Discriminative analysis with a limited number of MEG trials in depression

Electrophysiological predictors of early response to antidepressants in major depressive disorder

BACKGROUND

Psychomotor retardation (PMR) is a core feature of major depressive disorder (MDD), which is characterized by abnormalities in motor control and cognitive processes. PMR in MDD can predict a poor antidepressant response, suggesting that PMR may serve as a marker of the antidepressant response. However, the neuropathological relationship between treatment outcomes and PMR remains uncertain. Thus, this study examined electrophysiological biomarkers associated with poor antidepressant response in MDD.

METHODS

A total of 142 subjects were enrolled in this study, including 49 healthy controls (HCs) and 93 MDD patients. All participants performed a simple right-hand visuomotor task during magnetoencephalography (MEG) scanning. Patients who exhibited at least a 50 % reduction in disorder severity at the endpoint (>2 weeks) were considered to be responders. Motor-related beta desynchronization (MRBD) and inter- and intra-hemispheric functional connectivity were measured in the bilateral motor network.

RESULTS

An increased MRBD and decreased inter- and intra-hemispheric functional connectivity in the motor network during movement were observed in non-responders, relative to responders and HCs. This dysregulation predicted the potential antidepressant response.

CONCLUSION

Abnormal local activity and functional connectivity in the motor network indicate poor psychomotor function, which might cause insensitivity to antidepressant treatment. This could be regarded as a potential neural mechanism for the prediction of a patient's treatment response.

Spontaneous beta power, motor-related beta power and cortical thickness in major depressive disorder with psychomotor disturbance

INTRODUCTION

The psychomotor disturbance is a common symptom in patients with major depressive disorder (MDD). The neurological mechanisms of psychomotor disturbance are intricate, involving alterations in the structure and function of motor-related regions. However, the relationship among changes in the spontaneous activity, motor-related activity, local cortical thickness, and psychomotor function remains unclear.

METHOD

A total of 140 patients with MDD and 68 healthy controls performed a simple right-hand visuomotor task during magnetoencephalography (MEG) scanning. All patients were divided into two groups according to the presence of psychomotor slowing. Spontaneous beta power, movement-related beta desynchronization (MRBD), absolute beta power during movement and cortical characteristics in the bilateral primary motor cortex were compared using general linear models with the group as a fixed effect and age as a covariate. Finally, the moderated mediation model was tested to examine the relationship between brain metrics with group differences and psychomotor performance.

RESULTS

The patients with psychomotor slowing showed higher spontaneous beta power, movement-related beta desynchronization and absolute beta power during movement than patients without psychomotor slowing. Compared with the other two groups, significant decreases were found in cortical thickness of the left primary motor cortex in patients with psychomotor slowing. Our moderated mediation model showed that the increased spontaneous beta power indirectly affected impaired psychomotor performance by abnormal MRBD, and the indirect effects were moderated by cortical thickness.

CONCLUSION

These results suggest that patients with MDD have aberrant cortical beta activity at rest and during movement, combined with abnormal cortical thickness, contributing to the psychomotor disturbance observed in this patient population.

An insight into diagnosis of depression using machine learning techniques: a systematic review

BACKGROUND

In this modern era, depression is one of the most prevalent mental disorders from which millions of individuals are affected today. The symptoms of depression are heterogeneous and often coincide with other disorders such as bipolar disorder, Parkinson's, schizophrenia, etc. It is a serious mental illness that may lead to other health problems if left untreated. Currently, identifying individuals with depression is totally based on the expertise of the clinician's experience. In order to assist clinicians in identifying the characteristics and classifying depressed people, different types of data modalities and machine learning techniques have been incorporated by researchers in this field. This study aims to find the answers to some important questions related to the trend of publications, data modality, machine learning models, dataset usage, pre-processing techniques and feature extraction and selection techniques that are prevalent and guide the direction of future research on depression diagnosis.

METHODS

This systematic review was conducted using a broad range of articles from two major databases: IEEE Xplore and PubMed. Studies ranging from the years 2011 to April 2021 were retrieved from the databases resulting in a total of 590 articles (53 articles from the IEEE Xplore database and 537 articles from the PubMed database). Out of those, the articles which satisfied the defined inclusion criteria were investigated for further analysis.

RESULTS

A total of 135 articles were identified and analysed for this review. High growth in the number of publications has been observed in recent years. Furthermore, significant diversity in the use of data modalities and machine learning classifiers has also been noted in this study. fMRI data with an SVM classifier was found to be the most popular choice among researchers. In most of the studies, data scarcity and small sample size, particularly for neuroimaging data are major concerns. The use of identical data pre-processing tools for similar data modalities can be seen. This study also provides statistical analysis of the current framework with respect to the modality, machine learning classifier, sample size and accuracy by applying one-way ANOVA and the Tukey - Kramer test.

CONCLUSION

The results indicate that an effective fusion of machine learning techniques with a potential data modality has a promising future for assisting clinicians in automatic depression diagnosis.

Treatment Effect of Exercise Intervention for Female College Students with Depression: Analysis of Electroencephalogram Microstates and Power Spectrum

This paper aims to assess the effect of exercise intervention on the improvement of college students with depression and to explore the change characteristics of microstates and the power spectrum in their resting-state electroencephalogram (EEG). Forty female college students with moderate depression were screened according to the Beck Depression Inventory-II (BDI-II) and Depression Self-Rating Scale (SDS) scores, and half of them received an exercise intervention for 18 weeks. The study utilized an EEG to define the resting-state networks, and the scores of all the participants were tracked during the intervention. Compared with those in the depression group, the power spectrum values in the θ and α bands were significantly decreased (p < 0.05), and the duration of microstate C increased significantly (p < 0.05), while the frequency of microstate B decreased significantly (p < 0.05) in the exercise intervention group. The transition probabilities showed that the exercise intervention group had a higher probability from B to D than those in the depression group (p < 0.01). In addition, the power of the δ and α bands were negatively correlated with the occurrence of microstate C (r = −0.842, p < 0.05 and r = −0.885, p < 0.01, respectively), and the power of the β band was positively correlated with the duration of microstate C (r = 0.900, p < 0.01) after exercise intervention. Our results suggest that the decreased duration of microstate C and the increased α power in depressed students are associated with reduced cognitive ability, emotional stability, and brain activity. Depression symptoms were notably improved after exercise intervention, thus providing a more scientific index for the research, rehabilitation mechanisms, and treatment of depression.

Structural-functional decoupling predicts suicide attempts in bipolar disorder patients with a current major depressive episode

Bipolar disorder (BD) is associated with a high risk of suicidality, and it is challenging to predict suicide attempts in clinical practice to date. Although structural and functional connectivity alterations from neuroimaging studies have been previously reported in BD with suicide attempts, little is known about how abnormal structural and functional connectivity relates to each other. Here, we hypothesize that structure connectivity constrains functional connectivity, and structural-functional coupling is a more sensitive biomarker to detect subtle brain abnormalities than any single modality in BD patients with a current major depressive episode who had attempted suicide. By investigating structural and resting-state fMRI connectivity, as well as their coupling among 191 BD depression patients with or without a history of suicide attempts and 113 healthy controls, we found that suicide attempters in BD depression patients showed significantly decreased central-temporal structural connectivity, increased frontal-temporal functional connectivity, along with decreased structural-functional coupling compared with non-suicide attempters. Crucially, the altered structural connectivity network predicted the abnormal functional connectivity network profile, and the structural-functional coupling was significantly correlated with suicide risk but not with depression or anxiety severity. Our findings suggest that the structural connectome is the key determinant of brain dysfunction, and structural-functional coupling could serve as a valuable trait-like biomarker for BD suicidal predication over and above the intramodality network connectivity. Such a measure can have clinical implications for early identification of suicide attempters with BD depression and inform strategies for prevention.

Magnetoencephalography resting-state spectral fingerprints distinguish bipolar depression and unipolar depression

OBJECTIVES

In clinical practice, bipolar depression (BD) and unipolar depression (UD) appear to have similar symptoms, causing BD being frequently misdiagnosed as UD, leading to improper treatment decision and outcome. Therefore, it is in urgent need of distinguishing BD from UD based on clinical objective biomarkers as early as possible. Here, we aimed to integrate brain neuroimaging data and an advanced machine learning technique to predict different types of mood disorder patients at the individual level.

METHODS

Eyes closed resting-state magnetoencephalography (MEG) data were collected from 23 BD, 30 UD, and 31 healthy controls (HC). Individual power spectra were estimated by Fourier transform, and statistic spectral differences were assessed via a cluster permutation test. A support vector machine classifier was further applied to predict different mood disorder types based on discriminative oscillatory power.

RESULTS

Both BD and UD showed decreased frontal-central gamma/beta ratios comparing to HC, in which gamma power (30-75 Hz) was decreased in BD while beta power (14-30 Hz) was increased in UD vs HC. The support vector machine model obtained significant high classification accuracies distinguishing three groups based on mean gamma and beta power (BD: 79.9%, UD: 81.1%, HC: 76.3%, P < .01).

CONCLUSIONS

In combination with resting-state MEG data and machine learning technique, it is possible to make an individual and objective prediction for mode disorder types, which in turn has implications for diagnosis precision and treatment decision of mood disorder patients.

Spectral fingerprints of facial affect processing bias in major depression disorder

In major depressive disorder (MDD), processing of facial affect is thought to reflect a perceptual bias (toward negative emotion, away from positive emotion, and interpretation of neutral as emotional). However, it is unclear to what extent and which specific perceptual bias is represented in MDD at the behavior and neuronal level. The present report examined 48 medication naive MDD patients and 41 healthy controls (HCs) performing a facial affect judgment task while magnetoencephalography was recorded. MDD patients were characterized by overall slower response times and lower perceptual judgment accuracies. In comparison with HC, MDD patients exhibited less somatosensory beta activity (20–30 Hz) suppression, more visual gamma activity (40–80 Hz) modulation and somatosensory beta and visual gamma interaction deficit. Moreover, frontal gamma activity during positive facial expression judgment was found to be negatively correlated with depression severity. Present findings suggest that perceptual bias in MDD is associated with distinct spatio-spectral manifestations on the neural level, which potentially establishes aberrant pathways during facial emotion processing and contributes to MDD pathology.

Hyperactive frontolimbic and frontocentral resting-state gamma connectivity in major depressive disorder

BACKGROUND

Major depressive disorder (MDD) is a system-level disorder affecting multiple functionally integrated cerebral networks. Nevertheless, their temporospatial organization and potential disturbance remain mostly unknown. The present report tested the hypothesis that deficient temporospatial network organization separates MDD and healthy controls (HC), and is linked to symptom severity of the disorder.

METHODS

Eyes-closed resting-state magnetoencephalographic (MEG) recordings were obtained from twenty-two MDD and twenty-two HC subjects. Beamforming source localization and functional connectivity analysis were applied to identify frequency-specific network interactions. Then, a novel virtual cortical resection approach was used to pinpoint putatively critical network controllers, accounting for aberrant cerebral connectivity patterns in MDD.

RESULTS

We found significantly elevated frontolimbic and frontocentral connectivity mediated by gamma (30-48 Hz) activity in MDD versus HC, and the right amygdala was the key differential network controller accounting for aberrant cerebral connectivity patterns in MDD. Furthermore, this frontolimbic and frontocentral gamma-band hyper-connectivity was positively correlated with depression severity.

LIMITATIONS

The overall sample size was small, and we found significant effects in the deep limbic regions with resting-state MEG, the reliability of which was difficult to corroborate further.

CONCLUSIONS

Overall, these findings support a notion that the right amygdala critically controls the exaggerated gamma-band frontolimbic and frontocentral connectivity in MDD during the resting-state condition, which potentially constitutes pre-established aberrant pathways during task processing and contributes to MDD pathology.

Abnormal early dynamic individual patterns of functional networks in low gamma band for depression recognition

BACKGROUND

The functional networks are associated with emotional processing in depression. The mapping of dynamic spatio-temporal brain networks is used to explore individual performance during early negative emotional processing. However, the dysfunctions of functional networks in low gamma band and their discriminative potentialities during early period of emotional face processing remain to be explored.

METHODS

Functional brain networks were constructed from the MEG recordings of 54 depressed patients and 54 controls in low gamma band (30-48 Hz). Dynamic connectivity regression (DCR) algorithm analyzed the individual change points of time series in response to emotional stimuli and constructed individualized spatio-temporal patterns. The nodal characteristics of patterns were calculated and fed into support vector machine (SVM). Performance of the classification algorithm in low gamma band was validated by dynamic topological characteristics of individual patterns in comparison to alpha and beta band.

RESULTS

The best discrimination accuracy of individual spatio-temporal patterns was 91.01% in low gamma band. Individual temporal patterns had better results compared to group-averaged temporal patterns in all bands. The most important discriminative networks included affective network (AN) and fronto-parietal network (FPN) in low gamma band.

LIMITATIONS

The sample size is relatively small. High gamma band was not considered.

CONCLUSIONS

The abnormal dynamic functional networks in low gamma band during early emotion processing enabled depression recognition. The individual information processing is crucial in the discovery of abnormal spatio-temporal patterns in depression during early negative emotional processing. Individual spatio-temporal patterns may reflect the real dynamic function of subjects while group-averaged data may neglect some individual information.

Multivariate analysis of single-sweep evoked brain potentials for pharmaco-electroencephalography

BACKGROUND AND AIMS

Current findings on altered evoked potentials (EPs) caused by morphine are based on common alterations for a group of subjects after drug administration. However, this ignores the analysis of individual responses, which may explain the clinical differences in efficacy. Therefore, we explored the individual responses to morphine in terms of the altered single-sweep characteristics in a placebo-controlled crossover study. To account for multifactorial mechanisms, several characteristics were assessed simultaneously by multivariate pattern analysis (MVPA).

METHODS

EPs were recorded from 62 channels and obtained before and after morphine and placebo administration during repeated electrical stimulations of the oesophagus in 12 healthy males. Additionally, the pain detection threshold was recorded to reflect the subjective analgesic effect in each subject. The characteristics of the sweeps were extracted by a multivariate matching pursuit algorithm with Gabor atoms implemented with a variable amplitude and constant phase across the sweeps. The single-sweep amplitudes were used as input to an MVPA algorithm to discriminate individual responses. The accuracy of the MVPA for each individual subject was used for correlation analysis of the analgesic effect.

RESULTS

The mean classification accuracy when discriminating pre- and posttreatment morphine responses was 72% (p = 0.01). The individual classification accuracy was positively correlated to the analgesic effect of morphine (p = 0.03). Furthermore, the 2 posttreatment responses were classified and validated by the classification of the 2 pretreatment responses (p = 0.001).

CONCLUSIONS

The alterations in the single-sweep EPs after morphine reflect the analgesic effect. The MVPA approach is a novel methodology for monitoring the individual efficacy of analgesics.

Dynamic functional-structural coupling within acute functional state change phases: Evidence from a depression recognition study

BACKGROUND

Dynamic functional-structural connectivity (FC-SC) coupling might reflect the flexibility by which SC relates to functional connectivity (FC). However, during the dynamic acute state change phases of FC, the relationship between FC and SC may be distinctive and embody the abnormality inherent in depression. This study investigated the depression-related inter-network FC-SC coupling within particular dynamic acute state change phases of FC.

METHODS

Magnetoencephalography (MEG) and diffusion tensor imaging (DTI) data were collected from 26 depressive patients (13 women) and 26 age-matched controls (13 women). We constructed functional brain networks based on MEG data and structural networks from DTI data. The dynamic connectivity regression algorithm was used to identify the state change points of a time series of inter-network FC. The time period of FC that contained change points were partitioned into types of dynamic phases (acute rising phase, acute falling phase,acute rising and falling phase and abrupt FC variation phase) to explore the inter-network FC-SC coupling. The selected FC-SC couplings were then fed into the support vector machine (SVM) for depression recognition.

RESULTS

The best discrimination accuracy was 82.7% (P=0.0069) with FC-SC couplings, particularly in the acute rising phase of FC. Within the FC phases of interest, the significant discriminative network pair was related to the salience network vs ventral attention network (SN-VAN) (P=0.0126) during the early rising phase (70-170ms).

LIMITATIONS

This study suffers from a small sample size, and the individual acute length of the state change phases was not considered.

CONCLUSIONS

The increased values of significant discriminative vectors of FC-SC coupling in depression suggested that the capacity to process negative emotion might be more directly related to the SC abnormally and be indicative of more stringent and less dynamic brain function in SN-VAN, especially in the acute rising phase of FC. We demonstrated that depressive brain dysfunctions could be better characterized by reduced FC-SC coupling flexibility in this particular phase.

Automated model selection in covariance estimation and spatial whitening of MEG and EEG signals

Magnetoencephalography and electroencephalography (M/EEG) measure non-invasively the weak electromagnetic fields induced by post-synaptic neural currents. The estimation of the spatial covariance of the signals recorded on M/EEG sensors is a building block of modern data analysis pipelines. Such covariance estimates are used in brain-computer interfaces (BCI) systems, in nearly all source localization methods for spatial whitening as well as for data covariance estimation in beamformers. The rationale for such models is that the signals can be modeled by a zero mean Gaussian distribution. While maximizing the Gaussian likelihood seems natural, it leads to a covariance estimate known as empirical covariance (EC). It turns out that the EC is a poor estimate of the true covariance when the number of samples is small. To address this issue the estimation needs to be regularized. The most common approach downweights off-diagonal coefficients, while more advanced regularization methods are based on shrinkage techniques or generative models with low rank assumptions: probabilistic PCA (PPCA) and factor analysis (FA). Using cross-validation all of these models can be tuned and compared based on Gaussian likelihood computed on unseen data. We investigated these models on simulations, one electroencephalography (EEG) dataset as well as magnetoencephalography (MEG) datasets from the most common MEG systems. First, our results demonstrate that different models can be the best, depending on the number of samples, heterogeneity of sensor types and noise properties. Second, we show that the models tuned by cross-validation are superior to models with hand-selected regularization. Hence, we propose an automated solution to the often overlooked problem of covariance estimation of M/EEG signals. The relevance of the procedure is demonstrated here for spatial whitening and source localization of MEG signals.