儿童失神癫痫发作全程的默认模式网络脑磁图功率变化研究  

Dynamics of neuromagnetic power in the default mode network throughout the wholecourse of childhood absence epilepsy

作　　者：朱寅杰 李明昊 蒋佩琳 王小姗[1] ZHUYinjie;LI Minghao;JIANG Peilin;WANG Xiaoshan(Department of Neurology,The Affiliated Brain Hospital of Nanjing Medical University,Nanjing 210000,China)

机构地区：[1]南京医科大学附属脑科医院神经内科,南京210000

出　　处：《癫痫杂志》2025年第2期106-113,共8页Journal of Epilepsy

基　　金：国家自然科学基金(82071455)。

摘　　要：目的通过研究儿童失神癫痫(childhood absence epilepsy,CAE)发作全程中的默认模式网络(default mode network,DMN)节点脑磁图(magnetoencephalographic,MEG)信号时空动态特征,寻找区分CAE不同发作时期状态的生物学标志物,以期揭示CAE发作过程中意识状态变化的神经生理机制。方法本研究纳入36例未服药的CAE患儿,使用CTF-275通道的MEG设备采集CAE发作期及发作间期的脑磁信号数据。划分并选取发作间期、发作前期、发作起始期、发作期、发作终止期、发作后期共6个时期的14个MEG数据片段。选取12个DMN相关脑区作为感兴趣区(region of interest,ROI),运用最小范数估计结合Welch技术对ROI进行功率谱密度(power spectral density,PSD)计算,对相对PSD值进行差异分析,使用随机森林模型筛选区分癫痫状态的重要PSD特征。结果DMN相关脑区在不同频段的功率变化具有明显的同步性。发作过程中,δ频段功率在发作起始期迅速增加,并在发作终止期迅速下降;而θ到γ2频段的功率则在发作初期下降,发作终止后逐步回升。同时,在发作起始后的O+2阶段,β频段的功率出现了短暂的峰值,随后迅速回落。此外,内侧额叶在发作期的δ频段功率显著低于其他DMN脑区,而在α频段则表现出更高的功率水平。基于随机森林模型的特征重要性分析显示,楔前叶、外侧颞叶和内侧颞叶在区分不同发作状态时具有重要贡献,尤其是楔前叶在α和δ频段的功率变化对模型的分类精度提升最为显著。结论本研究探讨了DMN在CAE发作全程的时空动态,揭示了特定脑区和频段的功率变化在发作起始、维持及终止过程中的特征性表现,并发现楔前叶的功率指标可以作为区分CAE不同发作时期的重要特征。这些发现为理解CAE意识状态改变的病理生理机制提供了新的视角。Objective To investigate biological markers that differentiate states during various seizure periods of childhood absence epilepsy(CAE)by examining the spatiotemporal dynamics of magnetoencephalographic(MEG)signals from Default Mode Network(DMN)nodes,revealing the neurophysiological mechanisms underlying changes in consciousness during CAE seizures.Methods Thirty-six drug-native children diagnosed with CAE were recruited.The interictal data,ictal data of CAE children were collected using a CTF-225 channel MEG system.Conduct temporal homogeneity partitioning for all seizure period data,co-registering 14 distinct seizure states.Identify 12 brain regions associated with the default mode network(DMN)as regions of interest(ROI);employ minimum norm estimation in conjunction with the Welch method to compute the power spectral density(PSD)of the ROI;conduct differential analysis on the relative PSD values;and use a random forest model to identify significant PSD features that differentiate between states of epilepsy.Results Power changes in DMN-related brain regions across various frequency bands show significant synchrony.During a seizure,the power in the 8 band rapidly increases at the onset and quickly decreases at the end.

关 键 词：儿童失神癫痫 默认模式网络 脑磁图 频谱功率分析 特征分析 

分 类 号：R742.1[医药卫生—神经病学与精神病学]