Objective: In presurgical evaluation of temporal lobe epilepsy (TLE), selection of the resection side is challenging when bilateral temporal epileptiform discharges or structural abnormalities are present. We aim to evaluate the lateralization value of beamformer analysis of magnetoencephalography (MEG) in TLE.Methods: MEG data from 14 TLE patients were analyzed through beamformer analysis. We measured the hemispherical power distribution of beamformer sources and calculated the lateralization index (LI). We calculated the LI at multiple frequencies to explore the frequency dependency and at the delta frequency to define laterality. LI values ranging from −1 to −0.05 indicated right hemispheric dominance. LI values ranging from 0.05 to 1 indicated left hemispheric dominance. LI values ranging from −0.05 to 0.05 defined bilaterality. We measured the power of beamformer sources with a 9-s duration to explore time dependency.Results: The beamformer analysis showed that 10/14 patients had power dominance ipsilateral to resection. The delta frequency band had a higher lateralization value than other frequency bands. A time-dependent power fluctuation was found in the delta frequency band.Conclusions: MEG beamformer analysis, especially in the delta band, might efficiently provide additional information regarding lateralization in TLE.

Generalized Anxiety Disorder (GAD) is a highly prevalent yet poorly understood chronic mental disorder. Previous studies have associated GAD with excessive activation of the right dorsolateral prefrontal cortex (DLPFC). This study aimed to investigate the effect of low-frequency repetitive transcranial magnetic stimulation (repetitive TMS, rTMS) targeting the right DLPFC on clinical symptoms and TMS-evoked time-varying brain network connectivity in patients with GAD. Eleven patients with GAD received 1 Hz rTMS treatment targeting the right DLPFC for 10 days. The severity of the clinical symptoms was evaluated using the Hamilton Anxiety Scale (HAMA) and the Hamilton Depression Scale (HAMD) at baseline, right after treatment, and at the one-month follow-up. Co-registration of single-pulse TMS (targeting the right DLPFC) and electroencephalography (TMS-EEG) was performed pre- and post-treatment in these patients and 11 healthy controls. Time-varying brain network connectivity was analyzed using the adaptive directed transfer function. The scores of HAMA and HAMD significantly decreased after low-frequency rTMS treatment, and these improvements in ratings remained at the one-month follow-up. Analyses of the time-varying EEG network in the healthy controls showed a continuous weakened connection information outflow in the left frontal and mid-temporal regions. Compared with the healthy controls, the patients with GAD showed weakened connection information outflow in the left frontal pole and the posterior temporal pole at baseline. After 10-day rTMS treatment, the network patterns showed weakened connection information outflow in the left frontal and temporal regions. The time-varying EEG network changes induced by TMS perturbation targeting right DLPFC in patients with GAD were characterized by insufficient information outflow in the left frontal and temporal regions. Low-frequency rTMS targeting the right DLPFC reversed these abnormalities and improved the clinical symptoms of GAD.

In this study, a multi-feature fusion and decoupling solution based on the RNN is proposed from a discriminative perspective. This method can address the identity and age information extraction losses in cross-age face recognition. This method not only constrains the correlation between identity and age using correlation loss but also optimizes identity feature restoration using feature decoupling. The model was trained and simulated in CACD and CACD-VS datasets. The single-task learning model stabilized after 125 iterations of training, while the multi-task learning model reached a stable and convergent state after 75 iterations. In terms of performance analysis, the DE-RNN model had the highest recognition accuracy with a mAP of 92.4%. The Human Voting model had a value of 90.2%. The mAP of the Human Average model was 81.8%, whereas the mAP of the DAL model was the lowest at 78.1%. Experiments proved that the model constructed in this study has effective recognition and application value in the cross-age face recognition scenario.