Whole-brain resting-state functional connectivity identified major depressive disorder: A multivariate pattern analysis in two independent samples

The association between anhedonia and prefrontal cortex activation in patients with major depression: a functional near-infrared spectroscopy study

This study investigated cerebral blood flow characteristics using functional nearinfrared spectroscopy (fNIRS) in agomelatine-treated depressed patients with anhedonia. The level of anhedonia was assessed by the Snaith Hamilton Rating Scale (SHAPS) and Montgomery Åsberg Depression Rating Scale 5-item (MADRS 5-item) score. All 84 patients were evaluated on the day of the study initiation and followed at week 1, 4 and 8 after the study initiation. Graph theory-based network analysis showed 2-back task-modulated global efficiency (adjusted B = 0.055, 95% CI = 0.043 - 0.066) and local efficiency (adjusted B = 0.066, 95% CI = 0.050 - 0.081) were significantly increased 1 week after treatment compared with the baseline. Furthermore, the increased oxy-hemoglobin (oxy-Hb) values from the baseline to the one week after treatment was positively related to the total MADRAS 5-item score reductions in the left-hemispheric orbitofrontal cortex (r = - 0.307, p = 0.005). Our findings suggest that abnormal functional connectivity over OFC may reflect the pathophysiological characteristics of anhedonia and serve as a clinical biomarker for anhedonia.

The Clinical Utility of Biomarkers in Diagnosing Major Depressive Disorder in Adults: A Systematic Review of Literature From 2013 to 2023

OBJECTIVE

The variety and efficacy of biomarkers available that may be used objectively to diagnose major depressive disorder (MDD) in adults are unclear. This systematic review aims to identify and evaluate the variety of objective markers used to diagnose MDD in adults.

METHODS

The search strategy was applied via PubMed and PsycINFO over the past 10 years (2013-2023) to capture the latest available evidence supporting the use of biomarkers to diagnose MDD. Data was reported through narrative synthesis.

RESULTS

Forty-two studies were included in the review. Findings were synthesised based on the following measures: blood, neuroimaging/neurophysiology, urine, dermatological, auditory, vocal, cerebrospinal fluid and combinatory-and evaluated based on its sensitivity/specificity and area under the curve values. The best predictors of blood (MYT1 gene), neuroimaging/neurophysiological (5-HT1A auto-receptor binding in the dorsal and median raphe), urinary (combined albumin, AMBP, HSPB, APOA1), cerebrospinal fluid-based (neuron specific enolase, microRNA) biomarkers were found to be closely linked to the pathophysiology of MDD.

CONCLUSION

A large variety of biomarkers were available to diagnose MDD, with the best performing biomarkers intrinsically related to the pathophysiology of MDD. Potential for future research lies in investigating the joint sensitivity of the best performing biomarkers identified via machine learning methods and establishing the causal effect between these biomarkers and MDD.

The protective effect of vitamin D supplementation as adjunctive therapy to antidepressants on brain structural and functional connectivity of patients with major depressive disorder: a randomized controlled trial

BACKGROUND

Growing evidence points to the pivotal role of vitamin D in the pathophysiology and treatment of major depressive disorder (MDD). However, there is a paucity of longitudinal research investigating the effects of vitamin D supplementation on the brain of MDD patients.

METHODS

We conducted a double-blind randomized controlled trial in 46 MDD patients, who were randomly allocated into either VD (antidepressant medication + vitamin D supplementation) or NVD (antidepressant medication + placebos) groups. Data from diffusion tensor imaging, resting-state functional MRI, serum vitamin D concentration, and clinical symptoms were obtained at baseline and after an average of 7 months of intervention.

RESULTS

Both VD and NVD groups showed significant improvement in depression and anxiety symptoms but with no significant differences between the two groups. However, a greater increase in serum vitamin D concentration was found to be associated with greater improvement in depression and anxiety symptoms in VD group. More importantly, neuroimaging data demonstrated disrupted white matter integrity of right inferior fronto-occipital fasciculus along with decreased functional connectivity between right frontoparietal and medial visual networks after intervention in NVD group, but no changes in VD group.

CONCLUSIONS

These findings suggest that vitamin D supplementation as adjunctive therapy to antidepressants may not only contribute to improvement in clinical symptoms but also help preserve brain structural and functional connectivity in MDD patients.

Altered functional brain activity in first-episode major depressive disorder treated with electro-acupuncture: A resting-state functional magnetic resonance imaging study

Background

Previous studies have found electroacupuncture could improve the clinical symptoms of first-episode major depressive disorder (MDD), but the exact neural mechanism of action needs to be further elucidated.

Methods

Twenty-eight first-episode MDD patients were randomly divided into 14 electro-acupuncture stimulation (EAS) groups and 14 sham-acupuncture stimulation (SAS) groups, and clinical symptoms were assessed and functional magnetic resonance imaging (fMRI) scans were done in both groups. Amplitude of low-frequency fluctuations (ALFF) was used to observe the changes between the pre-treatment and post-treatment in the two groups, and the altered brain areas were selected as region of interest (ROI) to observe the FC changes. Meanwhile, the correlation between the altered clinical symptoms and the altered ALFF and FC of brain regions in the two groups was analyzed.

Results

The EAS significantly decreased the HAMD-24 and HAMA-14 scores of MDD than SAS group. The imaging results revealed that both groups were able to increase the ALFF of the left middle temporal gyrus and the left cerebellar posterior lobe. When using the left middle temporal gyrus and the left posterior cerebellar lobe as ROIs, EAS group increased the FC between the left middle temporal gyrus with the left superior frontal gyrus, the left middle frontal gyrus, and the left hippocampus, and decreased the FC between the left posterior cerebellar lobe and the left calcarine gyrus, while SAS group only increased the FC between the left middle temporal gyrus with the left superior frontal gyrus. The alternations in clinical symptoms after EAS treatment were positively correlated with the altered ALFF values in the left middle temporal gyrus and the altered FC values in the left middle temporal gyrus and the left middle frontal gyrus.

Conclusion

EA demonstrates modulation of functional activity in the default mode network (DMN), sensorimotor network (SMN), cognitive control network (CCN), limbic system, and visual network (VN) for the treatment of the first-episode MDD. Our findings contribute to the neuroimaging evidence for the efficacy of EAS.

Resting-state cerebral blood flow and functional connectivity abnormalities in depressed patients with childhood maltreatment: Potential biomarkers of vulnerability?

AIM

Childhood maltreatment (CM) is an important risk factor for major depressive disorder (MDD). This study aimed to explore the specific effect of CM on cerebral blood flow (CBF) and brain functional connectivity (FC) in MDD patients.

METHODS

A total of 150 subjects were collected including 55 MDD patients with CM, 34 MDD patients without CM, 19 healthy controls (HC) with CM, and 42 HC without CM. All subjects completed MRI scans and neuropsychological tests. Two-way analysis of covariance was used to detect the main and interactive effects of disease and CM on CBF and FC across subjects. Then, partial correlation analyses were conducted to explore the behavioral significance of altered CBF and FC in MDD patients. Finally, a support vector classifier model was applied to differentiate MDD patients.

RESULTS

MDD patients represented increased CBF in bilateral temporal lobe and decreased CBF in right visual cortex. Importantly, significant depression-by-CM interactive effects on CBF were primarily located in the frontoparietal regions, including orbitofrontal cortex (OFC), lateral prefrontal cortex (PFC), and parietal cortex. Moreover, significant FC abnormalities were seen in OFC-PFC and frontoparietal-visual cortex. Notably, the abnormal CBF and FC were significantly associated with behavioral performance. Finally, a combination of altered CBF and FC behaved with a satisfactory classification ability to differentiate MDD patients.

CONCLUSIONS

These results highlight the importance of frontoparietal and visual cortices for MDD with CM experience, proposing a potential neuroimaging biomarker for MDD identification.

A Siamese Network With Node Convolution for Individualized Predictions Based on Connectivity Maps Extracted From Resting-State fMRI Data

Deep learning has demonstrated great potential for objective diagnosis of neuropsychiatric disorders based on neuroimaging data, which includes the promising resting-state functional magnetic resonance imaging (RS-fMRI). However, the insufficient sample size has long been a bottleneck for deep model training for the purpose. In this study, we proposed a Siamese network with node convolution (SNNC) for individualized predictions based on RS-fMRI data. With the involvement of Siamese network, which uses sample pair (rather than a single sample) as input, the problem of insufficient sample size can largely be alleviated. To adapt to connectivity maps extracted from RS-fMRI data, we applied node convolution to each of the two branches of the Siamese network. For regression purposes, we replaced the contrastive loss in classic Siamese network with the mean square error loss and thus enabled Siamese network to quantitatively predict label differences. The label of a test sample can be predicted based on any of the training samples, by adding the label of the training sample to the predicted label difference between them. The final prediction for a test sample in this study was made by averaging the predictions based on each of the training samples. The performance of the proposed SNNC was evaluated with age and IQ predictions based on a public dataset (Cam-CAN). The results indicated that SNNC can make effective predictions even with a sample size of as small as 40, and SNNC achieved state-of-the-art accuracy among a variety of deep models and standard machine learning approaches.

Sampling inequalities affect generalization of neuroimaging-based diagnostic classifiers in psychiatry

BACKGROUND

The development of machine learning models for aiding in the diagnosis of mental disorder is recognized as a significant breakthrough in the field of psychiatry. However, clinical practice of such models remains a challenge, with poor generalizability being a major limitation.

METHODS

Here, we conducted a pre-registered meta-research assessment on neuroimaging-based models in the psychiatric literature, quantitatively examining global and regional sampling issues over recent decades, from a view that has been relatively underexplored. A total of 476 studies (n = 118,137) were included in the current assessment. Based on these findings, we built a comprehensive 5-star rating system to quantitatively evaluate the quality of existing machine learning models for psychiatric diagnoses.

RESULTS

A global sampling inequality in these models was revealed quantitatively (sampling Gini coefficient (G) = 0.81, p < .01), varying across different countries (regions) (e.g., China, G = 0.47; the USA, G = 0.58; Germany, G = 0.78; the UK, G = 0.87). Furthermore, the severity of this sampling inequality was significantly predicted by national economic levels (β =  - 2.75, p < .001, R²_adj = 0.40; r =  - .84, 95% CI: - .41 to - .97), and was plausibly predictable for model performance, with higher sampling inequality for reporting higher classification accuracy. Further analyses showed that lack of independent testing (84.24% of models, 95% CI: 81.0-87.5%), improper cross-validation (51.68% of models, 95% CI: 47.2-56.2%), and poor technical transparency (87.8% of models, 95% CI: 84.9-90.8%)/availability (80.88% of models, 95% CI: 77.3-84.4%) are prevailing in current diagnostic classifiers despite improvements over time. Relating to these observations, model performances were found decreased in studies with independent cross-country sampling validations (all p < .001, BF₁₀ > 15). In light of this, we proposed a purpose-built quantitative assessment checklist, which demonstrated that the overall ratings of these models increased by publication year but were negatively associated with model performance.

CONCLUSIONS

Together, improving sampling economic equality and hence the quality of machine learning models may be a crucial facet to plausibly translating neuroimaging-based diagnostic classifiers into clinical practice.

ICA-based Individualized Differential Structure Similarity Networks for Predicting Symptom Scores in Adolescents with Major Depressive Disorder

Major depressive disorder (MDD) is a complex mood disorder characterized by persistent and overwhelming depression. Previous studies have identified large scale structural brain alterations in MDD, yet most are group analyses with atlas-parcellated anatomical regions. Here we proposed a method to measure individual difference by independent component analysis (ICA)-based individual difference structural similarity network (IDSSN). This approach provided a data-adaptive, atlas-free solution that can be applied to new individual subjects. Specifically, we constructed individualized whole-brain structural covariance networks based on network perturbation approach using spatially constrained ICA. First, a set of benchmark independent components (ICs) were generated using gray matter volume (GMV) from all healthy controls. Then individual heterogeneity was obtained by calculating differences and other similarity metrics between ICs derived from "each one patient + all controls" and the benchmark ICs, resulting in 32 imaging features and structural similarity networks for each patient, which can be used for predicting multiple clinical symptoms. We estimated IDSSN for 189 adolescent MDD patients aged 10-20 years and compared them to 112 healthy adolescents. We tested their predictability of the Hamilton Anxiety Scale , the 17-item Hamilton Depression Scale and six clinical syndromes using connectome-based predictive modeling. The prediction results showed that ICA-based IDSSN features reveal more disease-specific information than those using other brain templates. We also found that depression-associated networks mainly involved the default-mode network and visual network. In conclusion, our study proposed an adaptive method that improves the ability to detect GMV deviations and specificity between one individual patient and healthy groups, providing a new perspectives and insights for evaluating individual-level clinical heterogeneity based on brain structures.

Frequency specificity of aberrant triple networks in major depressive disorder: a resting-state effective connectivity study

Major depressive disorder (MDD) has been associated with aberrant effective connectivity (EC) among the default mode network (DMN), salience network (SN), and central executive network (CEN)-collectively referred to as triple networks. However, prior research has predominantly concentrated on broad frequency bands (0.01-0.08 Hz or 0.01-0.15 Hz), ignoring the influence of distinct rhythms on triple network causal dynamics. In the present study, we aim to investigate EC alterations within the triple networks across various frequency bands in patients with MDD. Utilizing a data-driven frequency decomposition approach and a multivariate Granger causality analysis, we characterized frequency-specific EC patterns of triple networks in 49 MDD patients and 54 healthy controls. A support vector machine classifier was subsequently employed to assess the discriminative capacity of the frequency-specific EC features. Our findings revealed that, compared to controls, patients exhibited not only enhanced mean EC within the CEN in the conventional frequency band (0.01-0.08 Hz), but also decreased mean EC from the SN to the DMN in a higher frequency band (0.12-0.18 Hz), and increased mean EC from the CEN to the SN in a sub-frequency band (0.04-0.08 Hz); the latter was significantly correlated with disease severity. Moreover, optimal classification performance for distinguishing patients from controls was attained by combining EC features across all three frequency bands, with the area under the curve (AUC) value of 0.8831 and the corresponding accuracy, sensitivity, and specificity of 89.97%, 92.63%, and 87.32%, respectively. These insights into EC changes within the triple networks across multiple frequency bands offer valuable perspectives on the neurobiological basis of MDD and could aid in developing frequency-specific EC features as potential biomarkers for disease diagnosis.

Evaluation of Risk of Bias in Neuroimaging-Based Artificial Intelligence Models for Psychiatric Diagnosis: A Systematic Review

IMPORTANCE

Neuroimaging-based artificial intelligence (AI) diagnostic models have proliferated in psychiatry. However, their clinical applicability and reporting quality (ie, feasibility) for clinical practice have not been systematically evaluated.

OBJECTIVE

To systematically assess the risk of bias (ROB) and reporting quality of neuroimaging-based AI models for psychiatric diagnosis.

EVIDENCE REVIEW

PubMed was searched for peer-reviewed, full-length articles published between January 1, 1990, and March 16, 2022. Studies aimed at developing or validating neuroimaging-based AI models for clinical diagnosis of psychiatric disorders were included. Reference lists were further searched for suitable original studies. Data extraction followed the CHARMS (Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies) and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines. A closed-loop cross-sequential design was used for quality control. The PROBAST (Prediction Model Risk of Bias Assessment Tool) and modified CLEAR (Checklist for Evaluation of Image-Based Artificial Intelligence Reports) benchmarks were used to systematically evaluate ROB and reporting quality.

FINDINGS

A total of 517 studies presenting 555 AI models were included and evaluated. Of these models, 461 (83.1%; 95% CI, 80.0%-86.2%) were rated as having a high overall ROB based on the PROBAST. The ROB was particular high in the analysis domain, including inadequate sample size (398 of 555 models [71.7%; 95% CI, 68.0%-75.6%]), poor model performance examination (with 100% of models lacking calibration examination), and lack of handling data complexity (550 of 555 models [99.1%; 95% CI, 98.3%-99.9%]). None of the AI models was perceived to be applicable to clinical practices. Overall reporting completeness (ie, number of reported items/number of total items) for the AI models was 61.2% (95% CI, 60.6%-61.8%), and the completeness was poorest for the technical assessment domain with 39.9% (95% CI, 38.8%-41.1%).

CONCLUSIONS AND RELEVANCE

This systematic review found that the clinical applicability and feasibility of neuroimaging-based AI models for psychiatric diagnosis were challenged by a high ROB and poor reporting quality. Particularly in the analysis domain, ROB in AI diagnostic models should be addressed before clinical application.

Unsupervised cross-domain functional MRI adaptation for automated major depressive disorder identification

Resting-state functional magnetic resonance imaging (rs-fMRI) data have been widely used for automated diagnosis of brain disorders such as major depressive disorder (MDD) to assist in timely intervention. Multi-site fMRI data have been increasingly employed to augment sample size and improve statistical power for investigating MDD. However, previous studies usually suffer from significant inter-site heterogeneity caused for instance by differences in scanners and/or scanning protocols. To address this issue, we develop a novel discrepancy-based unsupervised cross-domain fMRI adaptation framework (called UFA-Net) for automated MDD identification. The proposed UFA-Net is designed to model spatio-temporal fMRI patterns of labeled source and unlabeled target samples via an attention-guided graph convolution module, and also leverage a maximum mean discrepancy constrained module for unsupervised cross-site feature alignment between two domains. To the best of our knowledge, this is one of the first attempts to explore unsupervised rs-fMRI adaptation for cross-site MDD identification. Extensive evaluation on 681 subjects from two imaging sites shows that the proposed method outperforms several state-of-the-art methods. Our method helps localize disease-associated functional connectivity abnormalities and is therefore well interpretable and can facilitate fMRI-based analysis of MDD in clinical practice.

Association between cognitive impairments and aberrant dynamism of overlapping brain sub-networks in unmedicated major depressive disorder: A resting-state MEG study

OBJECTIVE

Little is known about the pathogenesis underlying cognitive impairment in major depressive disorder (MDD). We aimed to explore the mechanisms of cognitive impairments among patients with MDD by investigating the dynamics of overlapping brain sub-networks.

METHODS

Forty unmedicated patients with MDD and 28 healthy controls (HC) were enrolled in this study. Cognitive function was measured using the Chinese versions of MATRICS Consensus Cognitive Battery (MCCB). All participants were scanned using a whole-head resting-state magnetoencephalography (MEG) machine. The dynamism of neural sub-networks was analyzed based on the detection of overlapping communities in five frequency bands of oscillatory brain signals.

RESULTS

MDD demonstrated poorer cognitive performance in six domains compared to HC. The difference in community detection (functional integration mode) in MDD was frequency-dependent. MDD showed significantly decreased community dynamics in all frequency bands compared to HC. Specifically, differences in the visual network (VN) and default mode network (DMN) were detected in all frequency bands, differences in the cognitive control network (CCN) were detected in the alpha2 and beta frequency bands, and differences in the bilateral limbic network (BLN) were only detected in the beta frequency band. Moreover, community dynamics in the alpha2 frequency band were positively correlated with verbal learning and reasoning problem solving abilities in MDD.

CONCLUSIONS

Our study found that decreasing in the dynamics of overlapping sub-networks may differ by frequency bands. The aberrant dynamics of overlapping neural sub-networks revealed by frequency-specific MEG signals may provide new information on the mechanism of cognitive impairments that result from MDD.

Investigating cognitive flexibility deficit in schizophrenia using task-based whole-brain functional connectivity

Background

Cognitive flexibility is a core cognitive control function supported by the brain networks of the whole-brain. Schizophrenic patients show deficits in cognitive flexibility in conditions such as task-switching. A large number of neuroimaging studies have revealed abnormalities in local brain activations associated with deficits in cognitive flexibility in schizophrenia, but the relationship between impaired cognitive flexibility and the whole-brain functional connectivity (FC) pattern is unclear.

Method

We investigated the task-based functional connectivity of the whole-brain in patients with schizophrenia and healthy controls during task-switching. Multivariate pattern analysis (MVPA) was utilized to investigate whether the FC pattern can be used as a feature to discriminate schizophrenia patients from healthy controls. Graph theory analysis was further used to quantify the degrees of integration and segregation in the whole-brain networks to interpret the different reconfiguration patterns of brain networks in schizophrenia patients and healthy controls.

Results

The results showed that the FC pattern classified schizophrenia patients and healthy controls with significant accuracy. Moreover, the altered whole-brain functional connectivity pattern was driven by a lower degree of network integration and segregation in schizophrenia, indicating that both global and local information transfers at the entire-network level were less efficient in schizophrenia patients than in healthy controls during task-switching processing.

Conclusion

These results investigated the group differences in FC profiles during task-switching and not only elucidated that FC patterns are changed in schizophrenic patients, suggesting that task-based FC could be used as a potential neuromarker to discriminate schizophrenia patients from healthy controls in cognitive flexibility but also provide increased insight into the brain network organization that may contribute to impaired cognitive flexibility.

The pattern glare and visual memory are disrupted in patients with major depressive disorder

BACKGROUND

Visual memory impairment is one of the most commonly complained symptoms in patients with major depressive disorder (MDD). Pattern glare is also a distorted visual phenomenon that puzzles patients with MDD. Nevertheless, how these two phenomena interact in MDD remains unknown. This study investigated the association between pattern glare and visual memory in MDD patients.

METHODS

Sixty-two patients with MDD and forty-nine age-, sex- and education level-matched healthy controls (HCs) were included in this study. The Pattern Recognition Memory (PRM) test and the Brief Visual Memory Test-Revised (BVMT-R) were applied to measure visual memory. The pattern glare test including three patterns with different spatial frequencies (SFs) was used to explore pattern glare levels.

RESULTS

Patients with MDD scored lower on the PRM-PCi, BVMT-R1, BVMT-R2, BVMT-R3, and BVMT-Rt and higher on the PRM-MCLd than HCs (all p < 0.05). Pattern glare scores for MDD patients were higher with mid-SF (p < 0.001), high-SF (p = 0.006) and mid-high SF differences (p = 0.01) than for HCs. A positive correlation between mid-SF and PRM-MCLd scores in all participants was observed (p = 0.01, r = 0.246). A negative correlation between mid-high difference scores and BVMT-R2 scores (p = 0.032, r = -0.317) was observed in HCs, but no significant correlation was observed in MDD patients.

CONCLUSIONS

The present study showed that visual memory and pattern glare are disrupted in MDD. Visual memory may be associated with pattern glare and needs to be studied in future work.

Fixed-effects inference and tests of correlation for longitudinal functional data

We propose an inferential framework for fixed effects in longitudinal functional models and introduce tests for the correlation structures induced by the longitudinal sampling procedure. The framework provides a natural extension of standard longitudinal correlation models for scalar observations to functional observations. Using simulation studies, we compare fixed effects estimation under correctly and incorrectly specified correlation structures and also test the longitudinal correlation structure. Finally, we apply the proposed methods to a longitudinal functional dataset on physical activity. The computer code for the proposed method is available at https://github.com/rli20ST758/FILF.

A voice without a mouth no more: The neurobiology of language and consciousness

Most research on the neurobiology of language ignores consciousness and vice versa. Here, language, with an emphasis on inner speech, is hypothesised to generate and sustain self-awareness, i.e., higher-order consciousness. Converging evidence supporting this hypothesis is reviewed. To account for these findings, a 'HOLISTIC' model of neurobiology of language, inner speech, and consciousness is proposed. It involves a 'core' set of inner speech production regions that initiate the experience of feeling and hearing words. These take on affective qualities, deriving from activation of associated sensory, motor, and emotional representations, involving a largely unconscious dynamic 'periphery', distributed throughout the whole brain. Responding to those words forms the basis for sustained network activity, involving 'default mode' activation and prefrontal and thalamic/brainstem selection of contextually relevant responses. Evidence for the model is reviewed, supporting neuroimaging meta-analyses conducted, and comparisons with other theories of consciousness made. The HOLISTIC model constitutes a more parsimonious and complete account of the 'neural correlates of consciousness' that has implications for a mechanistic account of mental health and wellbeing.

Formation and Schema Analysis of Oil Painting Style Based on Texture and Color Texture Features under Few Shot

Texture has strong expressiveness in picture art, and color texture features play an important role in composition. Together with texture, they can convey the artistic connotation of portrait, especially in oil painting. Therefore, in order to make the picture form oil painting style and oil painting schema, we need to study the texture and color texture in combination with the previous oil painting art images. But now, there are few samples of good oil paintings, so it is difficult to study the texture and color texture in oil paintings. Therefore, in order to form a unique artistic style of modern oil painting and promote the development of modern oil painting art, this paper studies the texture and color texture characteristics in the environment of few oil painting works. This paper establishes a model through deep neural network to extract the image incentive and color texture of oil painting art works, which provides guidance for promoting the development of oil painting art. The experiments in this paper show that the depth neural network has high definition for the extraction of texture and color texture of small sample oil painting images, which can reach more than 85%. It has high guiding significance for the research and creation of oil painting art.

A Multivariate Functional Connectivity Approach to Mapping Brain Networks and Imputing Neural Activity in Mice

Temporal correlation analysis of spontaneous brain activity (e.g., Pearson "functional connectivity," FC) has provided insights into the functional organization of the human brain. However, bivariate analysis techniques such as this are often susceptible to confounding physiological processes (e.g., sleep, Mayer-waves, breathing, motion), which makes it difficult to accurately map connectivity in health and disease as these physiological processes affect FC. In contrast, a multivariate approach to imputing individual neural networks from spontaneous neuroimaging data could be influential to our conceptual understanding of FC and provide performance advantages. Therefore, we analyzed neural calcium imaging data from Thy1-GCaMP6f mice while either awake, asleep, anesthetized, during low and high bouts of motion, or before and after photothrombotic stroke. A linear support vector regression approach was used to determine the optimal weights for integrating the signals from the remaining pixels to accurately predict neural activity in a region of interest (ROI). The resultant weight maps for each ROI were interpreted as multivariate functional connectivity (MFC), resembled anatomical connectivity, and demonstrated a sparser set of strong focused positive connections than traditional FC. While global variations in data have large effects on standard correlation FC analysis, the MFC mapping methods were mostly impervious. Lastly, MFC analysis provided a more powerful connectivity deficit detection following stroke compared to traditional FC.

Thalamic altered spontaneous activity and connectivity in obstructive sleep apnea syndrome

BACKGROUND AND PURPOSE

Obstructive sleep apnea (OSA) syndrome is a sleep disorder characterized by excessive snoring, repetitive apneas, and nocturnal arousals, that leads to fragmented sleep and intermittent nocturnal hypoxemia. Morphometric and functional brain alterations in cortical and subcortical structures have been documented in these patients via magnetic resonance imaging (MRI), even if correlational data between the alterations in the brain and cognitive and clinical indexes are still not reported.

METHODS

We examined the impact of OSA on brain spontaneous activity by measuring the fractional amplitude of low-frequency fluctuations (fALFF) in resting-state functional MRI data of 20 drug-naïve patients with OSA syndrome and 20 healthy controls matched for age, gender, and body mass index.

RESULTS

Patients showed a pattern of significantly abnormal subcortical functional activity as compared to controls, with increased activity selectively involving the thalami, specifically their intrinsic nuclei connected to somatosensory and motor-premotor cortical regions. Using these nuclei as seed regions, the subsequent functional connectivity analysis highlighted an increase in patients' thalamocortical connectivity at rest. Additionally, the correlation between fALFF and polysomnographic data revealed a possible link between OSA severity and fALFF of regions belonging to the central autonomic network.

CONCLUSIONS

Our results suggest a hyperactivation in thalamic diurnal activity in patients with OSA syndrome, which we interpret as a possible consequence of increased thalamocortical circuitry activation during nighttime due to repeated arousals.

A two-center radiomic analysis for differentiating major depressive disorder using multi-modality MRI data under different parcellation methods

BACKGROUND

The present study aimed to explore the difference in the brain function and structure between patients with major depressive disorder (MDD) and healthy controls (HCs) using two-center and multi-modal MRI data, which would be helpful to investigate the pathogenesis of MDD.

METHODS

The subjects were collected from two hospitals. One including 140 patients with MDD and 138 HCs was used as primary cohort. Another one including 29 patients with MDD and 52 HCs was used as validation cohort. Functional and structural magnetic resonance images (MRI) were acquired to extract four types of features: functional connectivity (FC), amplitude of low-frequency fluctuations (ALFF), regional homogeneity (ReHo), and gray matter volume (GMV). Then classifiers using different combinations among the four types of selected features were respectively built to discriminate patients from HCs. Different templates were applied and the results under different templates were compared.

RESULTS

The classifier built with the combination of FC, ALFF, and GMV under the AAL template discriminated patients from HCs with the best performance (AUC=0.916, ACC=84.8%). The regions selected in all the different templates were mainly located in the default mode network, affective network, prefrontal cortex.

LIMITATIONS

First, the sample size of the validation cohort was limited. Second, diffusion tensor imaging data were not collected.

CONCLUSION

The performance of classifier was improved by using multi-modal MRI imaging. Different templates would be suitable for different types of analysis. The regions selected in all the different templates are possibly the core regions to investigate the pathophysiology of MDD.

Hippocampal Resting-State Functional Connectivity Forecasts Individual Posttraumatic Stress Disorder Symptoms: A Data-Driven Approach

BACKGROUND

Posttraumatic stress disorder (PTSD) is a debilitating disorder, and there is no current accurate prediction of who develops it after trauma. Neurobiologically, individuals with chronic PTSD exhibit aberrant resting-state functional connectivity (rsFC) between the hippocampus and other brain regions (e.g., amygdala, prefrontal cortex, posterior cingulate), and these aberrations correlate with severity of illness. Previous small-scale research (n < 25) has also shown that hippocampal rsFC measured acutely after trauma is predictive of future severity using a region-of-interest-based approach. While this is a promising biomarker, to date, no study has used a data-driven approach to test whole-brain hippocampal FC patterns in forecasting the development of PTSD symptoms.

METHODS

A total of 98 adults at risk of PTSD were recruited from the emergency department after traumatic injury and completed resting-state functional magnetic resonance imaging (8 min) within 1 month; 6 months later, they completed the Clinician-Administered PTSD Scale for DSM-5 for assessment of PTSD symptom severity. Whole-brain rsFC values with bilateral hippocampi were extracted (using CONN) and used in a machine learning kernel ridge regression analysis (PRoNTo); a k-folds (k = 10) and 70/30 testing versus training split approach were used for cross-validation (1000 iterations to bootstrap confidence intervals for significance values).

RESULTS

Acute hippocampal rsFC significantly predicted Clinician-Administered PTSD Scale for DSM-5 scores at 6 months (r = 0.30, p = .006; mean squared error = 120.58, p = .006; R2 = 0.09, p = .025). In post hoc analyses, hippocampal rsFC remained significant after controlling for demographics, PTSD symptoms at baseline, and depression, anxiety, and stress severity at 6 months (B = 0.59, SE = 0.20, p = .003).

CONCLUSIONS

Findings suggest that functional connectivity of the hippocampus across the brain acutely after traumatic injury is associated with prospective PTSD symptom severity.

A neural network approach to optimising treatments for depression using data from specialist and community psychiatric services in Australia, New Zealand and Japan

This study investigated the application of a recurrent neural network for optimising pharmacological treatment for depression. A clinical dataset of 458 participants from specialist and community psychiatric services in Australia, New Zealand and Japan were extracted from an existing custom-built, web-based tool called Psynary . This data, which included baseline and self-completed reviews, was used to train and refine a novel algorithm which was a fully connected network feature extractor and long short-term memory algorithm was firstly trained in isolation and then integrated and annealed using slow learning rates due to the low dimensionality of the data. The accuracy of predicting depression remission before processing patient review data was 49.8%. After processing only 2 reviews, the accuracy was 76.5%. When considering a change in medication, the precision of changing medications was 97.4% and the recall was 71.4% . The medications with predicted best results were antipsychotics (88%) and selective serotonin reuptake inhibitors (87.9%). This is the first study that has created an all-in-one algorithm for optimising treatments for all subtypes of depression. Reducing treatment optimisation time for patients suffering with depression may lead to earlier remission and hence reduce the high levels of disability associated with the condition. Furthermore, in a setting where mental health conditions are increasing strain on mental health services, the utilisation of web-based tools for remote monitoring and machine/deep learning algorithms may assist clinicians in both specialist and primary care in extending specialist mental healthcare to a larger patient community.

Abnormal core functional connectivity on the pathology of MDD and antidepressant treatment: A systematic review

RATIONALE/IMPORTANCE

Researches have highlighted communication deficits between resting-state brain networks in major depressive disorder (MDD), as reflected in abnormal functional connectivity (FC). However, it is unclear whether impaired FC is associated with MDD pathology or is simply incidental to MDD symptoms. Moreover, there is no generalized theory to analyze the impact of treatment modalities on MDD.

OBJECTIVES

To address the issues, we conducted a systematic review of 49 eligible papers to provide insight into the pathological mechanisms of MDD patients by summarizing resting-state FC alterations involving mood and cognitive abnormalities and the effects of medications on them.

RESULTS

Mood disorders in MDD were characterized by abnormal FC between the amygdala, insula, anterior cingulate cortex (ACC), and prefrontal cortex (PFC). Cognitive impairment manifests as deficits in executive function, attention, memory, and rumination, primarily modulated by dysfunction between the fronto-parietal network and default mode network. Especially, we proposed the set of core abnormal FC (CA-FC) contributing to mood and cognitive impairment in MDD, currently including ACC-left precuneus/amygdala, rostral ACC-left dorsolateral PFC, left subgenual ACC-left cerebellar, left PFC- anterior subcallosal, and left precuneus-left pulvinar. After treatment, patients with normalized CA-FC showed remission of depressive symptoms.

CONCLUSIONS

We propose a CA-FC set for possible causative principle of MDD, which unifies the FC results from specific, difficult-to-analyze conditions into one outcome set for screening. Furthermore, CA-FC varies from person to person, and the low success rate of a single treatment may be due to the inability to cover too many CA-FC.

AUTOMATED RECOGNITION OF DEPRESSION FROM FEWER-SHOT LEANING IN RESTING-STATE fMRI WITH ReHo USING DEEP CONVOLUTIONAL NEURAL NETWORK

Up to now, there is still the absence of research about depression recognition using resting-state functional magnetic resonance imaging (rest_fMRI) and deep learning. Previous studies have shown that regional homogeneity (ReHo) of rest_fMRI (rest_ReHo_fMRI) is a characterization of the functional synchronization of adjacent voxels in brain regions, and the mental and behavioral abnormalities in depression are due to an imbalance of ReHo synchronization in some brain functional areas. Accordingly, this paper presents a method for depression recognition using rest_ReHo_fMRI. First, the rest_ReHo_fMRI is extracted from the preprocessed rest-fMRI by calculation. Then, deep convolutional networks (such as VGG16) pretrained on ImageNet are used to automatically complete extracting the classification features from rest_ReHo_fMRI. Finally, the Kernel Extreme Learning Machine (KELM) was used to classify the depression. The results of the test set show that the proposed method achieves 89.07% in sensitivity and 89.74% in specificity. This study suggests that features of rest_ReHo_fMRI can be used as biomarkers to distinguish depression from normal people.

Machine Learning-Based Definition of Symptom Clusters and Selection of Antidepressants for Depressive Syndrome

The current polythetic and operational criteria for major depression inevitably contribute to the heterogeneity of depressive syndromes. The heterogeneity of depressive syndrome has been criticized using the concept of language game in Wittgensteinian philosophy. Moreover, "a symptom- or endophenotype-based approach, rather than a diagnosis-based approach, has been proposed" as the "next-generation treatment for mental disorders" by Thomas Insel. Understanding the heterogeneity renders promise for personalized medicine to treat cases of depressive syndrome, in terms of both defining symptom clusters and selecting antidepressants. Machine learning algorithms have emerged as a tool for personalized medicine by handling clinical big data that can be used as predictors for subtype classification and treatment outcome prediction. The large clinical cohort data from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D), Combining Medications to Enhance Depression Outcome (CO-MED), and the German Research Network on Depression (GRND) have recently began to be acknowledged as useful sources for machine learning-based depression research with regard to cost effectiveness and generalizability. In addition, noninvasive biological tools such as functional and resting state magnetic resonance imaging techniques are widely combined with machine learning methods to detect intrinsic endophenotypes of depression. This review highlights recent studies that have used clinical cohort or brain imaging data and have addressed machine learning-based approaches to defining symptom clusters and selecting antidepressants. Potentially applicable suggestions to realize machine learning-based personalized medicine for depressive syndrome are also provided herein.

Cross-network interaction for diagnosis of major depressive disorder based on resting state functional connectivity

Previous studies have suggested that resting-state functional connectivity plays a central role in the physiopathology of major depressive disorder (MDD). However, the individualized diagnosis of MDD based on resting-state functional connectivity is still unclear, especially in first episode drug-naive patients with MDD. Resting state functional magnetic resonance imaging was enrolled from 30 first episode drug-naive patients with MDD and age- and gender-matched 31 healthy controls. Whole brain functional connectivity was computed and viewed as classification features. Multivariate pattern analysis (MVPA) was performed to discriminate patients with MDD from controls. The experimental results exhibited a correct classification rate of 82.25% (p < 0.001) with sensitivity of 83.87% and specificity of 80.64%. Almost all of the consensus connections (125/128) were cross-network interaction among default mode network (DMN), salience network (SN), central executive network (CEN), visual cortex network (VN), Cerebellum and Other. Moreover, the supramarginal gyrus exhibited high discriminative power in classification. Our findings suggested cross-network interaction can be used as an effective biomarker for MDD clinical diagnosis, which may reveal the potential pathological mechanism for major depression. The current study further confirmed reliable application of MVPA in discriminating MDD patients from healthy controls.

Classification of Unmedicated Bipolar Disorder Using Whole-Brain Functional Activity and Connectivity: A Radiomics Analysis

The aim of this study was to develop and validate a method of disease classification for bipolar disorder (BD) by functional activity and connectivity using radiomics analysis. Ninety patients with unmedicated BD II as well as 117 healthy controls underwent resting-state functional magnetic resonance imaging (rs-fMRI). A total of 4 types of 7018 features were extracted after preprocessing, including mean regional homogeneity (mReHo), mean amplitude of low-frequency fluctuation (mALFF), resting-state functional connectivity (RSFC), and voxel-mirrored homotopic connectivity (VMHC). Then, predictive features were selected by Mann-Whitney U test and removing variables with a high correlation. Least absolute shrinkage and selection operator (LASSO) method was further used to select features. At last, support vector machine (SVM) model was used to estimate the state of each subject based on the selected features after LASSO. Sixty-five features including 54 RSFCs, 7 mALFFs, 1 mReHo, and 3 VMHCs were selected. The accuracy and area under curve (AUC) of the SVM model built based on the 65 features is 87.3% and 0.919 in the training dataset, respectively, and the accuracy and AUC of this model validated in the validation dataset is 80.5% and 0.838, respectively. These findings demonstrate a valid radiomics approach by rs-fMRI can identify BD individuals from healthy controls with a high classification accuracy, providing the potential adjunctive approach to clinical diagnostic systems.

Migraine Is More Than Just Headache: Is the Link to Chronic Fatigue and Mood Disorders Simply Due to Shared Biological Systems?

Migraine is a symptomatically heterogeneous condition, of which headache is just one manifestation. Migraine is a disorder of altered sensory thresholding, with hypersensitivity among sufferers to sensory input. Advances in functional neuroimaging have highlighted that several brain areas are involved even prior to pain onset. Clinically, patients can experience symptoms hours to days prior to migraine pain, which can warn of impending headache. These symptoms can include mood and cognitive change, fatigue, and neck discomfort. Some epidemiological studies have suggested that migraine is associated in a bidirectional fashion with other disorders, such as mood disorders and chronic fatigue, as well as with other pain conditions such as fibromyalgia. This review will focus on the literature surrounding alterations in fatigue, mood, and cognition in particular, in association with migraine, and the suggested links to disorders such as chronic fatigue syndrome and depression. We hypothesize that migraine should be considered a neural disorder of brain function, in which alterations in aminergic networks integrating the limbic system with the sensory and homeostatic systems occur early and persist after headache resolution and perhaps interictally. The associations with some of these other disorders may allude to the inherent sensory sensitivity of the migraine brain and shared neurobiology and neurotransmitter systems rather than true co-morbidity.

A review of resting-state fMRI and its use to examine psychiatric disorders

Resting-state fMRI (rs-fMRI) has emerged as an alternative method to study brain function in human and animal models. In humans, it has been widely used to study psychiatric disorders including schizophrenia, bipolar disorder, autism spectrum disorders, and attention deficit hyperactivity disorders. In this review, rs-fMRI and its advantages over task based fMRI, its currently used analysis methods, and its application in psychiatric disorders using different analysis methods are discussed. Finally, several limitations and challenges of rs-fMRI applications are also discussed.

Translational application of neuroimaging in major depressive disorder: a review of psychoradiological studies

Major depressive disorder (MDD) causes great decrements in health and quality of life with increments in healthcare costs, but the causes and pathogenesis of depression remain largely unknown, which greatly prevent its early detection and effective treatment. With the advancement of neuroimaging approaches, numerous functional and structural alterations in the brain have been detected in MDD and more recently attempts have been made to apply these findings to clinical practice. In this review, we provide an updated summary of the progress in translational application of psychoradiological findings in MDD with a specified focus on potential clinical usage. The foreseeable clinical applications for different MRI modalities were introduced according to their role in disorder classification, subtyping, and prediction. While evidence of cerebral structural and functional changes associated with MDD classification and subtyping was heterogeneous and/or sparse, the ACC and hippocampus have been consistently suggested to be important biomarkers in predicting treatment selection and treatment response. These findings underlined the potential utility of brain biomarkers for clinical practice.

Exploring the Brain Lateralization in ADHD Based on Variability of Resting-State fMRI Signal

Objective: In this study, we investigate the brain lateralization in ADHD patients. Furthermore, we also explore the difference between male and female patients, and the difference among distinct ADHD subtypes, that is, ADHD-inattentive (ADHD-IA) and ADHD-combined (ADHD-C). Method: We employed the standard deviation to quantify the variability of resting-state functional magnetic resonance imaging (fMRI) signal and measure the lateralization index (LI). Results: ADHD patients showed significantly increased rightward lateralization in the inferior frontal gyrus (opercular), precuneus, and paracentral lobule, and decreased rightward lateralization in the insula. Compared with male patients, female patients showed significantly rightward lateralization in the putamen and lobule VII of cerebellar hemisphere. ADHD-C patients exhibited increased rightward lateralization in the inferior frontal gyrus (opercular), and decreased rightward lateralization in the inferior temporal gyrus, as compared with ADHD-IA. The LI was also found to be related to inattentive and hyper/impulsive scores. Conclusion: These key findings may aid in understanding the pathology of ADHD.

The alteration of cognitive function networks in remitted patients with major depressive disorder: an independent component analysis

INTRODUCTION

Dysfunctional connectivity of resting-state functional networks has been observed in patients with major depressive disorder (MDD), particularly in cognitive function networks including the central executive network (CEN), default mode network (DMN) and salience network (SN). Findings from studies examining how aberrant functional connectivity (FC) changed after antidepressant treatment, however, have been inconsistent. Thus, the purpose of the present study was to explore potential mechanisms of altered cognitive function networks during resting-state between remitted major depressive disorder (rMDD) patients and healthy controls (HCs) and furthermore, the relationship between dysfunctional connectivity patterns in rMDD and clinical symptoms.

METHODOLOGY

In this study, 19 HCs and 19 rMDD patients were recruited for resting-state functional magnetic resonance imaging (fMRI) scanning. FC was evaluated with independent component analysis for CEN, DMN and SN. Two sample t tests were conducted to compare differences between rMDD and HCs. A Pearson correlation analysis was also performed to examine the relationship between connectivity of networks and cognitive function scores and clinical symptoms.

RESULTS

Compared to healthy controls, remitted patients showed lower connectivity in CEN, mostly in the superior frontal gyrus (SFG), middle frontal gyrus (MFG), inferior parietal lobule (IPL) and part of the supramarginal gyrus (SMG). Conversely, the bilateral insula, part of the SMG (a key node of the CEN) and dorsal anterior cingulate cortex (dACC) of the DMN showed higher connectivity in rMDD patients. Pearson correlation results demonstrated that connectivity of the right IPL in CEN was positively correlated with cognitive function scores, and connectivity of the left insula was negatively correlated with BDI scores.

CONCLUSIONS

Though rMDD patients reached the standard of clinal remission, unique impairments of FC in cognitive function networks remained. Aberrant FC between cognitive function networks responsible for executive control was observed in rMDD and may be associated with residual clinical symptoms.

Predicting Response to Group Cognitive Behavioral Therapy in Asthma by a Small Number of Abnormal Resting-State Functional Connections

Group cognitive behavioral therapy (GCBT) is a successful psychotherapy for asthma. However, response varies considerably among individuals, and identifying biomarkers of GCBT has been challenging. Thus, the aim of this study was to predict an individual's potential response by using machine learning algorithms and functional connectivity (FC) and to improve the personalized treatment of GCBT. We use the lasso method to make the feature selection in the functional connections between brain regions, and we utilize t-test method to test the significant difference of these selected features. The feature selections are performed between controls (size = 20) and pre-GCBT patients (size = 20), pre-GCBT patients (size = 10) and post-GCBT patients (size = 10), and post-GCBT patients (size = 10) and controls (size = 10). Depending on these features, support vector classification was used to classify controls and pre- and post-GCBT patients. Pearson correlation analysis was employed to analyze the associations between clinical symptoms and the selected discriminated FCs in post-GCBT patients. At last, linear support vector regression was applied to predict the therapeutic effect of GCBT. After feature selection and significant analysis, five discriminated FC regarding neuroimaging biomarkers of GCBT were discovered, which are also correlated with clinical symptoms. Using these discriminated functional connections, we could accurately classify the patients before and after GCBT (classification accuracy, 80%) and predict the therapeutic effect of GCBT in asthma (predicted accuracy, 67.8%). The findings in this study would provide a novel sight toward GCBT response prediction and further confirm neural underpinnings of asthma. Moreover, our findings had clinical implications for personalized treatment by identifying asthmatic patients who will be appropriate for GCBT.

CLINICAL TRIAL REGISTRATION

The brain mechanisms of group cognitive behavioral therapy to improve the symptoms of asthma (Registration number: Chi-CTR-15007442, http://www.chictr.org.cn/index.aspx).

Identification of internet gaming disorder individuals based on ventral tegmental area resting-state functional connectivity

Objective neuroimaging markers are imminently in need for more accurate clinical diagnosis of Internet gaming disorder (IGD). Recent neuroimaging evidence suggested that IGD is associated with abnormalities in the mesolimbic dopamine (DA) system. As the key nodes of the DA pathways, ventral tegmental area (VTA) and substantia nigra (SN) and their connected brain regions may serve as potential markers to identify IGD. Therefore, we aimed to develop optimal classifiers to identify IGD individuals by using VTA and bilateral SN resting-state functional connectivity (RSFC) patterns. A dataset including 146 adolescents (66 IGDs and 80 healthy controls (HCs)) was used to build classification models and another independent dataset including 28 subjects (14 IGDs and 14 HCs) was employed to validate the generalization ability of the models. Multi-voxel pattern analysis (MVPA) with linear support vector machine (SVM) was used to select the features. Our results demonstrated that the VTA RSFC circuits successfully identified IGD individuals (mean accuracy: 86.1%, mean sensitivity: 84.5%, mean specificity: 86.6%, the mean area under the receiver operating characteristic curve: 0.91). Furthermore, the independent generalization ability of the VTA RSFC classifier model was also satisfied (accuracy = 78.5%, sensitivity = 71.4%, specificity = 85.8%). The VTA connectivity circuits that were selected as distinguishing features were mainly included bilateral thalamus, right hippocampus, right pallidum, right temporal pole superior gyrus and bilateral temporal superior gyrus. These findings demonstrated that the potential of the resting-state neuroimaging features of VTA RSFC as objective biomarkers for the IGD clinical diagnosis in the future.

Identifying resting-state effective connectivity abnormalities in drug-naïve major depressive disorder diagnosis via graph convolutional networks

Major depressive disorder (MDD) is a leading cause of disability; its symptoms interfere with social, occupational, interpersonal, and academic functioning. However, the diagnosis of MDD is still made by phenomenological approach. The advent of neuroimaging techniques allowed numerous studies to use resting-state functional magnetic resonance imaging (rs-fMRI) and estimate functional connectivity for brain-disease identification. Recently, attempts have been made to investigate effective connectivity (EC) that represents causal relations among regions of interest. In the meantime, to identify meaningful phenotypes for clinical diagnosis, graph-based approaches such as graph convolutional networks (GCNs) have been leveraged recently to explore complex pairwise similarities in imaging/nonimaging features among subjects. In this study, we validate the use of EC for MDD identification by estimating its measures via a group sparse representation along with a structured equation modeling approach in a whole-brain data-driven manner from rs-fMRI. To distinguish drug-naïve MDD patients from healthy controls, we utilize spectral GCNs based on a population graph to successfully integrate EC and nonimaging phenotypic information. Furthermore, we devise a novel sensitivity analysis method to investigate the discriminant connections for MDD identification in our trained GCNs. Our experimental results validated the effectiveness of our method in various scenarios, and we identified altered connectivities associated with the diagnosis of MDD.

Towards a brain-based predictome of mental illness

Neuroimaging-based approaches have been extensively applied to study mental illness in recent years and have deepened our understanding of both cognitively healthy and disordered brain structure and function. Recent advancements in machine learning techniques have shown promising outcomes for individualized prediction and characterization of patients with psychiatric disorders. Studies have utilized features from a variety of neuroimaging modalities, including structural, functional, and diffusion magnetic resonance imaging data, as well as jointly estimated features from multiple modalities, to assess patients with heterogeneous mental disorders, such as schizophrenia and autism. We use the term "predictome" to describe the use of multivariate brain network features from one or more neuroimaging modalities to predict mental illness. In the predictome, multiple brain network-based features (either from the same modality or multiple modalities) are incorporated into a predictive model to jointly estimate features that are unique to a disorder and predict subjects accordingly. To date, more than 650 studies have been published on subject-level prediction focusing on psychiatric disorders. We have surveyed about 250 studies including schizophrenia, major depression, bipolar disorder, autism spectrum disorder, attention-deficit hyperactivity disorder, obsessive-compulsive disorder, social anxiety disorder, posttraumatic stress disorder, and substance dependence. In this review, we present a comprehensive review of recent neuroimaging-based predictomic approaches, current trends, and common shortcomings and share our vision for future directions.

Quantitative Identification of Major Depression Based on Resting-State Dynamic Functional Connectivity: A Machine Learning Approach

Introduction

Developing a machine learning-based approach which could provide quantitative identification of major depressive disorder (MDD) is essential for the diagnosis and intervention of this disorder. However, the performances of traditional algorithms using static functional connectivity (SFC) measures were unsatisfactory. In the present work, we exploit the hidden information embedded in dynamic functional connectivity (DFC) and developed an accurate and objective image-based diagnosis system for MDD.

Methods

MRI images were collected from 99 participants including 56 healthy controls and 43 MDD patients. DFC was calculated using a sliding-window algorithm. A non-linear support vector machine (SVM) approach was then used with the DFC matrices as features to distinguish MDD patients from healthy controls. The spatiotemporal characteristics of the most discriminative features were then investigated.

Results

The area under the curve (AUC) of the SVM classifier with DFC measures reached 0.9913, while this value is only 0.8685 for the algorithm using SFC measures. Spatially, the most discriminative 28 connections distributed in the visual network (VN), somatomotor network (SMN), dorsal attention network (DAN), ventral attention network (VAN), limbic network (LN), frontoparietal network (FPN), and default mode network (DMN), etc. Notably, a large portion of these connections were associated with the FPN, DMN, and VN. Temporally, the most discriminative connections transited from the cortex to deeper regions.

Conclusion

The results clearly suggested that DFC is superior to SFC and provide a reliable quantitative identification method for MDD. Our findings may furnish a better understanding of the neural mechanisms of MDD as well as improve accurate diagnosis and early intervention of this disorder.

Abnormal Dynamic Functional Network Connectivity and Graph Theoretical Analysis in Major Depressive Disorder

Major depressive disorder (MDD) is a complex mood disorder characterized by persistent and overwhelming depression. Previous studies have identified abnormalities in large scale functional brain networks in MDD, yet most of them were based on static functional connectivity. By contrast, here we explored disrupted topological organization of dynamic functional network connectivity (dFNC) in MDD based on graph theory. 182 MDD patients and 218 healthy controls were included in this study, all Chinese Han people. By applying group information guided independent component analysis (GIG-ICA) on resting-state fMRI data, the dFNCs of each subject were estimated using a sliding window method and k-means clustering. Five dynamic functional states were identified, three of which demonstrated significant group difference on the percentage of state occurrence. Interestingly, MDD patients spent much more time in a weakly-connected state 2, which is associated with self-focused thinking, a representative feature of depression. In addition, the abnormal FNCs in MDD were observed connecting different networks, especially among prefrontal, sensorimotor and cerebellum networks. As to network properties, MDD patients exhibited increased node efficiency in prefrontal and cerebellum. Moreover, three dFNCs with disrupted node properties were commonly identified in different states, which are also correlated with depressive symptom severity and cognitive performance. This study is the first attempt to investigate the dynamic functional abnormalities in Chinese MDD using a relatively large sample size, which provides new evidence on aberrant time-varying brain activity and its network disruptions in MDD, which might underscore the impaired cognitive functions in this mental disorder.

Machine learning in major depression: From classification to treatment outcome prediction

AIMS

Major depression disorder (MDD) is the single greatest cause of disability and morbidity, and affects about 10% of the population worldwide. Currently, there are no clinically useful diagnostic biomarkers that are able to confirm a diagnosis of MDD from bipolar disorder (BD) in the early depressive episode. Therefore, exploring translational biomarkers of mood disorders based on machine learning is in pressing need, though it is challenging, but with great potential to improve our understanding of these disorders.

DISCUSSIONS

In this study, we review popular machine-learning methods used for brain imaging classification and predictions, and provide an overview of studies, specifically for MDD, that have used magnetic resonance imaging data to either (a) classify MDDs from controls or other mood disorders or (b) investigate treatment outcome predictors for individual patients. Finally, challenges, future directions, and potential limitations related to MDD biomarker identification are also discussed, with a goal of offering a comprehensive overview that may help readers to better understand the applications of neuroimaging data mining in depression.

CONCLUSIONS

We hope such efforts may highlight the need for an urgently needed paradigm shift in treatment, to guide personalized optimal clinical care.

Aberrant Dynamic Functional Network Connectivity and Graph Properties in Major Depressive Disorder

Major depressive disorder (MDD) is a complex mood disorder characterized by persistent and overwhelming depression. Previous studies have identified abnormalities in large scale functional brain networks in MDD, yet most of them were based on static functional connectivity. In contrast, here we explored disrupted topological organization of dynamic functional network connectivity (dFNC) in MDD based on graph theory. One hundred and eighty-two MDD patients and 218 healthy controls were included in this study, all Chinese Han people. By applying group information guided independent component analysis (GIG-ICA) to resting-state functional magnetic resonance imaging (fMRI) data, the dFNCs of each subject were estimated using a sliding window method and k-means clustering. Network properties including global efficiency, local efficiency, node strength and harmonic centrality, were calculated for each subject. Five dynamic functional states were identified, three of which demonstrated significant group differences in their percentage of state occurrence. Interestingly, MDD patients spent much more time in a weakly-connected State 2, which includes regions previously associated with self-focused thinking, a representative feature of depression. In addition, the FNCs in MDD were connected differently in different states, especially among prefrontal, sensorimotor, and cerebellum networks. MDD patients exhibited significantly reduced harmonic centrality primarily involving parietal lobule, lingual gyrus and thalamus. Moreover, three dFNCs with disrupted node properties were commonly identified in different states, and also correlated with depressive symptom severity and cognitive performance. This study is the first attempt to investigate the dynamic functional abnormalities in MDD in a Chinese population using a relatively large sample size, which provides new evidence on aberrant time-varying brain activity and its network disruptions in MDD, which might underscore the impaired cognitive functions in this mental disorder.

The Neural Association between Tendency to Forgive and Spontaneous Brain Activity in Healthy Young Adults

The tendency to forgive (TTF) refers to one’s global dispositional level of forgiveness across situations and relationships. Previous brain imaging studies examined activation patterns underlying forgiving process, yet the association between individual differences in the TTF and spontaneous brain activity at resting-state remains unknown. In this study, resting-state functional magnetic resonance imaging (fMRI) was used to investigate the correlation between the TTF and spontaneous brain activity in a young adult sample. Participants were 178 young students (55 men) who completed the TTF scale and underwent a resting-state fMRI scan. Multiple regression analysis was conducted to assess the association between the regional amplitude of low-frequency fluctuations (ALFF) and TTF scores corrected for age and sex. Results showed that the ALFF value in the right dorsomedial prefrontal cortex (dmPFC), precuneus and inferior parietal lobule (IPL) were negatively associated with TTF scores. These findings suggest that the spontaneous brain activity of brain regions like the dmPFC, precuneus and IPL which are implicated in mentalizing and empathic response are associated with individual differences in the TTF.

Sample heterogeneity in unipolar depression as assessed by functional connectivity analyses is dominated by general disease effects

OBJECTIVES

Combinations of resting-state fMRI and machine-learning techniques are increasingly employed to develop diagnostic models for mental disorders. However, little is known about the neurobiological heterogeneity of depression and diagnostic machine learning has mainly been tested in homogeneous samples. Our main objective was to explore the inherent structure of a diverse unipolar depression sample. The secondary objective was to assess, if such information can improve diagnostic classification.

MATERIALS AND METHODS

We analyzed data from 360 patients with unipolar depression and 360 non-depressed population controls, who were subdivided into two independent subsets. Cluster analyses (unsupervised learning) of functional connectivity were used to generate hypotheses about potential patient subgroups from the first subset. The relationship of clusters with demographical and clinical measures was assessed. Subsequently, diagnostic classifiers (supervised learning), which incorporated information about these putative depression subgroups, were trained.

RESULTS

Exploratory cluster analyses revealed two weakly separable subgroups of depressed patients. These subgroups differed in the average duration of depression and in the proportion of patients with concurrently severe depression and anxiety symptoms. The diagnostic classification models performed at chance level.

LIMITATIONS

It remains unresolved, if subgroups represent distinct biological subtypes, variability of continuous clinical variables or in part an overfitting of sparsely structured data.

CONCLUSIONS

Functional connectivity in unipolar depression is associated with general disease effects. Cluster analyses provide hypotheses about potential depression subtypes. Diagnostic models did not benefit from this additional information regarding heterogeneity.