摘要
突触后电位是神经元传递信息的重要载体,该信号的动力学分析是神经科学研究中的常用方法。但目前传统的分析方法中多采用手工提取动力学特征值,分析数据量有限。针对此缺陷,本文提出了一种神经元突触传递的动力学特征提取方法。在该方法中,采用低通滤波器提取有效数据,降低了数据计算量和复杂性;采用中值滤波算法去除了信号中随机噪声、刺激伪迹,并校正基线漂移,利用曲线拟合提取了突触后电位信号波形斜率特征值,实现突触后电位信号变化趋势的可视化。该特征提取方法的平均绝对百分比误差一般在5%左右,幅度特征值的平均绝对百分比误差一般在3%左右。同时取95%的置信区间,降低波形的测量误差,使得有效数据正检率指标高达98%。结果表明该特征提取方法可提取信号中的有效数据,并能保证特征值提取的精度,减小了人工提取存在的误差。通过神经元响应数据,验证了该方法的可行性,提出了太赫兹调控神经元突触传递的猜想。
Postsynaptic potentials are vital transmitters of information conveyed by neurons.In neuroscience research,kinetic analysis of this signal is a widely used technique.Nonetheless,conventional analysis methods depend on manual extraction of kinetic eigenvalues and are limited in terms of the analyzed data.To overcome this limitation,a kinetic feature extraction approach for neuronal synaptic transmission is proposed.In this approach,an efficient data extraction is achieved through the utilization of a low-pass filter.This method lessens the amount and intricacy of data calculation.Similarly,a median filtering algorithm is added to remove random noise and stimulus artifacts present in the signal.Mediation also corrects the baseline drift.Finally,slope eigenvalues of the post-synaptic potential signal waveforms are identified using curve-fitting.This allows for the data visualization and analysis of changes in the trend of post-synaptic potential signals.Technical terms are explained when initially introduced.The feature extraction method typically has an average absolute percentage error of approximately 5%.Meanwhile,the magnitude eigenvalues′average absolute percentage error is typically around 3%.When the 95%confidence interval is applied,the waveform′s measurement error is reduced,resulting in higher positive detection rates for effective data,with a rate of 98%.These findings demonstrate that the feature extraction method effectively extracts data from signals while ensuring the accuracy of feature value extractions,thus reducing errors that may occur during manual extraction.The method′s feasibility is confirmed through neuronal response data,while suggesting the hypothesis of terahertz modulation on neuronal synaptic transmission.
作者
马少卿
龚士香
路承彪
李小俚
李英伟
MA Shaoqing;GONG Shixiang;LU Chengbiao;LI Xiaoli;LI Yingwei(School of Information Science and Engineering,Yanshan University,Qinhuangdao,Hebei 066004,China;Hebei Key Laboratory of Information Transmission and Signal Processing,Yanshan University,Qinhuangdao,Hebei 066004,China;Henan International Joint Laboratory of Noninvasive Neural Modulation,Xinxiang Medical College,Xinxiang,Henan 453000,China;State Key Laboratory of Cognitive Neuroscience and Learning,Beijing Normal University,Beijing 100875,China)
出处
《燕山大学学报》
CAS
北大核心
2024年第3期255-264,共10页
Journal of Yanshan University
基金
国家自然科学基金资助项目(61827811)
河北省自然科学基金资助项目(F2020203099)
河北省重点实验室资助项目(202250701010046)
河北省引进留学人员资助项目(C20200364)。
关键词
太赫兹
突触传递
突触后电位
特征提取
terahertz synaptic transmission
postsynaptic potential
feature extraction
