Abnormal emotional reactions of the brain in patients with facial nerve paralysis have not yet been reported. This study aims to investigate this issue by applying a machine-learning algorithm that discriminates brain emotional activities that belong either to patients with facial nerve paralysis or to healthy controls. Beyond this, we assess an emotion rating task to determine whether there are differences in their experience of emotions. MEG signals of 17 healthy controls and 16 patients with facial nerve paralysis were recorded in response to picture stimuli in three different emotional categories (pleasant, unpleasant, and neutral). The selected machine learning technique in this study was the logistic regression with LASSO regularization. We demonstrated significant classification performances in all three emotional categories. The best classification performance was achieved considering features based on event-related fields in response to the pleasant category, with an accuracy of 0.79 (95% CI (0.70, 0.82)). We also found that patients with facial nerve paralysis rated pleasant stimuli significantly more positively than healthy controls. Our results indicate that the inability to express facial expressions due to peripheral motor paralysis of the face might cause abnormal brain emotional processing and experience of particular emotions.

The processing of emotions in the human brain is an extremely complex process that extends across a large number of brain areas and various temporal processing steps. In the case of magnetoencephalography (MEG) data, various frequency bands also contribute differently. Therefore, in most studies, the analysis of emotional processing has to be limited to specific sub-aspects. Here, we demonstrated that these problems can be overcome by using a nonparametric statistical test called the cluster-based permutation test (CBPT). To the best of our knowledge, our study is the first to apply the CBPT to MEG data of brain responses to emotional stimuli. For this purpose, different emotionally impacting (pleasant and unpleasant) and neutral pictures were presented to 17 healthy subjects. The CBPT was applied to the power spectra of five brain frequencies, comparing responses to emotional versus neutral stimuli over entire MEG channels and time intervals within 1500 ms post-stimulus. Our results showed significant clusters in different frequency bands, and agreed well with many previous emotion studies. However, the use of the CBPT allowed us to easily include large numbers of MEG channels, wide frequency, and long time-ranges in one study, which is a more reliable alternative to other studies that consider only specific sub-aspects.

Studies investigating human brain response to emotional stimuli—particularly high-arousing versus neutral stimuli—have obtained inconsistent results. The present study was the first to combine magnetoencephalography (MEG) with the bootstrapping method to examine the whole brain and identify the cortical regions involved in this differential response. Seventeen healthy participants (11 females, aged 19 to 33 years; mean age, 26.9 years) were presented with high-arousing emotional (pleasant and unpleasant) and neutral pictures, and their brain responses were measured using MEG. When random resampling bootstrapping was performed for each participant, the greatest differences between high-arousing emotional and neutral stimuli during M300 (270–320 ms) were found to occur in the right temporo-parietal region. This finding was observed in response to both pleasant and unpleasant stimuli. The results, which may be more robust than previous studies because of bootstrapping and examination of the whole brain, reinforce the essential role of the right hemisphere in emotion processing.

Surface electromyography (EMG) allows reliable detection of muscle activity in all nine intrinsic and extrinsic ear muscles during facial muscle movements. The ear muscles are affected by synkinetic EMG activity in patients with postparalytic facial synkinesis (PFS). The aim of the present work was to establish a machine-learning-based algorithm to detect eyelid closure and smiling in patients with PFS by recording sEMG using surface electromyography of the auricular muscles. Sixteen patients (10 female, 6 male) with PFS were included. EMG acquisition of the anterior auricular muscle, superior auricular muscle, posterior auricular muscle, tragicus muscle, orbicularis oculi muscle, and orbicularis oris muscle was performed on both sides of the face during standardized eye closure and smiling tasks. Machine-learning EMG classification with a support vector machine allowed for the reliable detection of eye closure or smiling from the ear muscle recordings with clear distinction to other mimic expressions. These results show that the EMG of the auricular muscles in patients with PFS may contain enough information to detect facial expressions to trigger a future implant in a closed-loop system for electrostimulation to improve insufficient eye closure and smiling in patients with PFS.

Here we aimed to automatically classify human emotion earlier than is typically attempted. There is increasing evidence that the human brain differentiates emotional categories within 100-300 ms after stimulus onset. Therefore, here we evaluate the possibility of automatically classifying human emotions within the first 300 ms after the stimulus and identify the time-interval of the highest classification performance. To address this issue, MEG signals of 17 healthy volunteers were recorded in response to three different picture stimuli (pleasant, unpleasant, and neutral pictures). Six Linear Discriminant Analysis (LDA) classifiers were used based on two binary comparisons (pleasant versus neutral and unpleasant versus neutral) and three different time-intervals (100-150 ms, 150-200 ms, and 200-300 ms post-stimulus). The selection of the feature subsets was performed by Genetic Algorithm and LDA. We demonstrated significant classification performances in both comparisons. The best classification performance was achieved with a median AUC of 0.83 (95 %- CI [0.71; 0.87]) classifying brain responses evoked by unpleasant and neutral stimuli within 100-150 ms, which is at least 850 ms earlier than attempted by other studies. Our results indicate that using the proposed algorithm, brain emotional responses can be significantly classified at very early stages of cortical processing (within 300 ms). Moreover, our results suggest that emotional processing in the human brain occurs within the first 100-150 ms.

Lithium therapy has been shown to affect imaging measures of brain function and microstructure in human immunodeficiency virus (HIV)-infected subjects with cognitive impairment. The aim of this proof-of-concept study was to explore whether changes in brain microstructure also entail changes in functional connectivity. Functional MRI data of seven cognitively impaired HIV infected individuals enrolled in an open-label lithium study were included in the connectivity analysis. Seven regions of interest (ROI) were defined based on previously observed lithium induced microstructural changes measured by Diffusion Tensor Imaging. Generalized partial directed coherence (gPDC), based on time-variant multivariate autoregressive models, was used to quantify the degree of connectivity between the selected ROIs. Statistical analyses using a linear mixed model showed significant differences in the average node strength between pre and post lithium therapy conditions. Specifically, we found that lithium treatment in this population induced changes suggestive of increased strength in functional connectivity. Therefore, by exploiting the information about the strength of functional interactions provided by gPDC we can quantify the connectivity changes observed in relation to a given intervention. Furthermore, in conditions where the intervention is associated with clinical changes, we suggest that this methodology could enable an interpretation of such changes in the context of disease or treatment induced modulations in functional networks.