Abstract: Objective  This study explores the electroencephalographic (EEG) characteristics of stroke patients during urination and defecation motor tasks using brain-computer interface (BCI) technology. By elucidating the central neural mechanisms underlying the control of excretion, this research aims to construct a BCI paradigm for urination and defecation motor tasks suitable for stroke patients, and to provide novel insights for the rehabilitation of post-stroke patients suffering from urination and defecation dysfunction.
Methods  Stroke patients admitted to the Huashan Hospital, Fudan University, and the Shandong Provincial Third Hospital between September 2024 and April 2025 were enrolled.  All patients were instructed to perform urination and defecation motor tasks, while their EEG features were recorded. By analyzing electroencephalography brain maps, time-frequency plots, and event-related spectral perturbation values, this study elucidates the temporal characteristics of EEG signals during the initiation and maintenance of these motor tasks, as well as the differences in frequency characteristics within the α (8-13 Hz), β1 (＞13-20 Hz), and β2 (＞20-30 Hz) frequency bands.
Results  A total of 14 stroke patients were enrolled in this study, comprising five females and nine males, with a mean age of (52.7±14.4) years. Electroencephalography brain maps analysis revealed that both urination/defecation retention and voiding tasks induced significant event-related desynchronization (ERD). These responses exhibited distinct spatiotemporal and spectral characteristics across the α, β1, and β2 frequency bands. Specifically, the task initiation phase (0-2 s) elicited widespread activation across all frequency bands, whereas the task maintenance phase (＞2-5 s) was predominantly concentrated in the α frequency band. Furthermore, the frontal cortex demonstrated sustained engagement in the β2 frequency band during the inhibitory control of urination and defecation.
Conclusions  This study elucidates the staged dynamic neural mechanisms underlying urination and defecation motor control. It reveals that, upon the initiation of urination and defecation motor tasks, stroke patients exhibit ERD bursts within the α, β1, and β2 frequency bands. These findings provide a significant theoretical foundation for the design and application of BCI systems.

Key words: Brain-computer interface; Urination and defecation dysfunction; Motor imagery; Electroencephalographic analysis; Stroke

摘要： 目的 基于脑机接口（brain-computer interface，BCI）探讨卒中患者进行二便运动任务时的脑电特征，揭示二便运动控制的中枢神经机制，构建适用于卒中患者的二便运动任务BCI范式，为卒中后二便障碍患者的康复提供新思路。
方法 纳入2024年9月—2025年4月于复旦大学附属华山医院和山东省立第三医院就诊的卒中患者，指导其进行二便运动任务，并记录脑电特征。通过分析脑电地形图、时频图和事件相关谱扰动值，揭示脑电信号在二便运动任务启动和维持状态的时间特征，以及在α（8～13 Hz）、β1（＞13～20 Hz）、β2（＞20～30 Hz）频段的频率特征差异。
结果 本研究共纳入卒中患者14例，其中女性5例，男性9例，平均年龄（52.7±14.4）岁。脑电地形图分析结果显示，憋/排二便运动任务均可诱发显著的事件相关去同步化现象，其响应在α、β1和β2频段均具有明显的时间-频率-空间特异性：二便运动任务启动状态（0～2 s）呈全频段广泛激活，任务维持状态（＞2～5 s）主要集中于α频段，且额叶皮质区在β2频段持续参与二便抑制控制。
结论 本研究揭示了二便运动控制分阶段的动态神经机制，表明卒中患者在二便运动任务启动状态下存在α、β1和β2频段的事件相关去同步化爆发启动现象，为BCI系统的设计与应用提供了重要的理论依据。

关键词: 脑机接口; 二便障碍; 运动想象; 脑电分析; 卒中