外围设备
肌电图
物理医学与康复
神经科学
判别式
冲程(发动机)
脑电图
眼电学
心理学
医学
固定(群体遗传学)
神经生理学
听力学
接收机工作特性
脑功能偏侧化
离解(化学)
临床神经生理学
眼球运动
神经影像学
感觉系统
神经康复
心脏病学
凝视
神经学
电生理学
级联
刺激(心理学)
作者
Youze He,Yudie Hu,Xuesen Wu,Jingchun Zhang,Zhihan Liu,Jinshan Huang,Xiuxu Lu,Xinyu Wu,Jingsong Wu,Lidian Chen,Jing Tao
标识
DOI:10.3389/fnhum.2026.2002490
摘要
Upper-limb motor impairment after stroke is compounded by a disrupted sensorimotor cascade linking visual attention, cortical motor preparation, and peripheral muscle activation. Whether these stages are temporally dissociated in subacute stroke, and whether their timing carries information about the preparation and execution of a reaching action, remains unclear. Thirty subacute stroke survivors and 30 healthy controls performed a full-plane nine-grid reaching task while eye-tracking, electroencephalography (EEG), and surface electromyography (EMG) were recorded synchronously. Three neurophysiological latencies were extracted: first fixation on the target, stimulus-locked lateralized readiness potential (s-LRP) onset, and peripheral muscle activation onset. Compared with controls, patients showed a stepwise delay across the cascade, with delayed visual engagement (616.73 ms vs. 367.86 ms) and delayed peripheral muscle activation (1,080 ms vs. 605 ms) bracketing a cortical preparation window. A random-forest classifier was trained on window-level gaze-timing features, computed separately for each target-presentation and inter-target interval of the task. Under participant-level leave-one-subject-out cross-validation, it achieved an area under the receiver operating characteristic curve (AUC) of 0.973 (95% CI 0.958–0.984) and an accuracy of 0.918, numerically the highest among the five classifiers compared. These findings describe a quantifiable temporal organization of the sensorimotor cascade in the subacute stage of stroke and indicate that gaze timing carries discriminative information about the occurrence of an instructed reaching action. They establish a descriptive, testable framework rather than a demonstrated clinical benefit. Validation in chronic stroke and across independent samples is further required.
科研通智能强力驱动
Strongly Powered by AbleSci AI