The functional reorganization of brain networks sustaining gait is poorly characterized in amyotrophic lateral sclerosis (ALS) despite ample evidence of progressive disconnection between brain regions. The main objective of this fMRI study is to assess gait imagery-specific networks in ALS patients using dynamic causal modeling (DCM) complemented by parametric empirical Bayes (PEB) framework. Seventeen lower motor neuron predominant (LMNp) ALS patients, fourteen upper motor neuron predominant (UMNp) ALS patients and fourteen healthy controls participated in this study. Each subject performed a dual motor imagery task: normal and precision gait. The Movement Imagery Questionnaire (MIQ-rs) and imagery time (IT) were used to evaluate gait imagery in each participant. In a neurobiological computational model, the circuits involved in imagined gait and postural control were investigated by modelling the relationship between normal/precision gait and connection strengths. Behavioral results showed significant increase in IT in UMNp patients compared to healthy controls (P_corrected < 0.05) and LMNp (P_corrected < 0.05). During precision gait, healthy controls activate the model's circuits involved in the imagined gait and postural control. In UMNp, decreased connectivity (inhibition) from basal ganglia (BG) to supplementary motor area (SMA) and from SMA to posterior parietal cortex (PPC) is observed. Contrary to healthy controls, DCM detects no cerebellar-PPC connectivity in neither UMNp nor LMNp ALS. During precision gait, bilateral connectivity (excitability) between SMA and BG is observed in the LMNp group contrary to UMNp and healthy controls. Our findings demonstrate the utility of implementing both DCM and PEB to characterize connectivity patterns in specific patient phenotypes. Our approach enables the identification of specific circuits involved in postural deficits, and our findings suggest a putative excitatory–inhibitory imbalance. More broadly, our data demonstrate how clinical manifestations are underpinned by network-specific disconnection phenomena in ALS.

Background: Transcranial direct current stimulation (tDCS) of the cerebellum dynamically modulates cerebello-thalamo-cortical excitability in a polarity-specific manner during motor, visuo- motor and cognitive tasks. It remains to be established whether tDCS of the cerebellum impact also on resting-state intrinsically connected networks (ICNs). Such impact would open novel research and therapeutical doors for the neuromodulation of ICNs in human. Method: We combined tDCS applied over the right cerebellum and fMRI to investigate tDCS- induced resting-state intrinsic functional reconfiguration, using a randomized, sham-controlled design. fMRI data were recorded both before and after real anodal stimulation (2 mA, 20 min) or sham tDCS in 12 right-handed healthy volunteers. We resorted to a region-of-interest static correlational analysis and to a sliding window analysis to assess temporal variations in resting state FC between the cerebellar lobule VII and nodes of the main ICNs. Results: After real tDCS and compared with sham tDCS, functional changes were observed between the cerebellum and ICNs. Static FC showed enhanced or decreased correlation between cerebellum and brain areas belonging to visual, default-mode (DMN), sensorimotor and salience networks (SN) (p-corrected < 0.05). The temporal variability (TV) of BOLD signal was significantly modified after tDCS displaying in particular a lesser TV between the whole lobule VII and DMN and central executive network and a greater TV between crus 2 and SN. Static and dynamic FC was also modified between cerebellar lobuli. Conclusion: These results demonstrate short- and long-range static and majorly dynamic effects of tDCS stimulation of the cerebellum affecting distinct resting-state ICNs, as well as intracerebellar functional connectivity, so that tDCS of the cerebellum appears as a non-invasive tool reconfigurating the dynamics of ICNs.

Objectives: The objective of this study is to investigate the efficacy of psychological Interventions – Mindfulness or Implementation Intention – associated with a Physical Activity program, delivered via internet, in reducing Multiple Sclerosis symptoms.Method: Thirty-five adults were randomly assigned to one of the three groups: a Mindfulness-Based Intervention group (N = 12), Implementation Intention group (N = 11), and a Control Group (N = 12). All the groups received the same Physical Activity program. The Mindfulness condition group received daily training in the form of pre-recorded sessions while the Implementation group elaborated their specific plans once a week. Mobility, fatigue, and the impact of the disease on the patient’s life were measured. Two measurement times are carried out in pre-post intervention, at baseline and after eight weeks.Results: Overall, after 8 weeks intervention, results show that there was a significant increase in Walking distance in the three groups. In addition, the within-group analysis showed a statistically significant improvement between pre and post intervention on the physical component of the Disease Impact scale in the Implementation Intention group (p = 0.023) with large effect size, in the Mindfulness-Based Intervention group (p = 0.008) with a medium effect size and in the control group (p = 0.028) with small effect size. In the Implementation Intention group, all physical, psychosocial and cognitive Fatigue Impact subscales scores decreased significantly (p = 0.022, p = 0.023, and p = 0.012, respectively) and the physical component was statistically and negatively correlated (r = −0.745; p = 0.008) when Implementation Intention group practice a mild to moderate physical activity. In the Mindfulness-Based Intervention group, the physical component (MFIS) showed a statistically significant improvement (p = 0.028) but no correlation with moderate-to-vigorous physical activity (MVPA); the control group outcomes did not reveal any significant change.Conclusion: The results of this study are very encouraging and show the feasibility of Mindfulness interventions associated with physical activity to improve the health of people with MS. Further study should assess Mindfulness interventions tailored to MS condition and using both hedonic and eudemonic measures of happiness.

Emotion detection using electroencephalogram (EEG) signals is a rapidly evolving field with significant applications in mental health diagnostics, affective computing, and human–computer interaction. However, existing approaches often face challenges related to accuracy, interpretability, and real-time feasibility. This study leverages the DEAP dataset to explore and evaluate various machine learning and deep learning techniques for emotion recognition, aiming to address these challenges. To ensure reproducibility, we have made our code publicly available. Extensive experimentation was conducted using K-Nearest Neighbors (KNN), Support Vector Machines (SVMs), Decision Tree (DT), Random Forest (RF), Bidirectional Long Short-Term Memory (BiLSTM), Gated Recurrent Units (GRUs), Convolutional Neural Networks (CNNs), autoencoders, and transformers. Our hybrid approach achieved a peak accuracy of 85–95%, demonstrating the potential of advanced neural architectures in decoding emotional states from EEG signals. While this accuracy is slightly lower than some state-of-the-art methods, our approach offers advantages in computational efficiency and real-time applicability, making it suitable for practical deployment. Furthermore, we employed SHapley Additive exPlanations (SHAP) to enhance model interpretability, offering deeper insights into the contribution of individual features to classification decisions. A comparative analysis with existing methods highlights the novelty and advantages of our approach, particularly in terms of accuracy, interpretability, and computational efficiency. A key contribution of this study is the development of a real-time emotion detection system, which enables instantaneous classification of emotional states from EEG signals. We provide a detailed analysis of its computational efficiency and compare it with existing methods, demonstrating its feasibility for real-world applications. Our findings highlight the effectiveness of hybrid deep learning models in improving accuracy, interpretability, and real-time processing capabilities. These contributions have significant implications for applications in neurofeedback, mental health monitoring, and affective computing. Future work will focus on expanding the dataset, testing the system on a larger and more diverse participant pool, and further optimizing the system for broader clinical and industrial applications.

This study explores the use of deep learning techniques for classifying EEG signals in the context of Alzheimer’s disease (AD) and frontotemporal dementia (FTD). We propose a novel classification pipeline that combines autoencoders for feature extraction and bidirectional long short-term memory (Bi-LSTM) networks for analyzing patterns over time in EEG data. Given the complexity and high dimensionality of EEG signals, we employed an autoencoder to reduce data dimensionality while preserving key diagnostic features. The Bi-LSTM model effectively identified subtle temporal patterns in brain activity that are indicative of AD and FTD. To enhance interpretability, we applied SHapley Additive exPlanations (SHAP), providing insights into how individual features contribute to the model’s predictions. We evaluated our approach on a publicly available EEG dataset from OpenNeuro, which includes resting-state EEG recordings from 88 elderly participants—36 with AD, 23 with FTD, and 29 cognitively normal controls. EEG provides a non-invasive, cost-effective tool for brain monitoring, but presents challenges such as noise sensitivity and inter-subject variability. Despite these challenges, our approach achieved 98% accuracy while maintaining transparency, making it a promising tool for clinical applications in the diagnosis of neurodegenerative diseases.