A systematic review on the potential use of machine learning to classify major depressive disorder from healthy controls using resting state fMRI measures

Reinforcement learning processes as forecasters of depression remission

BACKGROUND

Aspects of reinforcement learning have been associated with specific depression symptoms and may inform the course of depressive illness.

METHODS

We applied support vector machines to investigate whether blood‑oxygen-level dependent (BOLD) responses linked with neural prediction error (nPE) and neural expected value (nEV) from a probabilistic learning task could forecast depression remission. We investigated whether predictions were moderated by treatment use or symptoms. Participants included 55 individuals (n = 39 female) with a depression diagnosis at baseline; 36 of these individuals completed standard cognitive behavioral therapy and 19 were followed during naturalistic course of illness. All participants were assessed for depression diagnosis at a follow-up visit.

RESULTS

Both nPE and nEV classifiers forecasted remission significantly better than null classifiers. The nEV classifier performed significantly better than the nPE classifier. We found no main or interaction effects of treatment status on nPE or nEV accuracy. We found a significant interaction between nPE-forecasted remission status and anhedonia, but not for negative affect or anxious arousal, when controlling for nEV-forecasted remission status.

LIMITATIONS

Our sample size, while comparable to that of other studies, limits options for maximizing and evaluating model performance. We addressed this with two standard methods for optimizing model performance (90:10 train and test scheme and bootstrapped sampling).

CONCLUSIONS

Results support nEV and nPE as relevant biobehavioral signals for understanding depression outcome independent of treatment status, with nEV being stronger than nPE as a predictor of remission. Reinforcement learning variables may be useful components of an individualized medicine framework for depression healthcare.

Diagnosis of major depressive disorder using a novel interpretable GCN model based on resting state fMRI

The diagnosis and analysis of major depressive disorder (MDD) faces some intractable challenges such as dataset limitations and clinical variability. Resting-state functional magnetic resonance imaging (Rs-fMRI) can reflect the fluctuation data of brain activity in a resting state, which can find the interrelationships, functional connections, and network characteristics among brain regions of the patients. In this paper, a brain functional connectivity matrix is constructed using Pearson correlation based on the characteristics of multi-site Rs-fMRI data and brain atlas, and an adaptive propagation operator graph convolutional network (APO-GCN) model is designed. The APO-GCN model can automatically adjust the propagation operator in each hidden layer according to the data features to control the expressive power of the model. By adaptively learning effective information in the graph, this model significantly improves its ability to capture complex graph structural patterns. The experimental results on Rs-fMRI data from 1601 participants (830 MDD and 771 HC) and 16 sites of REST-meta-MDD project show that the APO-GCN achieved a classification accuracy of 91.8%, outperforming those of the state-of-the-art classifier methods. The classification process is driven by multiple significant brain regions, and our method further reveals functional connectivity abnormalities between these brain regions, which are important biomarkers of classification. It is worth noting that the brain regions identified by the classifier and the networks involved are consistent with existing research results, which suggest that the pathogenesis of depression may be related to dysfunction of multiple brain networks.

Generative dynamical models for classification of rsfMRI data

The growing availability of large-scale neuroimaging datasets and user-friendly machine learning tools has led to a recent surge in studies that use fMRI data to predict psychological or behavioral variables. Many such studies classify fMRI data on the basis of static features, but fewer try to leverage brain dynamics for classification. Here, we pilot a generative, dynamical approach for classifying resting-state fMRI (rsfMRI) data. By fitting separate hidden Markov models to the classes in our training data and assigning class labels to test data based on their likelihood under those models, we are able to take advantage of dynamical patterns in the data without confronting the statistical limitations of some other dynamical approaches. Moreover, we demonstrate that hidden Markov models are able to successfully perform within-subject classification on the MyConnectome dataset solely on the basis of transition probabilities among their hidden states. On the other hand, individual Human Connectome Project subjects cannot be identified on the basis of hidden state transition probabilities alone-although a vector autoregressive model does achieve high performance. These results demonstrate a dynamical classification approach for rsfMRI data that shows promising performance, particularly for within-subject classification, and has the potential to afford greater interpretability than other approaches.

Abnormal ReHo and ALFF values in drug-naïve depressed patients with suicidal ideation or attempts: Evidence from the REST-meta-MDD consortium

BACKGROUND

The assessment of suicide risk in patients with major depressive disorder (MDD) is somewhat subjective in clinical diagnosis and may lead to diagnostic bias and serious consequences. Therefore, the aim of this study was to determine whether MDD patients with suicidal ideation or suicide attempts exhibited local brain functional synchrony and spontaneous activity intensity, thus providing certain imaging basis for suicide assessment.

METHODS

This study was conducted using ReHo and ALFF analyses on 213 first episode drug-naïve MDD patients from the REST-meta-MDD consortium. All patients were categorized into MDD with SI or SA group and MDD without SI and SA. A voxel-based two-sample t-test was then used to identify brain regions with significant differences in ReHo or ALFF values. Finally, Reho or ALFF values of those brain regions in MDD with SI or SA group were extracted for correlation analysis with suicide severity.

RESULTS

Compared with MDD patients without SI or SA, MDD patients with SI or SA had increased ReHo in the triangular part of left inferior frontal gyrus, orbital part of right inferior frontal gyrus and right precuneus gyrus, and increased ALFF in the middle occipital gyrus. All of these brain region characteristics were positively correlated with suicide severity on the HAMD 3th item score and HAMD 9th item score.

CONCLUSION

Our findings suggest that abnormalities of regional spontaneous brain activity were found in IFG, precuneus gyrus, and MOG among MDD patients with suicidal thoughts or attempts, which provides a reliable imaging basis for identifying and preventing suicide.

A simple but tough-to-beat baseline for fMRI time-series classification

Current neuroimaging studies frequently use complex machine learning models to classify human fMRI data, distinguishing healthy and disordered brains, often to validate new methods or enhance prediction accuracy. Yet, where prediction accuracy is a concern, our results suggest that precision in prediction does not always require such sophistication. When a classifier as simple as logistic regression is applied to feature-engineered fMRI data, it can match or even outperform more sophisticated recent models. Classification of the raw time series fMRI data generally benefits from complex parameter-rich models. However, this complexity often pushes them into the class of black-box models. Yet, we found that a relatively simple model can consistently outperform much more complex classifiers in both accuracy and speed. This model applies the same multi-layer perceptron repeatedly across time and averages the results. Thus, the complexity and black-box nature of the parameter rich models, often perceived as a necessary trade-off for higher performance, do not invariably yield superior results on fMRI. Given the success of straightforward approaches, we challenge the merit of research that concentrates solely on complex model development driven by classification. Instead, we advocate for increased focus on designing models that prioritize the explainability of fMRI data or pursue applicable objectives beyond mere classification accuracy, unless they significantly outperform logistic regression or our proposed model. To validate our claim, we explore possible reasons for the superior performance of our straightforward model by examining the innate characteristics of fMRI time series data. Our findings suggest that the sequential information hidden in the temporal order may be far less important for the accurate fMRI classification than the stand-alone pieces of information scattered across the frames of the time series.

Self-supervised graph contrastive learning with diffusion augmentation for functional MRI analysis and brain disorder detection

Resting-state functional magnetic resonance imaging (rs-fMRI) provides a non-invasive imaging technique to study patterns of brain activity, and is increasingly used to facilitate automated brain disorder analysis. Existing fMRI-based learning methods often rely on labeled data to construct learning models, while the data annotation process typically requires significant time and resource investment. Graph contrastive learning offers a promising solution to address the small labeled data issue, by augmenting fMRI time series for self-supervised learning. However, data augmentation strategies employed in these approaches may damage the original blood-oxygen-level-dependent (BOLD) signals, thus hindering subsequent fMRI feature extraction. In this paper, we propose a self-supervised graph contrastive learning framework with diffusion augmentation (GCDA) for functional MRI analysis. The GCDA consists of a pretext model and a task-specific model. In the pretext model, we first augment each brain functional connectivity network derived from fMRI through a graph diffusion augmentation (GDA) module, and then use two graph isomorphism networks with shared parameters to extract features in a self-supervised contrastive learning manner. The pretext model can be optimized without the need for labeled training data, while the GDA focuses on perturbing graph edges and nodes, thus preserving the integrity of original BOLD signals. The task-specific model involves fine-tuning the trained pretext model to adapt to downstream tasks. Experimental results on two rs-fMRI cohorts with a total of 1230 subjects demonstrate the effectiveness of our method compared with several state-of-the-arts.

Explainable Multi-Layer Dynamic Ensemble Framework Optimized for Depression Detection and Severity Assessment

BACKGROUND

Depression is a pervasive mental health condition, particularly affecting older adults, where early detection and intervention are essential to mitigate its impact. This study presents an explainable multi-layer dynamic ensemble framework designed to detect depression and assess its severity, aiming to improve diagnostic precision and provide insights into contributing health factors.

METHODS

Using data from the National Social Life, Health, and Aging Project (NSHAP), this framework combines classical machine learning models, static ensemble methods, and dynamic ensemble selection (DES) approaches across two stages: detection and severity prediction. The depression detection stage classifies individuals as normal or depressed, while the severity prediction stage further classifies depressed cases as mild or moderate-severe. Finally, a confirmation depression scale prediction model estimates depression severity scores to support the two stages. Explainable AI (XAI) techniques are applied to improve model interpretability, making the framework more suitable for clinical applications.

RESULTS

The framework's FIRE-KNOP DES algorithm demonstrated high efficacy, achieving 88.33% accuracy in depression detection and 83.68% in severity prediction. XAI analysis identified mental and non-mental health indicators as significant factors in the framework's performance, emphasizing the value of these features for accurate depression assessment.

CONCLUSIONS

This study emphasizes the potential of dynamic ensemble learning in mental health assessments, particularly in detecting and evaluating depression severity. The findings provide a strong foundation for future use of dynamic ensemble frameworks in mental health assessments, demonstrating their potential for practical clinical applications.

Classifying Unconscious, Psychedelic, and Neuropsychiatric Brain States with Functional Connectivity, Graph Theory, and Cortical Gradient Analysis

Accurate and generalizable classification of brain states is essential for understanding their neural underpinnings and improving clinical diagnostics. Traditionally, functional connectivity patterns and graph-theoretic metrics have been utilized. However, cortical gradient features, which reflect global brain organization, offer a complementary approach. We hypothesized that a machine learning model integrating these three feature sets would effectively discriminate between baseline and atypical brain states across a wide spectrum of conditions, even though the underlying neural mechanisms vary. To test this, we extracted features from brain states associated with three meta-conditions including unconsciousness (NREM2 sleep, propofol deep sedation, and propofol general anesthesia), psychedelic states induced by hallucinogens (subanesthetic ketamine, lysergic acid diethylamide, and nitrous oxide), and neuropsychiatric disorders (attention-deficit hyperactivity disorder, bipolar disorder, and schizophrenia). We used support vector machine with nested cross-validation to construct our models. The soft voting ensemble model marked the average balanced accuracy (average of specificity and sensitivity) of 79% (62-98% across all conditions), outperforming individual base models (70-76%). Notably, our models exhibited varying degrees of transferability across different datasets, with performance being dependent on the specific brain states and feature sets used. Feature importance analysis across meta-conditions suggests that the underlying neural mechanisms vary significantly, necessitating tailored approaches for accurate classification of specific brain states. This finding underscores the value of our feature-integrated ensemble models, which leverage the strengths of multiple feature types to achieve robust performance across a broader range of brain states. While our approach offers valuable insights into the neural signatures of different brain states, future work is needed to develop and validate even more generalizable models that can accurately classify brain states across a wider array of conditions.

An rs-fMRI based neuroimaging marker for adult absence epilepsy

OBJECTIVE

Approximately 20-30 % of epilepsy patients exhibit negative findings on routine magnetic resonance imaging, and this condition is known as nonlesional epilepsy. Absence epilepsy (AE) is a prevalent form of nonlesional epilepsy. This study aimed to investigate the clinical diagnostic utility of regional homogeneity (ReHo) assessed through the support vector machine (SVM) approach for identifying AE.

METHODS

This research involved 102 healthy individuals and 93 AE patients. Resting-state functional magnetic resonance imaging was employed for data acquisition in all participants. ReHo analysis, coupled with SVM methodology, was utilized for data processing.

RESULTS

Compared to healthy control individuals, AE patients demonstrated significantly elevated ReHo values in the bilateral putamen, accompanied by decreased ReHo in the bilateral thalamus. SVM was used to differentiate patients with AE from healthy control individuals based on rs-fMRI data. A composite assessment of altered ReHo in the left putamen and left thalamus yielded the highest accuracy at 81.64 %, with a sensitivity of 95.41 % and a specificity of 69.23 %.

SIGNIFICANCE

According to the results, altered ReHo values in the bilateral putamen and thalamus could serve as neuroimaging markers for AE, offering objective guidance for its diagnosis.

Intelligent classification of major depressive disorder using rs-fMRI of the posterior cingulate cortex

Major Depressive Disorder (MDD) is a widespread psychiatric condition that affects a significant portion of the global population. The classification and diagnosis of MDD is crucial for effective treatment. Traditional methods, based on clinical assessment, are subjective and rely on healthcare professionals' expertise. Recently, there's growing interest in using Resting-State Functional Magnetic Resonance Imaging (rs-fMRI) to objectively understand MDD's neurobiology, complementing traditional diagnostics. The posterior cingulate cortex (PCC) is a pivotal brain region implicated in MDD which could be used to identify MDD from healthy controls. Thus, this study presents an intelligent approach based on rs-fMRI data to enhance the classification of MDD. Original rs-fMRI data were collected from a cohort of 430 participants, comprising 197 patients and 233 healthy controls. Subsequently, the data underwent preprocessing using DPARSF, and the amplitudes of low-frequency fluctuation values were computed to reduce data dimensionality and feature count. Then data associated with the PCC were extracted. After eliminating redundant features, various types of Support Vector Machines (SVMs) were employed as classifiers for intelligent categorization. Ultimately, we compared the performance of each algorithm, along with its respective optimal classifier, based on classification accuracy, true positive rate, and the area under the receiver operating characteristic curve (AUC-ROC). Upon analyzing the comparison results, we determined that the Random Forest (RF) algorithm, in conjunction with a sophisticated Gaussian SVM classifier, demonstrated the highest performance. Remarkably, this combination achieved a classification accuracy of 81.9 % and a true positive rate of 92.9 %. In conclusion, our study improves the classification of MDD by supplementing traditional methods with rs-fMRI and machine learning techniques, offering deeper neurobiological insights and aiding accuracy, while emphasizing its role as an adjunct to clinical assessment.

Diagnostic machine learning applications on clinical populations using functional near infrared spectroscopy: a review

Functional near-infrared spectroscopy (fNIRS) and its interaction with machine learning (ML) is a popular research topic for the diagnostic classification of clinical disorders due to the lack of robust and objective biomarkers. This review provides an overview of research on psychiatric diseases by using fNIRS and ML. Article search was carried out and 45 studies were evaluated by considering their sample sizes, used features, ML methodology, and reported accuracy. To our best knowledge, this is the first review that reports diagnostic ML applications using fNIRS. We found that there has been an increasing trend to perform ML applications on fNIRS-based biomarker research since 2010. The most studied populations are schizophrenia (n = 12), attention deficit and hyperactivity disorder (n = 7), and autism spectrum disorder (n = 6) are the most studied populations. There is a significant negative correlation between sample size (>21) and accuracy values. Support vector machine (SVM) and deep learning (DL) approaches were the most popular classifier approaches (SVM = 20) (DL = 10). Eight of these studies recruited a number of participants more than 100 for classification. Concentration changes in oxy-hemoglobin (ΔHbO) based features were used more than concentration changes in deoxy-hemoglobin (ΔHb) based ones and the most popular ΔHbO-based features were mean ΔHbO (n = 11) and ΔHbO-based functional connections (n = 11). Using ML on fNIRS data might be a promising approach to reveal specific biomarkers for diagnostic classification.

Depressed patients with childhood maltreatment display altered intra- and inter-network resting state functional connectivity

BACKGROUND

Childhood maltreatment (CM) is a major risk factor for the development of major depressive disorder (MDD). To gain more knowledge on how adverse childhood experiences influence the development of brain architecture, we studied functional connectivity (FC) alterations of neural networks of depressed patients with, or without the history of CM.

METHODS

Depressed patients with severe childhood maltreatment (n = 18), MDD patients without maltreatment (n = 19), and matched healthy controls (n = 20) were examined with resting state functional MRI. History of maltreatment was assessed with the 28-item Childhood Trauma Questionnaire. Intra- and inter-network FC alterations were evaluated using FMRIB Software Library and CONN toolbox.

RESULTS

We found numerous intra- and inter-network FC alterations between the maltreated and the non-maltreated patients. Intra-network FC differences were found in the default mode, visual and auditory networks, and cerebellum. Network modelling revealed several inter-network FC alterations connecting the default mode network with the executive control, salience and cerebellar networks. Increased inter-network FC was found in maltreated patients between the sensory-motor and visual, cerebellar, default mode and salience networks.

LIMITATIONS

Relatively small sample size, cross-sectional design, and retrospective self-report questionnaire to assess adverse childhood experiences.

CONCLUSIONS

Our findings confirm that severely maltreated depressed patients display numerous alterations of intra- and inter-network FC strengths, not only in their fronto-limbic circuits, but also in sensory-motor, visual, auditory, and cerebellar networks. These functional alterations may explain that maltreated individuals typically display altered perception and are prone to develop functional neurological symptom disorder (conversion disorder) in adulthood.

Predicting treatment outcome based on resting-state functional connectivity in internalizing mental disorders: A systematic review and meta-analysis

Predicting treatment outcome in internalizing mental disorders prior to treatment initiation is pivotal for precision mental healthcare. In this regard, resting-state functional connectivity (rs-FC) and machine learning have often shown promising prediction accuracies. This systematic review and meta-analysis evaluates these studies, considering their risk of bias through the Prediction Model Study Risk of Bias Assessment Tool (PROBAST). We examined the predictive performance of features derived from rs-FC, identified features with the highest predictive value, and assessed the employed machine learning pipelines. We searched the electronic databases Scopus, PubMed and PsycINFO on the 12th of December 2022, which resulted in 13 included studies. The mean balanced accuracy for predicting treatment outcome was 77% (95% CI: [72%- 83%]). rs-FC of the dorsolateral prefrontal cortex had high predictive value in most studies. However, a high risk of bias was identified in all studies, compromising interpretability. Methodological recommendations are provided based on a comprehensive exploration of the studies' machine learning pipelines, and potential fruitful developments are discussed.

Altered cortical myelin in the salience and default mode networks in major depressive disorder patients: A surface-based analysis

INTRODUCTION

Evidence from previous genetic and post-mortem studies suggested that the myelination abnormality contributed to the pathogenesis of major depressive disorder (MDD). However, image-level alterations in cortical myelin content associated with MDD are still unclear.

METHODS

The high-resolution T1-weighted (T1w) and T2-weighted (T2w) brain 3D structural images were obtained from 52 MDD patients and 52 healthy controls (HC). We calculated the vertex-based T1w/T2w ratio using the HCP structural pipelines to characterize individual cortical myelin maps at the fs_LR 32 k surface. We attempted to detect the clusters with significant differences in cortical myelin content between MDD and HC groups. We correlated the cluster-wise averaged myelin value and the clinical performances in MDD patients.

RESULTS

The MDD patients showed significantly lower cortical myelin content in the cluster involving the left insula, orbitofrontal cortex, superior temporal cortex, transverse temporal gyrus, inferior frontal cortex, superior frontal gyrus, anterior cingulate cortex, precentral cortex, and postcentral cortex. The correlation analysis showed a significantly positive correlation between the cluster-wise cortical myelin content and the onset age of MDD patients.

CONCLUSION

The MDD patients showed lower cortical myelin content in regions of the default mode network regions and salience network than healthy controls.

Classification of recurrent major depressive disorder using a new time series feature extraction method through multisite rs-fMRI data

BACKGROUND

Major depressive disorder (MDD) has a high rate of recurrence. Identifying patients with recurrent MDD is advantageous in adopting prevention strategies to reduce the disabling effects of depression.

METHOD

We propose a novel feature extraction method that includes dynamic temporal information, and inputs the extracted features into a graph convolutional network (GCN) to achieve classification of recurrent MDD. We extract the average time series using an atlas from resting-state functional magnetic resonance imaging (fMRI) data. Pearson correlation was calculated between brain region sequences at each time point, representing the functional connectivity at each time point. The connectivity is used as the adjacency matrix and the brain region sequences as node features for a GCN model to classify recurrent MDD. Gradient-weighted Class Activation Mapping (Grad-CAM) was used to analyze the contribution of different brain regions to the model. Brain regions making greater contribution to classification were considered to be the regions with altered brain function in recurrent MDD.

RESULT

We achieved a classification accuracy of 75.8 % for recurrent MDD on the multi-site dataset, the Rest-meta-MDD. The brain regions closely related to recurrent MDD have been identified.

LIMITATION

The pre-processing stage may affect the final classification performance and harmonizing site differences may improve the classification performance.

CONCLUSION

The experimental results demonstrate that the proposed method can effectively classify recurrent MDD and extract dynamic changes of brain activity patterns in recurrent depression.

Brain volumes in alcohol use disorder: Do females and males differ? A whole-brain magnetic resonance imaging mega-analysis

Emerging evidence suggests distinct neurobiological correlates of alcohol use disorder (AUD) between sexes, which however remain largely unexplored. This work from ENIGMA Addiction Working Group aimed to characterize the sex differences in gray matter (GM) and white matter (WM) correlates of AUD using a whole-brain, voxel-based, multi-tissue mega-analytic approach, thereby extending our recent surface-based region of interest findings on a nearly matching sample using a complementary methodological approach. T1-weighted magnetic resonance imaging (MRI) data from 653 people with AUD and 326 controls was analyzed using voxel-based morphometry. The effects of group, sex, group-by-sex, and substance use severity in AUD on brain volumes were assessed using General Linear Models. Individuals with AUD relative to controls had lower GM volume in striatal, thalamic, cerebellar, and widespread cortical clusters. Group-by-sex effects were found in cerebellar GM and WM volumes, which were more affected by AUD in females than males. Smaller group-by-sex effects were also found in frontotemporal WM tracts, which were more affected in AUD females, and in temporo-occipital and midcingulate GM volumes, which were more affected in AUD males. AUD females but not males showed a negative association between monthly drinks and precentral GM volume. Our results suggest that AUD is associated with both shared and distinct widespread effects on GM and WM volumes in females and males. This evidence advances our previous region of interest knowledge, supporting the usefulness of adopting an exploratory perspective and the need to include sex as a relevant moderator variable in AUD.

Disrupted brain gray matter connectome in social anxiety disorder: a novel individualized structural covariance network analysis

Phenotyping approaches grounded in structural network science can offer insights into the neurobiological substrates of psychiatric diseases, but this remains to be clarified at the individual level in social anxiety disorder (SAD). Using a recently developed approach combining probability density estimation and Kullback-Leibler divergence, we constructed single-subject structural covariance networks (SCNs) based on multivariate morphometry (cortical thickness, surface area, curvature, and volume) and quantified their global/nodal network properties using graph-theoretical analysis. We compared network metrics between SAD patients and healthy controls (HC) and analyzed the relationship to clinical characteristics. We also used support vector machine analysis to explore the ability of graph-theoretical metrics to discriminate SAD patients from HC. Globally, SAD patients showed higher global efficiency, shorter characteristic path length, and stronger small-worldness. Locally, SAD patients showed abnormal nodal centrality mainly involving left superior frontal gyrus, right superior parietal lobe, left amygdala, right paracentral gyrus, right lingual, and right pericalcarine cortex. Altered topological metrics were associated with the symptom severity and duration. Graph-based metrics allowed single-subject classification of SAD versus HC with total accuracy of 78.7%. This finding, that the topological organization of SCNs in SAD patients is altered toward more randomized configurations, adds to our understanding of network-level neuropathology in SAD.

Prediction of suicidality in bipolar disorder using variability of intrinsic brain activity and machine learning

Bipolar disorder (BD) is associated with marked suicidal susceptibility, particularly during a major depressive episode. However, the evaluation of suicidal risk remains challenging since it relies mainly on self-reported information from patients. Hence, it is necessary to complement neuroimaging features with advanced machine learning techniques in order to predict suicidal behavior in BD patients. In this study, a total of 288 participants, including 75 BD suicide attempters, 101 BD nonattempters and 112 healthy controls, underwent a resting-state functional magnetic resonance imaging (rs-fMRI). Intrinsic brain activity was measured by amplitude of low-frequency fluctuation (ALFF). We trained and tested a two-level k-nearest neighbors (k-NN) model based on resting-state variability of ALFF with fivefold cross-validation. BD suicide attempters had increased dynamic ALFF values in the right anterior cingulate cortex, left thalamus and right precuneus. Compared to other machine learning methods, our proposed framework had a promising performance with 83.52% accuracy, 78.75% sensitivity and 87.50% specificity. The trained models could also replicate and validate the results in an independent cohort with 72.72% accuracy. These findings based on a relatively large data set, provide a promising way of combining fMRI data with machine learning technique to reliably predict suicide attempt at an individual level in bipolar depression. Overall, this work might enhance our understanding of the neurobiology of suicidal behavior by detecting clinically defined disruptions in the dynamics of instinct brain activity.

Frequency-resolved connectome alterations in major depressive disorder: A multisite resting fMRI study

BACKGROUND

Functional connectome studies have revealed widespread connectivity alterations in major depressive disorder (MDD). However, the low frequency bandpass filtering (0.01-0.08 Hz or 0.01-0.1 Hz) in most studies have impeded our understanding on whether and how these alterations are affected by frequency of interest.

METHODS

Here, we performed frequency-resolved (0.01-0.06 Hz, 0.06-0.16 Hz and 0.16-0.24 Hz) connectome analyses using a large-sample resting-state functional MRI dataset of 1002 MDD patients and 924 healthy controls from seven independent centers.

RESULTS

We reported significant frequency-dependent connectome alterations in MDD in left inferior parietal, inferior temporal, precentral, and fusiform cortices and bilateral precuneus. These frequency-dependent connectome alterations are mainly derived by abnormalities of medium- and long-distance connections and are brain network-dependent. Moreover, the connectome alteration of left precuneus in high frequency band (0.16-0.24 Hz) is significantly associated with illness duration.

LIMITATIONS

Multisite harmonization model only removed linear site effects. Neurobiological underpinning of alterations in higher frequency (0.16-0.24 Hz) should be further examined by combining fMRI data with respiration, heartbeat and blood flow recordings in future studies.

CONCLUSIONS

These results highlight the frequency-dependency of connectome alterations in MDD and the benefit of examining connectome alteration in MDD under a wider frequency band.

EEG coherences of the fronto-limbic circuit between patients with major depressive disorder and healthy controls

BACKGROUND

Imaging studies found that patients with major depressive disorder (MDD) showed abnormal functional connectivity in the fronto-limbic circuit, including the prefrontal cortex (PFC), anterior cingulate cortex (ACC), and limbic system (amygdala). This study used electroencephalography (EEG) coherence as an indicator of functional connectivity in the fronto-limbic circuit and examined the group differences between the MDD group and healthy controls (HC group), and the associations between EEG coherence and depressive symptoms.

METHODS

125 and 132 participants in the MDD and HC groups have measured the symptoms of depression and anxiety, and delta, theta, alpha, and beta1-beta4 EEG coherences in the fronto-limbic circuit and examined the differences between the two groups, and the associations between the EEG coherence and depressive symptoms were examined.

RESULTS

Lower theta, alpha, beta1, beta3, and beta4 coherence in the fronto-limbic circuit and higher beta2 coherence between the PFC and limbic system in the MDD group than in the HC group. Negative correlations between delta, theta, beta1, beta3, and beta4 coherence and total depression, cognitive depression, and somatic depression; positive correlations between beta2 coherences in the PFC and limbic system, and total depression and cognitive depression scores in the MDD group.

LIMITATIONS

Whether low EEG coherence in the fronto-limbic circuit is applicable to other subtypes of MDD requires further study.

CONCLUSIONS

Low EEG coherences in the fronto-limbic circuit were related to depressive symptoms, and increased functional connectivity in the fronto-limbic circuit can be applied by neurofeedback in future studies.