Runhong Yao

Introduction: Mild cognitive impairment (MCI) is frequently accompanied by impairments in postural control and trunk muscle function. Although gamma–theta rhythmic stimulation has been reported to modulate neural oscillatory activity, its influence on brain–muscle interactions in individuals with MCI remains poorly understood. This exploratory pilot study investigated whether changes in EEG connectivity following gamma–theta stimulation were associated with alterations in trunk muscle electromyographic (EMG) activity in older adults with MCI. Methods: Nine older adults with MCI underwent two separate 10-min sessions of combined 40 Hz visual flicker and theta-frequency mechanical stimulation. Resting-state EEG and trunk muscle EMG were assessed before and after stimulation. EEG connectivity changes were calculated using Spearman correlations between predefined electrode pairs. Supplementary analyses using phase-lag index (PLI) and imaginary coherence (iCoh) were performed to address potential volume conduction effects. EMG activity was recorded bilaterally from the multifidus and lumbar paraspinal muscles during trunk extension tasks. Correlation analyses examined associations between EEG connectivity changes and EMG changes in RMS amplitude, mean frequency, and spectral power measures. Results: Although no predefined EEG connectivity pair demonstrated significant group-level change following stimulation, several EEG connectivity change scores were significantly associated with concurrent EMG changes. The most robust associations involved left multifidus activity: FC5–F3 connectivity changes were associated with left multifidus RMS amplitude ( ρ = +0.800, p = 0.010), and T7–T8 connectivity changes were associated with left multifidus mid-frequency spectral power ( ρ = −0.800, p = 0.010). Leave-one-out sensitivity analyses indicated that both associations remained statistically significant in all nine subsets. Supplementary phase-based connectivity analyses provided limited convergent support for the primary findings. Although several contralateral EEG–EMG associations were identified, the overall pattern was characterized by a predominance of ipsilateral associations. Discussion: These findings suggest that gamma–theta stimulation may influence coordinated brain–muscle responses in older adults with MCI, with a predominance of left-hemisphere EEG–EMG associations indicating relative hemispheric predominance rather than strict hemispheric exclusivity. Associations involving broader measures of muscle activation and spectral distribution were more robust than those involving specific spectral indices. Given the small sample size, absence of sham control, non-simultaneous EEG–EMG recording, and exploratory statistical approach, these findings should be interpreted as hypothesis-generating and require confirmation in larger controlled studies.

Mild cognitive impairment (MCI) is associated with age-related alterations in neural oscillatory activity and cerebral hemodynamics, while mobility limitations often restrict participation in conventional exercise-based interventions. This study investigated exploratory, single-arm, within-subject short-term neurovascular and electrophysiological responses to a combined visual and vibration stimulation protocol in older adults with MCI. Cerebral blood flow was assessed using near-infrared spectroscopy (NIRS) in 27 participants, and electroencephalography (EEG) was additionally recorded in a subgroup of nine participants ( n = 9) drawn from the same cohort. The intervention consisted of simultaneous 40 Hz visual flicker and low-frequency mechanical vibration delivered during a seated condition. Changes in total hemoglobin concentration and EEG power in predefined frequency bands were analyzed before and after the intervention, with additional time-window analyses to examine temporal dynamics. At baseline, NIRS revealed higher total hemoglobin levels in the left prefrontal cortex compared with the right. Following the intervention, total hemoglobin decreased significantly in both hemispheres, with a larger reduction observed in the right hemisphere. Time-resolved analyses demonstrated an earlier hemodynamic response in the right hemisphere, followed by a delayed response in the left. EEG analyses showed a reduction in frontal gamma-band power accompanied by an increase in theta-band power in temporoparietal regions after the intervention. These electrophysiological changes were stable across time windows. Together, these findings suggest that short-term combined visual and vibration stimulation was associated with asymmetric and time-dependent changes in cerebral hemodynamics and neural oscillatory activity in older adults with MCI. This non-volitional stimulation approach may provide a feasible experimental framework for investigating brain responses in aging populations with limited mobility. Given the exploratory, uncontrolled design, acute measurement window, and small EEG subgroup, these findings should be interpreted as hypothesis-generating rather than as evidence of clinical or neuroplastic efficacy.

In this study, we propose multiple deep learning models for classifying gray matter MRI images of healthy individuals, prodromal Parkinson’s disease (PD) subjects, and diagnosed PD patients. The two proposed models extend conventional deep learning architectures—MedicalNet3D and 3D ResNet18—by performing feature extraction separately for each class and inputting these features into distinct multilayer perceptron (MLP) classifiers constructed via fine-tuning. To mitigate overfitting problem and improve generalizability, we introduce a training method based on group-wise feature fusion, in which subject IDs are separated to avoid data leakage during training. Through comparative experiments using the PPMI database, the effectiveness of the proposed approach was validated.