，用于记录大脑活动时的电波变化㊂头皮脑电图包含丰富的生理㊁心理和病理信息，是评估癫痫和其他脑部疾病的有效工具［４］㊂在脑电图的记录中，癫痫发作和癫痫样放电（如棘波㊁尖波和棘慢波复合体）是癫痫的重要生物标志物［５］，并被广泛应用于临床评价㊂目前，临床上基于脑电图的识别与分析是医生进行癫痫检测的黄金标准，但对海量的临床脑电数据进行人工筛查，不仅给医生带来沉重的负担，还存在较强的主观性㊁判断标准不统一等问题［６－７］，影响分析的效率和准确性㊂因此，设计一种自动的癫痫发作检测方法是亟待解决的问题㊂为了克服传统诊断方法的局限性㊁提高医疗效率，伴随着机器学习的快速发展，癫痫发作的自动检测已成为行业内关注的重点㊂研究者们根据头皮脑电图的时域㊁频域或非线性特征建立了特征工程方法［８－１０］，并通过具有一个或多个特征的分类器检测癫痫发作㊂Ｍｕｒｓａｌｉｎ等学者［１１］从时域㊁频域和基于熵的特征中选择突出特征，使用随机森林分类器学习选定特征集合的特性，获得了更好的分类结果㊂杨舒涵等学者［１２］使用时域和非线性特征对脑电信号进行表征，结合ＸＧＢｏｏｓｔ分类器，实现了癫痫的自动检测㊂Ｚａｒｅｉ等学者［１３］使用离散小波变换ＤＷＴ和正交匹配追踪（ＯｒｔｈｏｇｏｎａｌＭａｔｃｈｉｎｇＰｕｒｓｕｉｔ，ＯＭＰ）提取ＥＥＧ中不同的系数，计算非线性特征和统计特征，使用ＳＶＭ进行分类，获得了较好的检测性能㊂吴端波等学者［１４］使用ａＥＥＧ尖峰和ｃＥＥＧ棘波提取的方法计算棘波率，使用阈值法对癫痫进行发作检测㊂上述模型虽然都能取得较好的分类结果，但是也存在以下问题：（１）多数研究在特征提取阶段仅从时域㊁频域或时频域中表征脑电信号信息，这些特征所涵盖的信息量并不足以全面描述一段ＥＥＧ信号㊂（２）在癫痫发作的自动检测中，强调周期性的信号转换对于有效㊁可靠地区分癫痫发作的重复特征至关重要，而互相关是时域上广泛用于表示信号周期性的方法㊂针对上述问题，本文提出一种基于平滑非线性能量算子划分的尖峰相关（ＳｐｉｋｅＣｏｒｒｅｌａｔｉｏｎ，ＳＣ）特征的癫痫发作自动检测算法㊂ＳＣ是关于自适应提取的脑电图尖峰信号段之间时间延迟的最大互相关㊂使用平滑非线性能量算子衡量癫痫脑电信号中出现的异常波，将脑电信号在癫痫发作期和非发作期的尖峰相关特征作为度量患者大脑活动的一个重要补充㊂本文提出的算法主要使用巴特沃斯滤波器对脑电信号进行滤波，去除外部伪迹的干扰，然后从传统特征角度出发，提取时域㊁频域特征，再结合提出的尖峰相关特征，进一步表征癫痫发作时的异常信息㊂最后结合有监督的机器学习分类模型，实现癫痫发作的自动检测㊂１㊀方法癫痫发作自动检测整体流程设计如图１所示，其中包含预处理㊁特征提取和分类等３个模块㊂脑电信号通道筛选滤波数据分割归一化预处理特征提取传统特征:时域、频域尖峰相关特征分类癫痫发作/非发作图１㊀癫痫发作自动检测流程图Ｆｉｇ．１㊀Ｆｌｏｗｃｈａｒｔｏｆｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎ１．１㊀脑电信号预处理头皮脑电数据通过放置在头皮固定位置的电极采集得到㊂由于外置电极，这种采集方式很容易受到外部干扰，导致采集到的数据被噪声污染㊂此外，由于受试者在采集过程中生理活动产生的内部伪迹（如：眨眼㊁心脏跳动等）［１５］，也会对数据产生干扰，影响分类结果㊂因此，针对内部伪迹，本文首先对采集到的脑电信号进行通道筛选，剔除受眼部运动干扰严重的２个电极ＦＴ１和ＦＴ２；同时，由于左侧耳电磁极易受到心电伪迹的干扰，因此也剔除了靠近耳部的２个电极ＦＴ９和ＦＴ１０㊂所以，在通道筛选阶段，共选择了脑电图中２０个通道信号㊂㊀㊀滤波是一种常见的去除脑电信号外部伪迹的方法，本文采用１ ４８Ｈｚ的带通巴特沃斯滤波器进行滤波，抑制其他频率范围的信号［１６］㊂根据数据集中标注的癫痫发作开始和结束时间，为了保证波形的完整性，设置重叠率为５０％的滑动窗口，将脑电信号分割成４ｓ的数据片段，最后对所有片段进行归一化处理㊂由于通道筛选和滤波后的脑电信号幅值的浮动一般是在可接受范围内，最大最小标准化能够较大程度地还原真实ＥＥＧ信号波形㊂因此，本文采用最大最小标准化对原始ＥＥＧ信号进行归一化操作，推得的公式为：Ｘｍｉｎ－ｍａｘ＝Ｘ－Ｘ－ｍａｘＸ()－ｍｉｎＸ()（１）１．２㊀特征提取原始脑电信号数据量庞大，且不具有代表性，而特征提取方法可以提炼出能够表征癫痫发作特征的数据，用于模型的建立㊂因此，本文主要使用传统时域㊁频域特征和基于平滑非线性能量算子的尖峰相关性特征，对脑电数据进行特征提取㊂１．２．１㊀传统特征提取研究主要从时域和频域两个角度对脑电信号进行传统特征提取㊂本文主要提取时域上每个通道的最大值㊁最小值㊁平均值㊁峰度（Ｋｕｒｔｏｓｉｓ）㊁偏斜度９２１第３期何雪兰，等：基于平滑非线性能量算子划分的尖峰相关特征癫痫发作自动检测算法（Ｓｋｎｅｗｎｅｓｓ）和线长（ＬｉｎｅＬｅｎｇｔｈ）；频域上主要提取每个信号频域分量的振幅㊂其中，峰度㊁偏斜度和线长的数学定义分别见式（２） （４）：Ｋｕｒｔｏｓｉｓ＝Ｅ［（ｘ－ｍｅａｎ（ｘ））４］｛Ｅ［（ｘ－ｍｅａｎ（ｘ））２］｝２（２）Ｓｋｎｅｗｎｅｓｓ＝Ｅ［（ｘ－ｍｅａｎ（ｘ）ｓｔｄ（ｘ））３］（３）ＬｉｎｅＬｅｎｇｔｈ＝１ｎðｎｉ＝１ａｂｓｘｉ＋１－ｘｉ()（４）㊀㊀其中，ｘ表示脑电信号片段；Ｅ表示对括号中数值求期望；ｘｉ表示采样点ｉ的值；ｎ表示片段ｘ中采样点数㊂１．２．２㊀尖峰相关特征提取根据癫痫发作时脑电信号异常波的振幅和频率发生改变的特异性现象，本文提出将尖峰相关特征作为表征癫痫发作时异常特征的补充㊂非线性能量算子（ＮＬＥＯ）是一种对信号进行能量度量的方法［１７］，能够跟踪信号的瞬时能量㊂对于离散信号ｘｎ，其非线性能量算子表达如式（５）所示：φ［ｘ（ｎ）］＝ｘ（ｎ－ｌ）ｘ（ｎ－ｐ）－ｘ（ｎ－ｑ）ｘ（ｎ－ｓ）（５）㊀㊀通常，当癫痫脑电信号中出现异常放电时，脑电波的振幅和频率会有所提高，可以更好地突出异常波在平稳状态下的放电波形，但非线性能量对脑电信号中可能存在的噪音信号也具有很高的敏感度㊂为了进一步提高ＮＬＥＯ对非平稳信号的表征能力和抗干扰能力，文献［１８］提出了一种ＮＬＥＯ的改进方法，即平滑非线性能量算子（ＳＮＬＥＯ），将计算所得的能量与一个窗函数进行卷积运算，在一定程度上减小低波幅噪音信号对输出结果的影响㊂ＳＮＬＥＯ计算见式（６）：φ［ｘ（ｎ）］＝ｗ（ｎ）∗φ［ｘ（ｎ）］（６）㊀㊀其中，ｗ是一个矩形的窗函数， ∗ 表示卷积操作㊂在非线性能量算子的计算中，本文使用的参数值为ｌ＝１，ｐ＝２，ｑ＝０和ｓ＝３，并采用７个点的窗函数进行卷积计算㊂获得ＳＮＬＥＯ后，需要设定一个合适的阈值，尽可能多地筛选出可能是尖峰的样本，同时最小化漏检率㊂本文使用自适应阈值，对ＳＮＬＥＯ进行尖峰筛选识别㊂本文采用影响检测尖峰数量没有大范围变化的阈值作为最优阈值㊂最优阈值的搜索范围为ＳＮＬＥＯ的１０％ ９０％［１９］，相邻２个峰值的中间被确定为一个尖峰的起始点或结束点㊂由于数据在划分过程中导致波形的不连续问题，本文将检测到的第一个和最后一个尖峰丢弃，以确保每个片段具有完整的尖峰形态㊂如果检测出尖峰，则将每个划分好的尖峰与后续５个尖峰片段相关联㊂本文使用尖峰相关性（ＳｐｉｋｅＣｏｒｒｅｌａｔｉｏｎ，ＳＣ）来定义该矩阵，并将ＳＣ的平均值和标准差作为癫痫发作检测的特征㊂ＳＣ计算见式（７ ８）：ＳＣｉ，ｊ＝ｍａｘｍＲｘｉｘｊ（ｍ）（７）Ｒｘｉｘｊ（ｍ；ｉ，ｊ）＝Ｅ［ｘｉ（ｎ）ｘｊ（ｎ＋ｍ）］σｘｉσｘｊ（８）㊀㊀其中，ｘｉ㊁ｘｊ是脑电ＥＥＧ信号的片段，这里ｉ＝［２， ，Ｓ－６］，ｊ＝［ｉ＋１， ，ｉ＋５］；Ｓ表示在一个片段中检测到的峰值数；σ表示脑电图片段的标准差㊂估计ＳＣ特征的处理过程如图２所示㊂将一个片段的第一个和最后一个丢弃，而后根据得到的尖峰计算其与后面５个尖峰的相关性㊂根据图２（ａ）中样例计算出的尖峰相关矩阵如图３所示㊂10050-501234时间/s(a)癫痫发作片段样例EEG/μV400200SNLEO/μV23224168尖峰数/个1234时间/s(b)片段(a)对应的S N L E Ot h(c)基于（b)确定的自适应阈值t h阈值104.87710050-50EEG/μV1234时间/s(d)划分好的尖峰片段(“*”表示丢弃的片段)12345678910t h图２㊀使用自适应阈值的ＳＮＬＥＯ计算尖峰相关性示意图Ｆｉｇ．２㊀ＳｃｈｅｍａｔｉｃｄｉａｇｒａｍｏｆＳＮＬＥＯｃａｌｃｕｌａｔｉｏｎｏｆｓｐｉｋｅｃｏｒｒｅｌａｔｉｏｎｓｕｓｉｎｇａｄａｐｔｉｖｅｔｈｒｅｓｈｏｌｄ1234567892345678910图３㊀尖峰片段得到的最大相关矩阵Ｆｉｇ．３㊀Ｍａｘｉｍｕｍｃｏｒｒｅｌａｔｉｏｎｍａｔｒｉｘｏｂｔａｉｎｅｄｆｒｏｍｓｐｉｋｅｆｒａｇｍｅｎｔｓ０３１智㊀能㊀计㊀算㊀机㊀与㊀应㊀用㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀第１４卷㊀㊀㊀此外，计算了ＳＮＬＥＯ的平均值㊁标准差和平均最大ＳＮＬＥＯ值ｓｐｉｋｉｎｅｓｓ㊂其中，ｓｐｉｋｉｎｅｓｓ被定义为ＳＮＬＥＯ中峰值的最大值除以ＳＮＬＥＯ的平均值［２０］，以及检测到的峰值数量（ｓｎｕｍ）㊁平均持续时间（ｓｗｉｄｔｈ）和平均峰值间间隔（ｓｇａｐ）㊂基于ＳＮＬＥＯ划分的尖峰相关特征的具体描述见表１㊂表１㊀尖峰相关特征的描述Ｔａｂｌｅ１㊀Ｄｅｓｃｒｉｐｔｉｏｎｏｆｓｐｉｋｅ－ｒｅｌａｔｅｄｃｈａｒａｃｔｅｒｉｓｔｉｃｓ特征描述ｍｅａｎ（ＳＣ）尖峰相关性矩阵的平均值ｓｔｄ（ＳＣ）尖峰相关性矩阵的标准差ｍｅａｎ（ＳＮＬＥＯ）ＳＮＬＥＯ的平均值ｓｔｄ（ＳＮＬＥＯ）ＳＮＬＥＯ的标准差ｓｐｉｋｉｎｅｓｓ平均最大ＳＮＬＥＯ值ｓｎｕｍ峰值数量ｓｗｉｄｔｈ平均持续时间ｓｇａｐ平均峰值间间距１．３㊀分类模型使用传统机器学习分类器ＲＦ和ＳＶＭ来评估本文提出的方法，这些分类器经常被用于癫痫发作的自动检测㊂２㊀实验２．１㊀数据集本研究采用公开的头皮脑电数据集ＣＨＢ－ＭＩＴ㊂该数据集共记录了美国波士顿儿童医院的２３名癫痫患者的头皮脑电数据，每个患者的数据都是由多个．ｅｄｆ文件组成，采样频率２５６Ｈｚ，共含有１５７次癫痫发作㊂大多数文件包含有２３个ＥＥＧ通道信号，并采用国际标准１０－２０系统使用的ＥＥＧ电极位置命名这些通道记录㊂由于癫痫发作时间远小于发作间期的时间，为了保证数据集正负样本的均衡性，本文采用欠采样的方式在发作间期随机采样和癫痫发作样本数量相当的负样本㊂２．２㊀评价指标为了验证本文方法的有效性，采用准确率（Ａｃｃ）㊁敏感性（Ｓｅｎ）㊁特异性（Ｓｐｅ）㊁Ｆ１值和ＡＵＣ等指标进行实验评估㊂计算方法见式（９） 式（１１）：Ａｃｃ＝ＴＰ＋ＴＮＴＰ＋ＴＮ＋ＦＰ＋ＦＮ（９）Ｓｅｎ＝ＴＰＴＰ＋ＦＮ（１０）Ｓｐｅ＝ＴＮＴＮ＋ＦＰ（１１）㊀㊀其中，ＴＰ㊁ＦＰ㊁ＦＮ和ＴＮ分别为真阳性㊁假阳性㊁假阴性和真阴性㊂本文产生的所有实验结果都是在配置为Ｉｎｔｅｌ（Ｒ）Ｃｏｒｅ（ＴＭ）ｉ７－９７００ＣＰＵ＠３．００ＧＨｚ，１６ＧＢＲＡＭ的计算机上实现的㊂实验模型使用Ｐｙｔｈｏｎ３．７和Ｓｃｉｋｉｔ－ｌｅａｒｎ构建㊂２．３㊀结果分析本文先对提取的传统时域㊁频域特征分别使用ＲＦ和ＳＶＭ分类模型进行测试，所得实验结果见表２㊂由表２可知，ＳＶＭ分类模型表现最佳㊂表２㊀基于传统特征的实验结果Ｔａｂｌｅ２㊀Ｅｘｐｅｒｉｍｅｎｔａｌｒｅｓｕｌｔｓｂａｓｅｄｏｎｔｒａｄｉｔｉｏｎａｌｃｈａｒａｃｔｅｒｉｓｔｉｃｓ特征分类器ＡｃｃＳｅｎＳｐｅ传统特征ＲＦ０．８６２１０．７５３３０．９３３９ＳＶＭ０．９５９００．９３８５０．９７３６㊀㊀在确定分类模型ＳＶＭ的基础上，将传统特征和尖峰相关特征结合，探讨尖峰相关特征对癫痫脑电信号的表征能力㊂添加前后对比结果见表３㊂表３㊀尖峰相关特征对比的分类结果Ｔａｂｌｅ３㊀Ｃｌａｓｓｉｆｉｃａｔｉｏｎｒｅｓｕｌｔｓｏｆｓｐｉｋｅ－ｒｅｌａｔｅｄｆｅａｔｕｒｅｃｏｍｐａｒｉｓｏｎ分类器特征ＡｃｃＳｅｎＳｐｅＳＶＭ传统特征０．９５９００．９３８５０．９７３６传统特征＋尖峰相关特征０．９６５２０．９５６５０．９７０９㊀㊀由表３可知，尖峰相关特征能够对癫痫脑电信号信息进行表征㊂加入尖峰相关特征后，检测结果在Ａｃｃ上提升了０．６２％，在Ｓｅｎ上提升了１．８％，在Ｓｐｅ上有所降低㊂在实际的临床应用中，正确识别发作样本比正确识别非发作样本更重要，因此Ｓｅｎ指标更能准确衡量方法的优劣㊂本文提出的方法虽然在Ｓｐｅ上略有降低，但Ｓｅｎ指标上有一定程度的提升㊂３㊀结束语本文提出了一种基于平滑非线性能量算子划分的尖峰相关特征的癫痫发作自动检测算法㊂该算法使用传统的时域㊁频域特征，结合尖峰相关性特征对脑电信号进行刻画，使用ＲＦ和ＳＶＭ分类器来测试癫痫发作自动检测的有效性和可靠性㊂将所提方法在开源数据集ＣＨＢ－ＭＩＴ上进行了评估，ＳＶＭ分类器获得了更好的结果，其准确率㊁敏感性和特异性分别为９６．５２％，９５．６５％和９７．０９％㊂此外，研究开展的特征消融实验结果表明，提出的基于平滑非线性能１３１第３期何雪兰，等：基于平滑非线性能量算子划分的尖峰相关特征癫痫发作自动检测算法量算子划分的尖峰相关特征，能够作为癫痫脑电信息的补充，进一步提高癫痫发作检测的性能㊂参考文献［１］ＰＡＴＥＬＤＣ，ＴＥＷＡＲＩＢＰ，ＣＨＡＵＮＳＡＬＩＬ，ｅｔａｌ．Ｎｅｕｒｏｎ–ｇｌｉａｉｎｔｅｒａｃｔｉｏｎｓｉｎｔｈｅｐａｔｈｏｐｈｙｓｉｏｌｏｇｙｏｆｅｐｉｌｅｐｓｙ［Ｊ］．ＮａｔｕｒｅＲｅｖｉｅｗｓＮｅｕｒｏｓｃｉｅｎｃｅ，２０１９，２０（５）：２８２－２９７．［２］ＳＰＥＣＣＨＩＯＮ，ＷＩＲＲＥＬＬＥＣ，ＳＣＨＥＦＦＥＲＩＥ，ｅｔａｌ．ＩｎｔｅｒｎａｔｉｏｎａｌＬｅａｇｕｅＡｇａｉｎｓｔＥｐｉｌｅｐｓｙｃｌａｓｓｉｆｉｃａｔｉｏｎａｎｄｄｅｆｉｎｉｔｉｏｎｏｆｅｐｉｌｅｐｓｙｓｙｎｄｒｏｍｅｓｗｉｔｈｏｎｓｅｔｉｎｃｈｉｌｄｈｏｏｄ：ＰｏｓｉｔｉｏｎｐａｐｅｒｂｙｔｈｅＩＬＡＥＴａｓｋＦｏｒｃｅｏｎＮｏｓｏｌｏｇｙａｎｄＤｅｆｉｎｉｔｉｏｎｓ［Ｊ］．Ｅｐｉｌｅｐｓｉａ，２０２２，６３（６）：１３９８－１４４２．［３］ＳＣＨＡＤＡ，ＳＣＨＩＮＤＬＥＲＫ，ＳＣＨＥＬＴＥＲＢ，ｅｔａｌ．Ａｐｐｌｉｃａｔｉｏｎｏｆａｍｕｌｔｉｖａｒｉａｔｅｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎａｎｄｐｒｅｄｉｃｔｉｏｎｍｅｔｈｏｄｔｏｎｏｎ－ｉｎｖａｓｉｖｅａｎｄｉｎｔｒａｃｒａｎｉａｌｌｏｎｇ－ｔｅｒｍＥＥＧｒｅｃｏｒｄｉｎｇｓ［Ｊ］．ＣｌｉｎｉｃａｌＮｅｕｒｏｐｈｙｓｉｏｌｏｇｙ，２００８，１１９（１）：１９７－２１１．［４］ＢＥＮＢＡＤＩＳＳＲ，ＢＥＮＩＣＺＫＹＳ，ＢＥＲＴＲＡＭＥ，ｅｔａｌ．ＴｈｅｒｏｌｅｏｆＥＥＧｉｎｐａｔｉｅｎｔｓｗｉｔｈｓｕｓｐｅｃｔｅｄｅｐｉｌｅｐｓｙ［Ｊ］．ＥｐｉｌｅｐｔｉｃＤｉｓｏｒｄｅｒｓ，２０２０，２２（２）：１４３－１５５．［５］王学峰．癫癎的脑电图：传统观点㊁新认识和新领域［Ｊ］．中华神经科杂志，２００４，３７（３）：７－９．［６］刘晓燕，黄珍妮，秦炯．不同类型小儿癫痫持续状态的临床及脑电图分析［Ｊ］．中华神经科杂志，２０００，３３（２）：７３－７３．［７］ＭＡＴＵＲＡＮＡＭＩ，ＭＥＩＳＥＬＣ，ＤＥＬＬＫ，ｅｔａｌ．Ｃｒｉｔｉｃａｌｓｌｏｗｉｎｇｄｏｗｎａｓａｂｉｏｍａｒｋｅｒｆｏｒｓｅｉｚｕｒｅｓｕｓｃｅｐｔｉｂｉｌｉｔｙ［Ｊ］．ＮａｔｕｒｅＣｏｍｍｕｎｉｃａｔｉｏｎｓ，２０２０，１１（１）：２１７２．［８］彭睿旻，江军，匡光涛，等．基于ＥＥＧ的癫痫自动检测：综述与展望［Ｊ］．自动化学报，２０２２，４８（２）：３３５－３５０．［９］ＨＯＳＳＥＩＮＩＭＰ，ＨＯＳＳＥＩＮＩＡ，ＡＨＩＫ．ＡｒｅｖｉｅｗｏｎｍａｃｈｉｎｅｌｅａｒｎｉｎｇｆｏｒＥＥＧｓｉｇｎａｌｐｒｏｃｅｓｓｉｎｇｉｎｂｉｏｅｎｇｉｎｅｅｒｉｎｇ［Ｊ］．ＩＥＥＥＲｅｖｉｅｗｓｉｎＢｉｏｍｅｄｉｃａｌＥｎｇｉｎｅｅｒｉｎｇ，２０２０，１４：２０４－２１８．［１０］ＡＣＨＡＲＹＡＵＲ，ＨＡＧＩＷＡＲＡＹ，ＤＥＳＨＰＡＮＤＥＳＮ，ｅｔａｌ．ＣｈａｒａｃｔｅｒｉｚａｔｉｏｎｏｆｆｏｃａｌＥＥＧｓｉｇｎａｌｓ：Ａｒｅｖｉｅｗ［Ｊ］．ＦｕｔｕｒｅＧｅｎｅｒａｔｉｏｎＣｏｍｐｕｔｅｒＳｙｓｔｅｍｓ，２０１９，９１：２９０－２９９．［１１］ＭＵＲＳＡＬＩＮＭ，ＺＨＡＮＧＹ，ＣＨＥＮＹ，ｅｔａｌ．Ａｕｔｏｍａｔｅｄｅｐｉｌｅｐｔｉｃｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎｕｓｉｎｇｉｍｐｒｏｖｅｄｃｏｒｒｅｌａｔｉｏｎ－ｂａｓｅｄｆｅａｔｕｒｅｓｅｌｅｃｔｉｏｎｗｉｔｈｒａｎｄｏｍｆｏｒｅｓｔｃｌａｓｓｉｆｉｅｒ［Ｊ］．Ｎｅｕｒｏｃｏｍｐｕｔｉｎｇ，２０１７，２４１：２０４－２１４．［１２］杨舒涵，李博，周丰丰．基于机器学习的跨患者癫痫自动检测算法［Ｊ］．吉林大学学报（理学版），２０２１，５９（１）：１０１－１０６．［１３］ＺＡＲＥＩＡ，ＡＳＬＢＭ．Ａｕｔｏｍａｔｉｃｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎｕｓｉｎｇｏｒｔｈｏｇｏｎａｌｍａｔｃｈｉｎｇｐｕｒｓｕｉｔ，ｄｉｓｃｒｅｔｅｗａｖｅｌｅｔｔｒａｎｓｆｏｒｍ，ａｎｄｅｎｔｒｏｐｙｂａｓｅｄｆｅａｔｕｒｅｓｏｆＥＥＧｓｉｇｎａｌｓ［Ｊ］．ＣｏｍｐｕｔｅｒｓｉｎＢｉｏｌｏｇｙａｎｄＭｅｄｉｃｉｎｅ，２０２１，１３１：１０４２５０．［１４］吴端坡，王紫萌，董芳，等．基于ａＥＥＧ尖峰和ｃＥＥＧ棘波提取的癫痫发作检测算法［Ｊ］．实验技术与管理，２０２０，３７（１２）：５７－６２．［１５］骆睿鹏，冯铭科，黄鑫，等．脑电信号预处理方法研究综述［Ｊ］．电子科技，２０２３，３６（４）：３６－４３．［１６］ＯＣＢＡＧＡＢＩＲＨＴ，ＡＢＯＡＬＡＹＯＮＫＡＩ，ＦＡＥＺＩＰＯＵＲＭ．ＥｆｆｉｃｉｅｎｔＥＥＧａｎａｌｙｓｉｓｆｏｒｓｅｉｚｕｒｅｍｏｎｉｔｏｒｉｎｇｉｎｅｐｉｌｅｐｔｉｃｐａｔｉｅｎｔｓ［Ｃ］／／２０１３ＩＥＥＥＬｏｎｇＩｓｌａｎｄＳｙｓｔｅｍｓ，ＡｐｐｌｉｃａｔｉｏｎｓａｎｄＴｅｃｈｎｏｌｏｇｙＣｏｎｆｅｒｅｎｃｅ（ＬＩＳＡＴ）．Ｆａｒｍｉｎｇｄａｔｅ，ＵＳＡ：ＩＥＥＥ，２０１３：１－６．［１７］ＢＯＯＮＹＡＫＩＴＡＮＯＮＴＰ，ＬＥＫ－ＵＴＨＡＩＡ，ＣＨＯＭＴＨＯＫ，ｅｔａｌ．ＡｒｅｖｉｅｗｏｆｆｅａｔｕｒｅｅｘｔｒａｃｔｉｏｎａｎｄｐｅｒｆｏｒｍａｎｃｅｅｖａｌｕａｔｉｏｎｉｎｅｐｉｌｅｐｔｉｃｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎｕｓｉｎｇＥＥＧ［Ｊ］．ＢｉｏｍｅｄｉｃａｌＳｉｇｎａｌＰｒｏｃｅｓｓｉｎｇａｎｄＣｏｎｔｒｏｌ，２０２０，５７：１０１７０２．［１８］ＭＵＫＨＯＰＡＤＨＹＡＹＳ，ＲＡＹＧＣ．Ａｎｅｗｉｎｔｅｒｐｒｅｔａｔｉｏｎｏｆｎｏｎｌｉｎｅａｒｅｎｅｒｇｙｏｐｅｒａｔｏｒａｎｄｉｔｓｅｆｆｉｃａｃｙｉｎｓｐｉｋｅｄｅｔｅｃｔｉｏｎ［Ｊ］．ＩＥＥＥＴｒａｎｓａｃｔｉｏｎｓｏｎＢｉｏｍｅｄｉｃａｌＥｎｇｉｎｅｅｒｉｎｇ，１９９８，４５（２）：１８０－１８７．［１９］ＴＡＰＡＮＩＫＴ，ＶＡＮＨＡＴＡＬＯＳ，ＳＴＥＶＥＮＳＯＮＮＪ．ＩｎｃｏｒｐｏｒａｔｉｎｇｓｐｉｋｅｃｏｒｒｅｌａｔｉｏｎｓｉｎｔｏａｎＳＶＭ－ｂａｓｅｄｎｅｏｎａｔａｌｓｅｉｚｕｒｅｄｅｔｅｃｔｏｒ［Ｃ］／／ＥＭＢＥＣ＆ＮＢＣ２０１７：ＪｏｉｎｔＣｏｎｆｅｒｅｎｃｅｏｆｔｈｅＥｕｒｏｐｅａｎＭｅｄｉｃａｌａｎｄＢｉｏｌｏｇｉｃａｌＥｎｇｉｎｅｅｒｉｎｇＣｏｎｆｅｒｅｎｃｅ（ＥＭＢＥＣ）ａｎｄｔｈｅＮｏｒｄｉｃ－ＢａｌｔｉｃＣｏｎｆｅｒｅｎｃｅｏｎＢｉｏｍｅｄｉｃａｌＥｎｇｉｎｅｅｒｉｎｇａｎｄＭｅｄｉｃａｌＰｈｙｓｉｃｓ（ＮＢＣ）．Ｓｉｎｇａｐｏｒｅ：Ｓｐｒｉｎｇｅｒ，２０１８：３２２－３２５．［２０］ＴＡＰＡＮＩＫＴ，ＶＡＮＨＡＴＡＬＯＳ，ＳＴＥＶＥＮＳＯＮＮＪ．Ｔｉｍｅ－ｖａｒｙｉｎｇＥＥＧｃｏｒｒｅｌａｔｉｏｎｓｉｍｐｒｏｖｅａｕｔｏｍａｔｅｄｎｅｏｎａｔａｌｓｅｉｚｕｒｅｄｅｔｅｃｔｉｏｎ［Ｊ］．ＩｎｔｅｒｎａｔｉｏｎａｌＪｏｕｒｎａｌｏｆＮｅｕｒａｌＳｙｓｔｅｍｓ，２０１９，２９（４）：１８５００３０．２３１智㊀能㊀计㊀算㊀机㊀与㊀应㊀用㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀㊀第１４卷㊀。

Noise-Induced Synchronization among Sub-RF CMOS Neural Oscillators forSkew-Free Clock DistributionAkira Utagawa,Tetsuya Asai,Tetsuya Hirose,and Yoshihito AmemiyaGraduate School of Information Science and Technology,Hokkaido UniversityKita14,Nishi9,Kita-ku,Sapporo060-0814,JAPANEmail:**********************.hokudai.ac.jpAbstract—A possible idea is presented here for deal-ing with clock skew problems on synchronous digital sys-tems.Nakao et al.recently reported that independent neu-ral oscillators can be synchronized by applying temporal random impulses to the oscillators[1].We regard neural oscillators as independent clock sources on LSIs;i.e.,clock sources are distributed on LSIs,and they are forced to syn-chronize through the use of random noises.We designed neuron-based clock generators operating at sub-RF region (<1GHz)by modifying the original neuron model to a new model that is suitable for CMOS implementation with 0.25-µm CMOS parameters.Through circuit simulations, we demonstrate that the clock generators are certainly syn-chronized by pseudo-random noises.1.IntroductionSynchronous sequential circuits with global clock-distribution systems are the mainstream of implementation in present digital VLSI systems where the clock distribu-tion is the core of synchronous digital operations.Practical clocks given through external pads are distributed to se-quential circuits being synchronous to the same clocks via distributed clock networks.System clocks for synchronous digital circuits must arrive at all the registers simultane-ously.In practice,time mismatches of clock arrival which are called‘clock skew’occur in LSIs[2].The major rea-sons for these mismatches derive from the system clock distribution(wiring defects or asymmetric clock paths),the propagation delay of the clock chip,and the clock traces on the board.The propagation delay is dependent on the fab-rication process,voltage,temperature,and loading,which makes the clock skew even more complicated.Small clock skews prevent us from increasing the clock frequency,and large skews may result in severe malfunctions.Indeed clock-skew effects on the circuit performance rise as the in-tegration density(∼miniaturization)or the clock frequency increases.To resolve these clock-skew issues,various technolo-gies on clock distribution are widely used in present digi-tal systems such as zero-skew clock distribution[3],insert-ing buffers for skew compensation[4]and controlling the clock-wire length[5].In regular circuit structures,clock skews are effectively reduced by designing clock paths based on H trees(see[6]for details including statistical analysis).For large-scale complex clock networks,opti-mizing buffers in the clock distribution tree usually reduces clock skew.One possible way to cancel clock skew is to use asynchronous digital circuits where only local clocks are used instead of global system clocks[7].However, the functions of these circuits currently cannot satisfy var-ious sophisticated demands.Moreover,major LSI design-ers have recently started using advanced genetic algorithms in their post-manufacturing processes to calculate the re-quired margin[8].The present solutions for the skew problems may in-crease both the total length of clock distribution wires and the power consumption,as well as optimization and post-processing costs.In this paper,we propose another solution for the skew problems.Nakao et al.recently reported that independent neural oscillators can be synchronized by ap-plying appropriate noises to the oscillators[1].We here regard neural oscillators as independent clock sources on LSIs;i.e.,clock sources are distributed on LSIs,and they are forced to synchronize with the addition of artificial (or natural if possible)noises.In the following sections, we show a modified neuron-based model that are suitable for hardware implementation,neuron-based clock genera-tor for sub-RF operations(<1GHz),and circuit simulation results representing synchronous(or asynchronous)oscil-lations with(or without)external noises.2.The ModelIn the original model[1],FitzHugh-Nagumo neuron was used to demonstrate the noise-induced synchronization be-tween the time courses of N trials under different initial conditions.Instead we use N Wilson-Cowan oscillators in our model that are suitable for analog CMOS implementa-tion.The dynamics are given bydu idt=−u i+fβ(u i−v i),(1) dv idt=−v i+fβ(u i−θ)+I(t),(2)where u i and v i represent the system variables of the i-th os-cillator,θthe threshold,I(t)the common temporal random impulse and fβ(·)the sigmoid function with slopeβ.2007 International Symposium on Nonlinear Theory and its A pplications NOLTA'07, Vancouver, Canada, September 16-19, 20070.510.5 1vuu nullcline v nullcline trajectoryFigure 1:Nullclines and trajectories of single Wilson-Cowan type oscillator receiving random impulses.0 0.5 1 160170180 190200utime0 0.51 160170180 190200vtimeFigure 2:Time courses of system variables of single Wilson-Cowan type oscillator receiving random impulses.Figure 1shows numerical simulation results of a single Wilson-Cowan oscillator receiving temporal random im-pulses given by I (t )=α j δ(t −t (1)j )−δ(t −t (2)j )where δ(t )=Θ(t )−Θ(t −w )(Θ,w and t j represent the step func-tion,the pulse width and the positive random number witht (1)j t (2)j for all j s,respectively).The system parameters were θ=0.5,β=10,α=0.1,w =1,and the averaged inter-spike interval of |I (t )|was set at 100.We observed the limit-cycle oscillations,and confirmed that the trajectory was certainly fluctuated by I (t ).The time courses of u and v are shown in Fig.2.We conducted numerical simulations using 10oscillators (N =10).All the oscillators have the same parameters,and accept (or do not accept)the common random impulse I (t ).The initial condition of each oscillator was randomly chosen.Figure 3shows the raster plots of 10oscillators (vertical bars were plotted at which u i >0.5and du i /dt >0).When the oscillators did not accept I (t )(α=0),they exhibited independent oscillations as shown in Fig.3(a);1510 240024502500 25502600o s c i l l a t o r n o .time(b) with noise1510 240024502500 25502600o s c i l l a t o r n o .time(a) without noiseFigure 3:Raster plots of 10oscillators.(a)independent oscillations without random impulses,(b)synchronous os-cillations with random impulses.0 0.5 1 0100020003000 40005000R (t )time(b) with noise0 0.5 1 010002000300040005000R (t )time(a) without noiseFigure 4:Time courses of order parameter values (a)with-out random impulses and (b)with random impulses.however,all the oscillators were synchronized when α=0.1as shown in Fig.3(b).To evaluate the degree of the synchronization,we use the following order parameter:R (t )=1N j exp(i θj ) ,where N represents the number of oscillators,i the imagi-nary unit and θj =tan −1[(v j −v ∗)/(u j −u ∗)]((u ∗,v ∗)repre-sents the fixed point of the oscillator).When all the oscilla-tor are synchronized,R (t )equals 1because of the uniformθj s,while R (t )is less than 1if the oscillators are not syn-chronized.Figure 4shows the time courses of the order parameter values.When α=0,R (t )was unstable and was always less than 1[Fig.4(a)],whereas R (t )remained at 1after it became stable at t ≈2000when α=0.1[Fig.4(b)].These results indicate that if we implemented these oscilla-tors as clock generators on CMOS LSIs,applying common random pulses to the oscillators could synchronize them.Figure 5:Wilson-Cowan circuit for sub-RF operations.0.5 1 1.5 2 2.5 00.511.5 22.5v (V )u (V)u nullcline v nullcline trajectoryFigure 6:Nullclines and trajectories of oscillator circuit receiving pseudo-random impulse.3.The circuit and simulation resultsWe designed a Wilson-Cowan oscillator circuit for sub-RF operations (Fig.5).The circuit consists of a di fferen-tial pair (M1to M3)and a bu ffer circuit composed of two standard inverters.In the following simulations,we used TSMC’s 0.25-µm CMOS parameters with W /L =0.36µm /0.24µm except for M3’s channel length (L =2.4µm).Pseudo-random sequences (V mseq )were generated using a 4-bit M-sequence circuit,and were distributed to the circuit through a RC filter.The supply voltage was fixed at 2.5V .Figure 6shows SPICE results of the nullclines and tra-jectories receiving random impulses (C =10fF,R =100k Ω,the clock frequency of the M-sequence circuit was 50MHz,which resulted in a 300-ns pseudo-random se-quence).Time courses of u and v are shown in Fig.7.We observed qualitatively-equivalent nullclines and trajec-toriesto those of the Wilcon-Cowan oscillators.We con-firmed the limit-cycle oscillations where the trajectory was e ffectively fluctuated by the M-sequence circuit with the0 0.5 1 1.5 2 2.5 5051 52 5354 55 56 57 58 59 60u (V )time (ns)0 0.5 1 1.5 2 2.5 5051 52 5354 55 56 57 58 59 60v (V )time (ns)Figure 7:Time courses of system variables of oscillator circuit receiving pseudo-random impulses.1510 1.951.961.97 1.981.992o s c i l l a t o r n o .time (µs)(b) with noise1510 1.951.961.97 1.981.992o s c i l l a t o r n o .time (µs)(a) without noiseFigure 8:Raster plots of 10oscillator circuits.(a)in-dependent oscillations without random impulses,(b)syn-chronous oscillations with random impulses.RC filter.The oscillation frequency was about 1GHz when the reference voltage V ref was set at 1V .Figure 8shows the raster plots of 10oscillator circuits (vertical bars were plotted at which v i >1.25V and dv i /dt >0).All the cir-cuits exhibited independent oscillations when random se-quence V mseq was not given to them [Fig.8(a)],whereas they exhibited complete synchronization when V mseq was given [Fig.8(b)].Time courses of the order parameter val-ues were shown in Fig.9.When random impulse was not given to the circuit,R (t )was not stable and was always less than 1[Fig.9(a)],while R (t )remained at 1after it be-came stable at t ≈700µs when random impulse was given [Fig.9(b)].Our results indicate that if we distributed these circuits as ubiquitous clock sources on CMOS LSIs,they could be synchronized when common random impulses were given to the circuits.Although this may cancel out the present clock skew problems,device mismatches between0 0.5 1 012 345R (t )time (µs)(b) with noise0 0.5 1 012 345R (t )time (µs)(a) without noiseFigure 9:Time courses of order parameter values (a)with-out random impulses and (b)with random impulses.0.90.92 0.94 0.96 0.98 1 0 0.5 11.5 22.5 3〈 R (t ) 〉σ (mV)Figure 10:Synchrony dependence on parameter mismatch.the clock sources may prevent the sources from complete synchronization.Therefore,we investigated the device-mismatch dependence of the proposed circuits.For our dis-tributing purposes,local mismatches in a single oscillator circuit would be negligible;i.e.,mismatches in a di fferen-tial pair (M1and M2)and a current mirror.Mismatches in inverters corresponding to threshold θin Wilson-Cowan model would also be negligible because they only shift the fixed point,and do not vastly change the oscillation fre-quency.However,mismatches of M3between the oscilla-tors may drastically change each oscillator’s intrinsic fre-quency.Therefore,we distributed threshold voltages of M3s of all the oscillators.Zero-bias threshold voltages (VTO)of M3s were randomly chosen from the Gaussian distribution (mean:0.37V and standard deviation:σ).Fig-ure 10shows the dependence of averaged order-parameter values 〈R (t )〉(from 0to 1µs)on σ.We generated 10ran-dom VTO sets for each σ,and plotted the error bars and the mean values in the figure.We confirmed that 〈R (t )〉was gradually decreased when σwas increased.4.ConclusionWe designed CMOS sub-RF oscillators that could be synchronized using common random impulses,based on a theory in [1].We proposed a modified Wilson-Cowan model for implementing FitzHugh-Nagmo oscillators.We confirmed that the synchronization properties of the mod-ified model were qualitatively equivalent to those of the original model.We then designed sub-RF oscillator cir-cuits based on the modified model.Through circuit simula-tions,we demonstrated that the circuits exhibited the same synchronization properties as in the original and modified models.For our clock-distributing purposes,we investi-gated the synchrony dependence on device mismatches be-tween the distributed oscillator circuits.The result showed that the synchrony was gradually decreased when variance of the mismatch was linearly increased,which indicated that our ‘ubiquitous’clock sources with small device mis-matches would be synchronized by optimizing our param-eter sets.References[1]H.Nakao,K.Arai,and K.Nagai,“Synchrony oflimit-cycle oscillators induced by random external impulses,”Phys.Rev.E vol.72,026220,2005.[2]D.E.Brueske and S.H.K.Embabi,“A dynamic clocksynchronization technique for large systems,”IEEE p.,Packag.,Manufact.Technol.B ,vol.17,pp.350-361,1994.[3]R.S.Tsay,“An exact zero-skew clock routing algo-rithm,”IEEE Trans.on Comp.-Aided Design of Inte-grated Cir.Syst.,vol.12,no.2,pp.242-249,1993.[4]R.B.Watson,Jr.,and R.B.Iknaian,“Clock bu fferchip with multiple target automatic skew compensa-tion,”IEEE J.Solid-State Circuits ,vol.30,pp.1267-1276,1995.[5]T.-H.Chao,Y .-C.Hsu,J.-M.Ho,and A.B.Kahng,“Zero skew clock routing with minimum wirelength,”IEEE Trans.Circuits and Systems II ,vol.39,no.11,pp.799-814,1992.[6]M.Hashimoto,T.Yamamoto,and H.Onodera,“Sta-tistical analysis of clock skew variation in H-tree structure,”IEICE Trans.on Fundamentals of Elec-tronics,Communications and Computer Sciences ,vol.E88-A,no.12,pp.3375-3381,2005.[7]C.J.Myers,Asynchronous Circuit Design ,Wiley-Interscience,2001.[8]E.Takahashi,Y .Kasai,M.Murakawa and T.Higuchi,“Post-fabrication clock-timing adjustment using ge-netic algorithms,”IEEE J.Solid-State Circuits ,vol.39,no.4,pp.643-649,2004.。

257ISSN: 1749-3889 ( print ) 1749-3897 (online)BimonthlyVol.7 (2009) No.3JuneEngland, UK ***************************.uk International Journal ofN o n l i n e a r S c i e n c e Edited by International Committee for Nonlinear Science, WAUPublished by World Academic Union (World Academic Press)CONTENTS259.New Exact Travelling Wave Solutions for Some Nonlinear Evolution EquationsA. Hendi268.A New Hierarchy of Generalized Fisher Equations and Its Bi- Hamiltonian StructuresLu Sun274.New Exact Solutions of Nonlinear Variants of the RLW, the PHI-four and Boussinesq Equations Based on Modified Extended Direct Algebraic MethodA. A. Soliman, H. A. Abdo283. Monotone Methods in Nonlinear Elliptic Boundary Value ProblemG.A.Afrouzi, Z.Naghizadeh, S.Mahdavi290. Influence of Solvents Polarity on NLO Properties of Fluorone Dye Ahmad Y. Nooraldeen1, M. Palanichant, P. K. Palanisamy301.Projective Synchronization of Chaotic Systems with Different Dimensions via Backstepping DesignXuerong Shi, Zhongxian Wang307.Adaptive Control and Synchronization of a Four-Dimensional Energy Resources System of JiangSu ProvinceLin Jia , Huanchao Tang312.Optimal Control of the Viscous KdV-Burgers Equation Using an Equivalent Index MethodAnna Gao, Chunyu Shen，Xinghua Fan319.Adaptive Control of Generalized Viscous Burgers’ Equation Xiaoyan Deng, Wenxia Chen, Jianmei Zhang327. Wavelet Density Degree of a Class of Wiener Processes Xuewen Xia, Ting Dai332.Niches’ Similarity Degree Based on Type-2 Fuzzy Niches’ Model Jing Hua, Yimin Li340.Full Process Nonlinear Analysis the Fatigue Behavior of the Crane Beam Strengthened with CFRPHuaming Zhu, Peigang Gu, Jinlong Wang, Qiyin Shi345.The Infinite Propagation Speed and the Limit Behavior for the B-family Equation with Dispersive TermXiuming Li353.The Classification of all Single Traveling Wave Solutions to Fornberg-Whitham EquationChunxiang Feng, Changxing Wu360.New Jacobi Elliptic Functions Solutions for the Higher-orderNonlinear Schrodinger EquationBaojian Hong, Dianchen Lu368.A Counterexample on Gap Property of Bi-Lipschitz Constants Ying Xiong, Lifeng Xi371.A Method for Recovering the Shape for Inverse Scattering Problem of Acoustic WavesLihua Cheng, Tieyan Lian, Ping Li379.An Approach of Image Hiding and Encryption Based on a New Hyper-chaotic SystemHongxing Yao, Meng Li258International Journal of Nonlinear Science (IJNS)BibliographicISSN: 1749-3889 (print), 1749-3897 (online), BimonthlyEdited by International Editorial Committee of Nonlinear Science, WAUPublished by World Academic Union (World Academic Press)Publisher Contact:Academic House, 113 Mill LaneWavertree Technology Park, Liverpool L13 4AH, England, UKEmail:******************.uk,*********************************URL: www. World Academic Union .comContribution enquiries and submittingThe paper(s) could be submitted to the managing editor ***************************.uk. Author also can contact our editorial offices by mail or email at addresses below directly.For more detail to submit papers please visit Editorial BoardEditor in Chief: Boling Guo, Institute of Applied Physics and Computational Mathematics, Beijing, 100088, China;************.cnCo-Editor in Chief: Lixin Tian, Nonlinear Scientific Research Center, Faculty of Science, Jiangsu University, Zhenjiang, Jiangsu,212013;China;**************.cn,************.cnStanding Members of Editorial Board:Ghasem Alizahdeh Afrouzi, Department ofMathematics, Faculty of Basic Sciences, Mazandaran University,Babolsar,Iran;**************.ir Stephen Anco, Department of Mathematics, Brock University, 500 Glenridge Avenue St. Catharines, ON L2S3A1,Canada;***************Adrian Constantin ,Department of Mathematics, Lund University,22100Lund,****************.seSweden;*************************.se,Ying Fan, Department of Management Science, Institute of Policy and Management, ChineseAcademy of Sciences, Beijing 100080,China,**************.cn.Juergen Garloff, University of Applied Sciences/ HTWG Konstanz, Faculty of Computer Science, Postfach100543, D-78405 Konstanz, Germany;************************Tasawar Hayat, Department of mathematics,Quaid-I-AzamUniversity,Pakistan,*****************Y Jiang, William Lee Innovation Center, University of Manchester, Manchester, M60 1QD UK;*******************Zhujun Jing,Institute of Mathematics, Academy of Mathematics and Systems Sciences, ChineseAcademy of Science, Beijing, 100080,China;******************Yue Liu,Department of Mathematics, University of Texas, Arlington,TX76019,USA;************Zengrong Liu,Department of Mathematics, Shanghai University, Shanghai, 201800,China;******************.cnNorio Okada, Disaster Prevention Research Institute, KyotoUniversity,****************.kyoto-u.ac.jp Jacques Peyriere,Université Paris-Sud, Mathématique, bˆa t. 42591405 ORSAY Cedex , France;************************,****************************.frWeiyi Su, Department of Mathematics, NanjingUniversity, Nanjing，Jiangsu, 210093,China;*************.cnKonstantina Trivisa ,Department of Mathematics, University of Maryland College Park，MD20472-4015，USA;****************.eduYaguang Wang,Department of Mathematics, Shanghai Jiao Tong University, Shanghai， 200240，China;***************.cnAbdul-Majid Wazwaz, 3700 W. 103rd Street Department of Mathematics and Computer Science, Saint Xavier University, Chicago, IL 60655 ,USA;**************Yiming Wei, Institute of Policy and management, Chinese Academy of Science, Beijing, 100080,China;*****************Zhiying Wen,Department of Mathematics, Tsinghua University, Beijing, 100084, China;*******************Zhenyuan Xu, Faculty of Science, Southern Yangtze University, Wuxi , Jiangsu 214063 ,China;*********************Huicheng Yi n, Department of Mathematics, Nanjing University, Nanjing， Jiangsu, 210093, China;****************.cnPingwen Zhang, School of Mathematic Sciences, Peking University, Beijing, 100871, China;**************.cnSecretary: Xuedi Wang, Xinghua FanEditorial office:Academic House, 113 Mill Lane Wavertree Technology Park Liverpool L13 4AH, England, UK Email:***************************.uk **************************.uk ************.cn。

*国家自然科学基金资助项目(63,66);广东省自然科学基金资助项目(5)通信作者z y @y 基于近似熵的癫痫发作预测研究*刘治远1,解玲丽2,陈子怡3,黄瑞梅1,李小江1,周毅1(1.中山大学中山医学院生物医学工程系,广州510080;2.中山大学数学与计算科学学院,广州510275;3.中山大学附属第一医院神经内科,广州510080)摘要:临床采用诱发方法检测获得的失神发作患者EEG 信号,研究其发作前脑电信号的动力学变化的规律,寻找预测癫痫失神发作一般规律和方法。

我们选择合适的电极对,使用非线性动力学的方法,采用复杂度变化度量的近似熵指标,通过闪光刺激癫痫患者获得的EEG 信号进行动力学特征研究,根据EE G 信号表现出的同步情况实现对癫痫发作的预测。

结果表明,本研究可以实现对临床光刺激进行诱发的癫痫失神患者的发作进行预测。

研究中采用优化电极的方法优于采用固定电极对。

关键词:癫痫;脑电图;T 检验;近似熵;发作预测中图分类号:R318文献标识码:A 文章编号:1672-6278(2011)01-0020-04Prediction of Epileptic Seizure based on Approximate Entropy of EEGLIU Zhiyuan 1,XIE Lingli 2,CHEN Ziyi 3,HUA NG Ruimei 1,LI Xiao jiang 1,ZHOU Yi1(1.D epartment o f Biomedical Engineering ,Zhon gs han Schoo l o f M edicine,Sun Y at -s en U nive rsity,G uan gzho u 510080,China ;2.School o f Mathematics and Com putational Science,Sun Y at -s en U niv ersity,G uan gzho u 510275;3.De partment o f Neu r o lo gy,the F irs t A ff liate d Hos pital,Sun Y at -sen U niv ersity,G ua n gzhou 510080)Abstr act:In this study,we try to find out the general regularity and method to predict absences seizure by studying regularities of dynamics of spatiotemporal transitio ns of EEG during epilep tic seizures.With no nlinear dynamical method,using Appro ximate Entropy (A pEn),w e s tudied the dy namical features o f EEG signal w hich w as o btained fro m clinical data.The selection of cri tical cortical cites involved a model w hich could opti mize the pro bability to be best.Before epileptic absence seizure,the T-index of those critical cites will change to so me deg ree and will progressively conv erge.Then,a pro bably predictio n can be given.The prediction scheme o f opti ma electrode is better than using fi xed electrodeduring tw o seizures is considerable.Key wor ds:Epilepsy;Electroencephalog ram(EEG);T-test;Appro xi mate entropy;Prediction of seizures1引言癫痫患者一般有两种不同的状态:表现正常的发作间期和发作期,有些病例还含有明显的发作前期和发作后期。

Action ....................... 动作Animator .................. 原画者，动画设计Assista .................... 动画者Antic ....................... 预备动作Air Brushing …喷效Angle .............................. 角度Animated Zoom ……画面扩大或缩小Animation Film......................... ........ 动画片Animation Computer …电脑控制动画摄影Atmosphere Sketch .............. 气氛草图B．P．(Bot Pegs) ................... 下定位Bg(Background) ................... 背景Blurs ............................. 模糊Blk(Blink) ..................眨眼Brk Dn(B．D．)(Break-Down) …中割Bg Layout .............................. 背景设计稿Background Keys ................... ...背景样本Background Hookup ............. 衔接背景Background Pan ...................... 长背景Background Still 短背景Bar Sheets ............................ 音节表Beat................... 节拍Blank ..................空白Bloom ............................ 闪光Blow Up ..............................放大Camera Notes ................. 摄影注意事项C．U．(Close-Up) …特写Clean Up .............. 清稿，修形，作监Cut ............................ 镜头结束Cel=Celluloid ............................. 化学板Cycle ................................ 循环Cw(Clock-Wise) …顺时针转动Ccw(Counter Clock-Wise) …逆时针转动Continue(Cont ，Con‘D)…继续Cam(Camera) ................. 摄影机Cush(Cushion) ……缓冲C=Center ................ 中心点Camera Shake ……镜头振动Checker ................... ....... 检查员Constant .................... 等速持续Color Keys=Color Mark-Ups 色指定Color Model ................... 彩色造型Color Flash(Paint Flash) …跳色Camera Animation………动画摄影机Cel Level .......................化学板层次Character .................... 人物造型Dialog (Dialogue ............... 双重曝光Multi Runs ............. 多重曝光1st Run .................. 第一次曝光2nd Run........................ 地二次曝光Dry Brushing ……干刷Diag Pan(Diagonal) .................. 斜移Dwf(Drawing) .......................... 画，动画纸Double Image ............... 双重影像Dailies (Rushes) ……样片Director ................................. 导演Dissolve(X. D) .................... ......溶景，叠化Distortion ............................................. 变形Double Frame ......................................... 双（画）格Drawing Disc .................................... 动画圆盘E.C. U = Extreme Close Up 大特写Ext(Exterior) ................ ... 外面；室外景Eft(Effect) .......................... 特效Editing ......................... 剪辑Exit(Moves Out, O. S. ) …出去Enter(In) ................. 入画Ease-In....................... ... 渐快Ease-Out .................. 渐慢Editor...................... 剪辑师Episode ……片集Field(Fld) .............................. 安全框Fade(In/On) ……画面淡入Fade(Out/Off) ……画面淡出Fin(Finish) ..................... 完成Folos(Follows) …跟随，跟着Fast; Quickly ……快速Field Guide ……安全框指示Finial Check ........... ...... 总检Footage .................... 尺数（英尺）F.G. (Foreground)…前景Focal Length ……焦距Frame …格数Freeze Frame ..................... 停格Gain In ……移入Head Up ............... 抬头Hook Up ...................... 接景；衔接Hold ...............画面停格Halo ............................... 光圈Int(Interior) ......................... 里面；室内景Inb(In Between) .................................. 动画In-Betweener ……动画员I&P(Ink & Paint) …描线和着色Inking ..................描线In Sync .................... 同步Intermittent ..................... 间歇Iris Out ..................... 画面旋逝Jiggle .................. 摇动Jump …跳Jitter ................ 跳动Lip Sync(Synchronization) 口形Level ........................... 层Look ……看Listen ........................ 听Layout .......................... 设计稿；构图Laughs(Laffs) ……笑L/S(Light Source) ……光源Line Test(Pencil Test) …铅笔稿试拍；线拍M. S. (Medium Shot) ..................... 中景M. C. U. (Mediium Close Up) …近景Moves Out(Exit; O. S. ) ...... ........... 出去Moves In ................................... 进入Match Line ......................... 组合线Multi Runs ................. 多重拍摄Mouth ............................. 嘴Mouth Charts ........................ 口形图Mag T rack(Magnetic S ound T rack) 音轨Multicel Levels …多层次化学板Multiplane ....................... 多层设计N／S Pegs ...................... 南北定位器N．G．(No Good) ..................... 不好的，作废Narration ……旁白叙述Ol(Overlay) ...................... .. 前层景Out Of Scene ................. 到画外面O．S．(Off Stage Off Scene) …出景Off Model .......................... 走型Ol／Ul(Underlay) 前层与中层间的景Overlap Action …重叠动作Ones ...................... 一格；单格Pose ...................... 姿势Pos(Position) ……位置；定点Pan ........................... 移动Pops In ／On ..................... 突然出现Pause ....................... 停顿；暂停Perspective ……透视Peg Bar ...................... 定位尺P．T．(Painting) ...... ........... 着色Paint Flashes(Color Flashes) 跳色Papercut ................... ...... 剪纸片Pencil Test ............. 铅笔稿试拍Persistence Of Vision 视觉暂留Post-Synchronized Sound后期同步录音Puppet .......................... ...... 木偶片Ripple Glass ................... 水纹玻璃Re-Peg ............................ 重新定位Ruff(Rough-Drawing) …草稿Run ...................... 跑Reg(Register) ..................... .组合Rpt(Repeat) ................................. 重复Retakes ...................... 重拍；修改Registration Pegs ……定位器Registration Holes ……定位洞Silhouette(Silo) .................... 剪影Speed Line ................... 流线Storm Out ............................. 速转出Sparkle ....................... 火花；闪光Shadow ................ 阴影Smile ....................... 微笑Smoke ……烟Stop .............................. 停止Slow ...................... ..... 慢慢的Sc(Scene) .......................... 镜号S／A(SameAs)............................... 兼用S．S(Screen Shake) …画面振动Size Comparison ……大小比例Storyboard(Sab) …分镜头台本Sfx(Sound Effect) …声效；音效Settle ..................................... 定姿；定置Self-Line(Self-Trace Line) 色线Sound Chart(Bar Sheeets) 音节表Special Effect ....................... 特效Spin ................................. 旋转T．A．(Top Aux) .............. 上辅助定位T．P．(Top Pegs) .............. 上定位Track ........................................................ 声带Turns ........................................................ 转向Take …………拍摄(一般指拍摄顺序) Truck I n................................................ 镜头推人Truck Out ................................ 镜头拉出Tr(Trace) .................同描Tapers ................................. 渐Taper-Up ......................... 渐快Taper-Down .................... 渐慢Tight Field .......................... 小安全框Tap(Beat) ...................节拍Tittle ..................... 片名；字幕Ul(Underlay) ............................ 中景；后景Up .................... 上面Use ....................... 用Vert Up ................... 垂直上移V.O. (V oice Over) …旁白；画外音Value .................................... 明暗度Wipe .........................转（换）景方式Work Print …工作样片X(X-Diss) (X. D. ) ……两景交融Xerox Down ................ .. 缩小Xerox Up(Xerox Paste-Ups) 放大X-Sheet ...................... 摄影表Zoom Out ........................... 拉出Zoon Chart.................镜头推拉轨迹Zoom In ……推进Zoom Lens …变焦距镜头MMagnetic Tape 磁性录音带Makeup A rtist 美容师Manipulation 操纵Markup 固定利润Matte 影像形板Maysles Films 梅思利电影公司Memory-Hook 回马枪Memory-Jogger 回马枪Merrill Lynch 美林动画Metamorphic A nimation 变形动画Metamorphosis 变形Micro-Markets 微众市场Mixer 混音师Modeling 模型制作Montage 蒙太奇Morph 型变Mos 不需要现场收音的无声取景Motion Board 活动脚本或动作脚本Motion Capture 动作资料截取Motion Cintrol 电脑控制拍摄系统Motion Picture Film 动画影片Motion Tests 动作测试Motor Home 移动居住车Mouse 滑鼠Mouthpiece 发言人Multi-City Bidding 多城市竟标Music Bookends 音乐书签Music First 以音乐为优先Musical Instrument Digital Interface Midi电子乐器一的数位介面NNational Association O f Broadc国家广播电子技师协会National Cash Register 国家收银机公司Nbc 国家广播公司Negative Conformer 底片组合员Ng 不好的镜头Nonlinear Editing 非线性剪辑OOfff-Camera 镜外表演Off-Key 走调Offline System 线外系统Offline System 线外剪辑系统One-Stop Operation 一贯作业On Camera 镜内表演On-Camera Sag Rates 演员同业公会规定的上镜费On Location 出外景Online Editing 线上剪辑One-Light 单一光度One-Light Film Print 单光影片洗印One-Stop Operation 一次作业Opaquer 著色人员Open Camera 公开摄影Optical House 视觉效果工作室Optical Printer 光学印片室Original Arrangment 编曲著作Original Recording 录音著作Original Score 总谱制作Out-Of-Pocket 现款支付Outside Props 棚外道具师Outtakes 借用镜头PPacific Data Images 太平洋影像公司Pegs 过场用之画面Pencil Test 铅笔测试稿Perceived Value 知觉价值Personalities 知名人士Personality Testimonials 名人见证Petsuasion 说服Photo Cd 影像光碟Pickup Footage 从旧有的广告借凑而来的影片Pictures First 以画面为优先Pixels 像素Playback 播放Playback Person 录影机播放员Post-Scoring 后制配乐Posttesting 后测Pre-Lite 预先排演Pre-Production Meeting 拍制前会议Pre-Production Stage 制前阶段Prescoring Music 拍摄前配乐Pretesting 前测Price-Quote 报价或喊价Printed Circuiry 印刷电路Producer 广告公司的制片，制作人Product Shot 商品展示镜头Production Assistant P .A 制作助理Production Boutique 制片工作室Production Notes 制作住记Production Package 制作议价组合Production Specification Sheets 制作分工明细表Promotions 促销Prop People 道具师Peoperties 舞台道具Props 道具Public-Domain Music大众共有或版权公有的音乐Publisher's Fee 发行费用Pulldowm 抓片RRanddom A ccess 随机存取Random Access Memory Ra随m机存取记忆体Raster 屏面Read Only Memory Rom 唯读记忆体Real Opinions 真实反应的意见Real People 消费大众或一般人Real People Reactions And Opinions消费大众的真实反应及意见Recordist 录音师Reebok 锐跑Reflections 反光Rendering 算图Rental Facilities 出租公司Residual 后续付款Rhapsody In Blue 《蓝色狂想曲》Rhythm And Hues 莱休电脑动画公司Right-To-Work 自由工作权Ripomatic/Stealomatic Storyboard 借境脚本Roll Camera 开动摄影机Rotoscope 逐格帖合的重覆动画动作Rough Cut 粗剪SSample Reels 作品集Scencs 场景Scenics A rtist 布景设计师Scratch Track 临时音轨Screen Actors Guild Sag 电影演员同业公会Screen Extra's Guild Sag 电影临时演员同业公会Scripts 剧本Script Clerk 场记Set Construction Costs 搭景费用Set Designer 布景设计师Set Dresser 布影装饰师Shadows 阴影Shape Library 清晰对焦Shooting Board 模型资料库Shooting Day 制作脚本拍片日Shooting In Two 一次两画格的方式拍摄Shot List 拍摄程序表Shutter 快门Sides 台词表Silent Scenes 无声场景Silent Takes 无声取景Slate 开拍板Slice-Of-Life Episodes 生活片段式对白Snapshot 快照拍摄Solid State Screensound 数位录音工作站Song-And-Dance 歌舞片Sound People 音效人员Sound Stage 隔音场Sound Take 有声摄影Special Effects Person 特殊效果人员Special-Effects 特效Specification Sheet 职责明细表Speed 运转正常Splice 捻接Sprint斯布林特电话公司Stand-In 替身Stand-Up Presenters 播报员推荐Standing Sets 常备的布景配置Star Personality 知名人物Stereo-Mixing 立体声混音Sticks 排字手托Stills 剧照Still Photos 静态照片Stock Footage 底片材料、库存影片Stop-Motion 单格拍制Story Line 故事情节Storyboard 故事脚本Strobe-Lighe Photography 频闪闪光灯摄影法Subaru Autombile 速霸陆汽车Super 16mm Format 超16 厘米底片规格Sync Sound 同步收音Synchronized 同步TThe Screening Room 试播室Takes 取景镜头Talent Reports 劳务报价单Teamsters 卡车驾驶员Teamsters Union 卡车驾驶员工会Telepromrter 读稿机Test Commercial 测试性广告Testimonial Release Print 电影院放映片Three-Dimensional 3d 三度空间Ight Close-Up 大特写Time-Code 时码Tissue S heets 薄绵纸Top Light 顶光Trim 剪修Trims 修剪下来的片头尾Tracing Paper 扫图纸Track Left 摄影机左移Track R ight 摄影机右移Track Time 音轨时限Trade 通路Tri-X 柯打tri-X 底片Turnarounds 转场Unique Selling Proposition 独特的销售主张VVideo 视觉或影像部分Video Master 影像母带Video Tape Recording Person 录音带录制员Vignetters 集锦式快接画面处理Virtual Reality 虚拟实境Visual Timeline 视觉时间尺Visually Oriented 视觉导向Voiceover Announcer 旁白播音员WWardrobe Attendant 服装师West And Brady 威布广告公司Wild Wall 活动墙板Window Burn-In 叠印框Wire-Frame 立体线稿Words-And-Music 旁白加音乐Words First 以文案为优先Zoom 变焦Zoom In 镜头向前推进。

专业英语Specialized English一、将词组译成英语信道容量Channel capacity信息量Amount of information信号功率Signal power噪声功率Noise power噪声谱密度Noise spectral density通信保真Fidelity of communication光波系统Optical system中继距离Distance spacing半导体激光器Semiconductor laser光纤放大器Fiber amplifier波分复用Wavelength-division multiplexing 光纤损耗Fiber loss光纤色散Fiber dispersion掺饵光纤放大器Erbium-doped fiber amplifier同步数字系列Synchronous digital hierarchy 支路信号Tributary signals数字交叉连接Digital cross connect网络维护Network maintenance支路映射Tributary mapping同步传输帧Transmission frame线路终端复用器Line Terminal Multiplexer灵敏度Sensitivity虚容器Virtual container成帧字节framing bytes段开销Section overhead端到端传输End of transmission误码监视Error monitoring信号处理节点Signal processing node净负荷Net load指针Pointer离线率The rate of off-time软交换Soft switching功率谱密度Power spectral density开环功率控制Open loop power control抗干扰能力Anti jamming ability拦截Blocking rate虚电路Virtual circuit时隙Time slot时分复用Time division multiplexing局域网Local area network服务质量Service quality广域网Wide area network公众交换电话网Public switched telephone network分组交换Packet switching蓝牙规范Bluetooth specification免提电话 A hands-free phone通用接入框架Universal access framework接入控制协议Access control protocol业务发现协议Service Discovery Protocol立体声耳机Stereo headset网络电视Network television数字用户线接入复用器Digital subscriber line access multiplexer 视频点播Video on demandIP组员协议IP crew agreement机顶盒The set-top box前向纠错Forward error correction高清电视High definition television实时流协议Real time streaming protocol通信信道Communication channel光发送机Optical transmitter光接收机Optical receiver光脉冲Optical pulse光源Light source非线性效应Nonlinear effect信噪比Signal to noise ratio误码率Bit error rate强度调制直接检验Intensity modulation direct inspection嵌入式系统Embedded system特定用途集成电路Application specific integrated circuit数字助理Digital assistant通信协议Communication protocol微控制器Micro controller实时系统Real time system二、将词组译成应为pulse code modulation 脉冲编码调制the highest frequency component 最高频率分量signaling and synchronization information 信号和同步信息per-channel codec system 每个信道编解码系统two-to-four wire conversion 两到四线转换the lower-frequency portion of the spectrum 低频部分的频谱nonlinear A/D converter 非线性模数转换器amplitude distortion振幅失真to prevent power-line frequency noise from being transmitted防止电力线频率噪声的传播resolution of the resulting digital signal解决由此产生的数字信号the resulting serial bit stream由此产生的串行位流line-to-line crosstalk 线间串扰in a fully integrated form 在一个完全集成的形式coaxial systems同轴系统multimode fiber多模光纤single-mode fiber单模光纤fiber losses 光纤损耗fiber dispersion光纤色散coherent lightwave systems 相干光通信系统fiber amplifiers光纤放大器wavelength-division multiplexing 波分复用erbium-doped fiber amplifier 掺铒光纤放大器propagation mode传播模式refractive index profile折射率剖面optical receiver光接收机dielectric介质destructive interference破坏性干涉stepped-index fiber加强指数纤维synchronous transmission system同步传输系统the equipment supplied by different manufacturers不同厂商提供的设备terminal multiplexer 终端复用器synchronous DXC 同步数字交叉连接设备individual tributary signals各支路信号section overhead段开销central processing unit中央处理单元Local area network 局部区域网络Network topology 网络拓扑Token ring network 令牌环网络reed relay簧片继电器electromechanical switching device机电开关装置crosstalk串扰labour-intensive 劳动密集型semiconductor lasers 半导体激光器light-emitting diode 发光二极管semiconductor photodiodes 半导体光电二极管intensity modulation with direct detection 强度调制直接检测error-correction codes 纠错码receiver sensitivity 接收机灵敏度三、简写PCM (Pulse-code modulation) 脉冲编码调制CDMA (Code division multiple access)码分多址PAM (Pulse amplitude modulation) 脉冲振幅调制ATM (Asynchronous Transfer Mode) 异步传输模式GPRS（General Packet Radio Service）通用分组无线服务USB （Universal serial Bus）通用串行总线AON (Active optical network) 有源光纤网FTTC (fiber to the curb) 光纤到路边WWW ( world wide web ) 全球资讯网LAN (Local Side Band) 局域网WAN （wide area network）广域网WLAN 无线局域网SSB (single side band) 单边带DSP ( digital signal processing) 数字信号处理LASER ( light amplification by stimulated emission of radiation )受激辐射的光放大CCITT（Telephone Consultative Committee）国际电话与电报顾问委员会SLIC （subscriber-line interface circuit）用户线接口电路EDFA （erbium-doped fiber amplifier）掺饵光纤放大器DSF （Dispersion shifted fiber）色散位移光纤ADM （Add and drop multiplexer）分插复用IEEE （Institute of Electrical and Electronic Engineers）电气与电子工程师学会ITU (International Telecommunications Union) 国际电信联盟CATV (Cable Television)有线电视ISDN (Integrated Services Digital Network) 综合数字业务网AGC 自动增益控制器TDM （time-division multiplexer）时分复用SDH（Asynchronous Digital Hierarchy）同步数字系列CCS （Common-Channel Signaling）公共信道信令WDM (wavelength-division multiplexing) 波分复用NNI (Network Node Interface) 网络节点接口LTM (Line Terminal Multiplexer) 线路终端复用器MEMS (Micro Electro Mechanical switching) 微机电开关CPU （central processing unit）中央处理单元CSMA/CD （Multiple Access with collision detection）多路存取/碰撞检测PDCP (Personal Digital cellular packet) 分组数据聚合协议PN 伪码GSM 全球移动通信系统CELP (Code Excited Linear Predictive) 线性预测HDSL (High-bit-rate Digital Subscriber Line) 高比特率数字用户线ADSL (Asymmetrical Digital Subscriber Line) 非对称数字用户线CAP ( Carrierless amplitude modulation) 无载波振幅调制DMT (Discrete multi-tone modulation) 离散多音调制FTTC (Fiber to the curb) 路用光纤FTTH (Fiber to the home) 家用光纤IDN (integrated digital network) 集成数字网络PSTN (public telephone switched network) 公共交换电话网WiFi 无线上网。

脑机接口技术的无创性脑电刺激The field of brain-computer interface (BCI) technology has witnessed remarkable advancements in recent years, particularly in the realm of noninvasive brain stimulation through electroencephalography (EEG). This technique allows for direct communication between the human brain and external devices, opening up new horizons in the treatment of neurological disorders, rehabilitation, and even cognitive enhancement.近年来，脑机接口（BCI）技术领域取得了显著进展，特别是在通过脑电图（EEG）进行的无创性脑刺激方面。

这种技术实现了人脑与外部设备之间的直接交流，为神经系统疾病的治疗、康复甚至认知增强开辟了新的领域。

Noninvasive brain stimulation using EEG-based BCI involves the recording of electrical activity in the brain through electrodes placed on the scalp. These electrodes capture the brain's neural signals, which are then processed and translated into commands that can control external devices. This process allows for the precise targeting of specific brain regions, enabling the delivery of tailored stimuli to modulate neural activity.基于EEG的BCI无创性脑刺激技术涉及通过放置在头皮上的电极记录大脑的电活动。