Resting-state electroencephalography (EEG) microstates reflect sub-second, quasi-stable states of brain activity. Several studies have reported alterations of microstate features in patients with schizophrenia (SZ). Based on these findings, it has been suggested that microstates may represent neurophysiological biomarkers for the classification of SZ. To explore this possibility, machine learning approaches can be employed. Bayesian optimization is a machine learning approach that selects the best-fitted machine learning model with tuned hyperparameters from existing models to improve the classification. In this proof-of-concept preliminary study based on secondary analysis, 20 microstate features were extracted from 14 SZ patients and 14 healthy controls’ EEG signals. These parameters were then ranked as predictors based on their importance, and an optimized machine learning approach was applied to evaluate the performance of the classification. SZ patients had altered microstate features compared to healthy controls. Furthermore, Bayesian optimization outperformed conventional multivariate analyses and showed the highest accuracy (90.93%), AUC (0.90), sensitivity (91.37%), and specificity (90.48%), with reliable results using just six microstate predictors. Altogether, in this proof-of-concept study, we showed that machine learning with Bayesian optimization can be utilized to characterize EEG microstate alterations and contribute to the classification of SZ patients.

Non-invasive brain stimulation (NIBS) techniques are designed to precisely and selectively target specific brain regions, thus enabling focused modulation of neural activity. Among NIBS technologies, low-intensity transcranial focused ultrasound (tFUS) has emerged as a promising new modality. The application of tFUS can safely and non-invasively stimulate deep brain structures with millimetric precision, offering distinct advantages in terms of accessibility to non-cortical regions over other NIBS methods. However, to date, several tFUS aspects still need to be characterized; furthermore, there are only a handful of studies that have utilized tFUS in psychiatric populations. This narrative review provides an up-to-date overview of key aspects of this NIBS technique, including the main components of a tFUS system, the neuronavigational tools used to precisely target deep brain regions, the simulations utilized to optimize the stimulation parameters and delivery of tFUS, and the experimental protocols employed to evaluate the efficacy of tFUS in psychiatric disorders. The main findings from studies in psychiatric populations are presented and discussed, and future directions are highlighted.

Study Objectives: Microstates are semi-stable voltage topographies that account for most of electroencephalogram (EEG) variance. However, the impact of time of the day and sleep on microstates has not been examined. To address this gap, we assessed whether microstates differed between the evening and morning and whether sleep slow waves correlated with microstate changes in healthy participants. Methods: Forty-five healthy participants were recruited. Each participant underwent 6 minutes of resting state EEG recordings in the evening and morning, interleaved by sleep EEGs. Evening-to-morning changes in microstate duration, coverage, and occurrence were assessed. Furthermore, correlation between microstate changes and sleep slow-wave activity (SWA) and slow-wave density (SWD) were performed. Results: Two-way ANOVAs with microstate class (A, B, C, and D) and time (evening and morning) revealed significant microstate class × time interaction for duration (F(44) = 5.571, p = 0.002), coverage (F(44) = 6.833, p = 0.001), and occurrence (F(44) = 5.715, p = 0.002). Post hoc comparisons showed significant effects for microstate C duration (padj = 0.048, Cohen’s d = −0.389), coverage (padj = 0.002, Cohen’s d = −0.580), and occurrence (padj = 0.002, Cohen’s d = -0.606). Topographic analyses revealed inverse correlations between SWD, but not SWA, and evening-to-morning changes in microstate C duration (r = −0.51, padj = 0.002), coverage (r = −0.45, padj = 0.006), and occurrence (r = −0.38, padj = 0.033). Conclusions: Microstate characteristics showed significant evening-to-morning changes associated with, and possibly regulated by, sleep slow waves. These findings suggest that future microstate studies should control for time of day and sleep effects.

Background: Transcranial Magnetic Stimulation (TMS) with simultaneous Electroencephalogram (TMS-EEG) allows assessing the neurophysiological properties of cortical neurons. However, TMS-evoked EEG potentials (TEPs) can be affected by components unrelated to TMS direct neuronal activation. Accurate, automatic tools are therefore needed to establish the quality of TEPs. Objective: To assess the discriminability of EEG responses to TMS vs. EEG responses to sham stimulations using sequence-to-sequence machine learning (ML). Methods: Two indipendent TMS-EEG datasets including TMS and several sham stimulation conditions were obtained from the left motor area of healthy volunteers (N=33 across datasets). A Bi-directional Long Short-Term Memory (BiLSTM) ML network was used to label each time point of the EEG signals as pertaining to TMS or sham conditions. Main outcome measures included accuracy at single-trial level and after averaging five to twenty trials. Results: For TMS conditions, post-stimulus vs. baseline/pre-stimulus EEG comparisons yielded moderate (60%-75%) single-trial accuracy and high-accuracy (>75%) for 20 trials across datasets, while for sham conditions post- vs. baseline/pre-stimulus EEG comparisons yielded lower accuracy rates than for TMS conditions, except for unmasked auditory stimulation. Furthermore, baseline/pre-stimulus TMS vs. baseline/pre-stimulus sham EEG comparisons showed chance-level accuracy, whereas post-stimulus TMS vs. post-stimulus sham EEG comparisons had moderate (single trial) to high (20 trial) accuracy, except for TMS with and without the click noise masking. Single-subject findings were comparable to group-level results across datasets. Conclusions: TEPs after active TMS are discernible from various sham stimulations even after a handful of trials and at the single-subject level using a BiLSTM ML approach.