Distinguishing bipolar and major depressive disorders by brain structural morphometry: a pilot study

Machine learning applied to the prediction of relapse, hospitalization, and suicide in bipolar disorder using neuroimaging and clinical data: A systematic review

BACKGROUND

Bipolar disorder (BD) is associated with increased morbidity/mortality. Adverse outcome prediction might help with the management of patients with BD.

METHODS

We systematically reviewed the performance of machine learning (ML) studies in predicting adverse outcomes (relapse or recurrence, hospital admission, and suicide-related events) in patients with BD. Demographic, clinical, and neuroimaging-related poor outcome predictors were also reviewed. Three databases (PubMed, Scopus, and Web of Science) were explored from inception to July 2023.

RESULTS

Eighteen studies, accounting for >30,000 patients, were included. Support vector machine, decision trees, random forest, and logistic regression were the most frequently used ML algorithms. ML models' area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, and specificity ranged from 0.71 to 0.98, 72.7-92.8 %, and 59.0-95.2 % for relapse/recurrence prediction (4 studies (3 on relapses and 1 on recurrences). The corresponding values were 0.78-0.88, 21.4-100 %, and 77.0-99.7 % for hospital admissions (3 studies, 21,266 patients), and 0.71-0.99, 44.4-97.9 %, and 38.9-95.0 % for suicide-related events (10 studies, 5558 patients). Also, one study addressed a combination of the interest outcomes. Adverse outcome predictors included early onset BD, BD type I, comorbid psychiatric or substance use disorder, circadian rhythm disruption, hospitalization characteristics, and neuroimaging parameters, including increased dynamic amplitude of low-frequency fluctuation, decreased frontolimbic functional connectivity and aberrant dynamic functional connectivity in corticostriatal circuitry.

CONCLUSIONS

ML models can predict adverse outcomes of BD with relatively acceptable performance measures. Future studies with larger samples and nested cross-validation validation should be conducted to reach more reliable results.

Multi-site benchmark classification of major depressive disorder using machine learning on cortical and subcortical measures

Machine learning (ML) techniques have gained popularity in the neuroimaging field due to their potential for classifying neuropsychiatric disorders. However, the diagnostic predictive power of the existing algorithms has been limited by small sample sizes, lack of representativeness, data leakage, and/or overfitting. Here, we overcome these limitations with the largest multi-site sample size to date (N = 5365) to provide a generalizable ML classification benchmark of major depressive disorder (MDD) using shallow linear and non-linear models. Leveraging brain measures from standardized ENIGMA analysis pipelines in FreeSurfer, we were able to classify MDD versus healthy controls (HC) with a balanced accuracy of around 62%. But after harmonizing the data, e.g., using ComBat, the balanced accuracy dropped to approximately 52%. Accuracy results close to random chance levels were also observed in stratified groups according to age of onset, antidepressant use, number of episodes and sex. Future studies incorporating higher dimensional brain imaging/phenotype features, and/or using more advanced machine and deep learning methods may yield more encouraging prospects.

Identifying misdiagnosed bipolar disorder using support vector machine: feature selection based on fMRI of follow-up confirmed affective disorders

Nearly a quarter of bipolar disorder (BD) patients were misdiagnosed as major depressive disorder (MDD) patients, which cannot be corrected until mania/hypomania develops. It is important to recognize these obstacles so that the appropriate treatment can be initiated. Thus, we sought to distinguish patients with BD from MDD, especially to identify misdiagnosed BD before mania/hypomania, and further explore potential trait features that allow accurate differential diagnosis independent of state matters. Functional magnetic resonance imaging scans were performed at baseline on 92 MDD patients and 48 BD patients. The MDD patients were then followed up for more than two years. After follow-up, 23 patients transformed into BD (tBD), and 69 patients whose diagnoses remained unchanged were eligible for unipolar depression (UD). A support vector machine classifier was trained on the amygdala-based functional connectivity (FC) of 48 BD and 50 UD patients using a novel region-based feature selection. Then, the classifier was tested on the dataset, encompassing tBD and the remaining UD. It performed well for known BD and UD and can also distinguish tBD from UD with an accuracy of 81%, sensitivity of 82.6%, specificity of 79%, and AUC of 74.6%, respectively. Feature selection results revealed that ten regions within the cortico-limbic neural circuit contributed most to classification. Furthermore, in the FC comparisons among diseases, BD and tBD shared almost overlapped FC patterns in the cortico-limbic neural circuit, and both of them presented pronounced differences in most regions within the circuit compared with UD. The FC values of the most discriminating brain regions had no prominent correlations with the severity of depression, anxiety, and mania/hypomania (FDR correction). It suggests that BD possesses some trait features in the cortico-limbic neural circuit, rendering it dichotomized by the classifier based on known-diagnosis data.

Explainable machine-learning algorithms to differentiate bipolar disorder from major depressive disorder using self-reported symptoms, vital signs, and blood-based markers

BACKGROUND AND OBJECTIVE

Caused by shared genetic risk factors and similar neuropsychological symptoms, bipolar disorder (BD) and major depressive disorder (MDD) are at high risk of misdiagnosis, which is associated with ineffective treatment and worsening of outcomes. We aimed to develop a machine learning (ML)-based diagnostic system, based on electronic medical records (EMR) data, to mimic the clinical reasoning of human physicians to differentiate MDD and BD (especially BD depressive episodes) patients about to be admitted to a hospital and, hence, reduce the misdiagnosis of BD as MDD on admission. In addition, we examined to what extent our ML model could be made interpretable by quantifying and visualizing the features that drive the predictions.

METHODS

By identifying 16,311 patients admitted to a hospital located in western China between 2009 and 2018 with a recorded main diagnosis of MDD or BD, we established three sub-cohorts with different combinations of features for both the MDD-BD cohort and the MDD-BD depressive episodes cohort, respectively. Four different ML algorithms (logistic regression, extreme gradient boosting (XGBoost), random forest, and support vector machine) and four train-test splits were used to train and validate diagnostic models, and explainable methods (SHAP and Break Down) were utilized to analyze the contribution of each of the features at both population-level and individual-level, including feature importance, feature interaction, and feature effect on prediction decision for a specific subject.

RESULTS

The XGBoost algorithm provided the best test performance (AUC: 0.838 (0.810-0.867), PPV: 0.810 and NPV: 0.834) for separating patients with BD from those with MDD. Core predictors included symptoms (mood-up, exciting, bad sleep, loss of interest, talking, mood-down, provoke), along with age, job, myocardial enzyme markers (creatine kinase, hydroxybutyrate dehydrogenase), diabetes-associated marker (glucose), bone function marker (alkaline phosphatase), non-enzymatic antioxidant (uric acid), markers of immune/inflammation (white blood cell count, lymphocyte count, basophil percentage, monocyte count), cardiovascular function marker (low density lipoprotein), renal marker (total protein), liver biochemistry marker (indirect bilirubin), and vital signs like pulse. For separating patients with BD depressive episodes from those with MDD, the test AUC was 0.777 (0.732-0.822), with PPV 0.576 and NPV 0.899. Additional validation in models built with self-reported symptoms removed from the feature set, showed test AUC of 0.701 (0.666-0.736) for differentiating BD and MDD, and AUC of 0.564 (0.515-0.614) for detecting patients in BD depressive episodes from MDD patients. Validation in the datasets without removing the patients with comorbidity showed an AUC of 0.826 (0.806-0.846).

CONCLUSION

The diagnostic system accurately identified patients with BD in various clinical scenarios, and differences in patterns of peripheral markers between BD and MDD could enrich our understanding of potential underlying pathophysiological mechanisms of them.

Association of the plasma complement system with brain volume deficits in bipolar and major depressive disorders

BACKGROUND

Inflammation plays a crucial role in the pathogenesis of major depressive disorder (MDD) and bipolar disorder (BD). This study aimed to examine whether the dysregulation of complement components contributes to brain structural defects in patients with mood disorders.

METHODS

A total of 52 BD patients, 35 MDD patients, and 53 controls were recruited. The human complement immunology assay was used to measure the levels of complement factors. Whole brain-based analysis was performed to investigate differences in gray matter volume (GMV) and cortical thickness (CT) among the BD, MDD, and control groups, and relationships were explored between neuroanatomical differences and levels of complement components.

RESULTS

GMV in the medial orbital frontal cortex (mOFC) and middle cingulum was lower in both patient groups than in controls, while the CT of the left precentral gyrus and left superior frontal gyrus were affected differently in the two disorders. Concentrations of C1q, C4, factor B, factor H, and properdin were higher in both patient groups than in controls, while concentrations of C3, C4 and factor H were significantly higher in BD than in MDD. Concentrations of C1q, factor H, and properdin showed a significant negative correlation with GMV in the mOFC at the voxel-wise level.

CONCLUSIONS

BD and MDD are associated with shared and different alterations in levels of complement factors and structural impairment in the brain. Structural defects in mOFC may be associated with elevated levels of certain complement factors, providing insight into the shared neuro-inflammatory pathogenesis of mood disorders.

Increased cortical surface area but not altered cortical thickness or gyrification in bipolar disorder following stabilisation from a first episode of mania

BACKGROUND

Despite reports of altered brain morphology in established bipolar disorder (BD), there is limited understanding of when these morphological abnormalities emerge. Assessment of patients during the early course of illness can help to address this gap, but few studies have examined surface-based brain morphology in patients at this illness stage.

METHODS

We completed a secondary analysis of baseline data from a randomised control trial of BD individuals stabilised after their first episode of mania (FEM). The magnetic resonance imaging scans of n = 35 FEM patients and n = 29 age-matched healthy controls were analysed. Group differences in cortical thickness, surface area and gyrification were assessed at each vertex of the cortical surface using general linear models. Significant results were identified at p < 0.05 using cluster-wise correction.

RESULTS

The FEM group did not differ from healthy controls with regards to cortical thickness or gyrification. However, there were two clusters of increased surface area in the left hemisphere of FEM patients, with peak coordinates falling within the lateral occipital cortex and pars triangularis.

CONCLUSIONS

Cortical thickness and gyrification appear to be intact in the aftermath of a first manic episode, whilst cortical surface area in the inferior/middle prefrontal and occipitoparietal cortex is increased compared to age-matched controls. It is possible that increased surface area in the FEM group is the outcome of abnormalities in a premorbidly occurring process. In contrast, the findings raise the hypothesis that cortical thickness reductions seen in past studies of individuals with more established BD may be more attributable to post-onset factors.

Identifying the neural marker of chronic sciatica using multimodal neuroimaging and machine learning analyses

INTRODUCTION

Sciatica is a pain disorder often caused by the herniated disk compressing the lumbosacral nerve roots. Neuroimaging studies have identified functional abnormalities in patients with chronic sciatica (CS). However, few studies have investigated the neural marker of CS using brain structure and the classification value of multidimensional neuroimaging features in CS patients is unclear.

METHODS

Here, structural and resting-state functional magnetic resonance imaging (fMRI) was acquired for 34 CS patients and 36 matched healthy controls (HCs). We analyzed cortical surface area, cortical thickness, amplitude of low-frequency fluctuation (ALFF), regional homogeneity (REHO), between-regions functional connectivity (FC), and assessed the correlation between neuroimaging measures and clinical scores. Finally, the multimodal neuroimaging features were used to differentiate the CS patients and HC individuals by support vector machine (SVM) algorithm.

RESULTS

Compared to HC, CS patients had a larger cortical surface area in the right banks of the superior temporal sulcus and rostral anterior cingulate; higher ALFF value in the left inferior frontal gyrus; enhanced FCs between somatomotor and ventral attention network. Three FCs values were associated with clinical pain scores. Furthermore, the three multimodal neuroimaging features with significant differences between groups and the SVM algorithm could classify CS patients and HC with an accuracy of 90.00%.

DISCUSSION

Together, our findings revealed extensive reorganization of local functional properties, surface area, and network metrics in CS patients. The success of patient identification highlights the potential of using artificial intelligence and multimodal neuroimaging markers in chronic pain research.

Personalized Diagnosis and Treatment for Neuroimaging in Depressive Disorders

Depressive disorders are highly heterogeneous in nature. Previous studies have not been useful for the clinical diagnosis and prediction of outcomes of major depressive disorder (MDD) at the individual level, although they provide many meaningful insights. To make inferences beyond group-level analyses, machine learning (ML) techniques can be used for the diagnosis of subtypes of MDD and the prediction of treatment responses. We searched PubMed for relevant studies published until December 2021 that included depressive disorders and applied ML algorithms in neuroimaging fields for depressive disorders. We divided these studies into two sections, namely diagnosis and treatment outcomes, for the application of prediction using ML. Structural and functional magnetic resonance imaging studies using ML algorithms were included. Thirty studies were summarized for the prediction of an MDD diagnosis. In addition, 19 studies on the prediction of treatment outcomes for MDD were reviewed. We summarized and discussed the results of previous studies. For future research results to be useful in clinical practice, ML enabling individual inferences is important. At the same time, there are important challenges to be addressed in the future.

Associations of PM2.5 and road traffic noise with mental health: Evidence from UK Biobank

BACKGROUND

The associations of atmospheric particulate matter with diameters of 2.5 μm or less (PM2.5) and road traffic noise with mental disorders in men and women are not well studied.

OBJECTIVES

We aim to examine the cross-sectional associations of PM2.5 and road traffic noise with mental disorders in men and women.

METHODS

The baseline data of the UK Biobank study (2006-2010) were used. Mental disorders including symptoms of nerves, anxiety, tension or depression (NATD), major depression, and bipolar disorder were assessed by validated questions. Verified models were used to estimate PM2.5 and road traffic noise.

RESULTS

A total of 334,986 participants with measurements of NATD and 90,706 participants with measurements of major depression and bipolar disorder were included in the analysis. After adjusting for covariates, the odds for the risk of NATD symptoms increased by 2.31 (95% CI: 2.15-2.50) times per 10 μg/m³ increase in PM2.5. The odds for the risk of major depression and bipolar disorder increased by 2.26 and 4.99 times per 10 μg/m³ increase in PM2.5. On the other hand, higher road traffic noise exposure was significantly associated with a higher risk of NATD symptoms (Decile 6-8 (54.9-57.8 dB), OR: 1.03, 95% CI: 1.01-1.06; Decile 9-10 (≥57.8 dB), OR: 1.04, 95% CI: 1.01-1.07) and bipolar disorder (Decile 2-5 (52.1-54.9 dB), OR: 1.26, 95% CI: 1.00-1.59; Decile 6-8 (54.9-57.8 dB), OR: 1.30, 95% CI: 1.02-1.65; Decile 9-10 (≥57.8 dB), OR: 1.54, 95% CI: 1.21-1.97). Interestingly, a negative association was observed between moderate road traffic noise and major depression (Decile 2-5 (52.1-54.9 dB), OR: 0.95, 95% CI: 0.90-1.00). Interactions between PM2.5 exposure with age, gender, and sleeplessness for NATD symptoms were observed (P < 0.05), while interactions between road traffic noise exposure with age and gender were observed (P < 0.05).

CONCLUSIONS

We found a positive association between PM2.5 and mental disorders. Meanwhile, we found a positive association of road traffic noise with NATD symptoms and bipolar disorder and a negative association of moderate road traffic noise with major depression. Also, the effect modifications of these associations by age, gender, or sleeplessness may exist.

Divergent and convergent imaging markers between bipolar and unipolar depression based on Machine Learning

Distinguishing bipolar depression (BD) from unipolar depression (UD) based on symptoms only is challenging. Brain functional connectivity (FC), especially dynamic FC, has emerged as a promising approach to identify possible imaging markers for differentiating BD from UD. However, most of such studies utilized conventional FC and group-level statistical comparisons, which may not be sensitive enough to quantify subtle changes in the FC dynamics between BD and UD. In this paper, we present a more effective individualized differentiation model based on machine learning and the whole-brain high-order functional connectivity (HOFC) network. The HOFC, capturing temporal synchronization among the dynamic FC time series, a more complex chronnectome metric compared to the conventional FC, was used to classify 52 BD, 73 UD, and 76 healthy controls (HC). We achieved a satisfactory accuracy (70.40%) in BD vs. UD differentiation. The resultant contributing features revealed the involvement of the coordinated flexible interactions among sensory (e.g., olfaction, vision, and audition), motor, and cognitive systems. Despite sharing common chronnectome of cognitive and affective impairments, BD and UD also demonstrated unique dynamic FC synchronization patterns. UD is more associated with abnormal visual-somatomotor inter-network connections, while BD is more related to impaired ventral attention-frontoparietal inter-network connections. Moreover, we found that the illness duration modulated the BD vs. UD separation, with the differentiation performance hampered by the secondary disease effects. Our findings suggest that BD and UD may have divergent and convergent neural substrates, which further expand our knowledge of the two different mental disorders.

Alteration of cortical functional networks in mood disorders with resting-state electroencephalography

Studies comparing bipolar disorder (BD) and major depressive disorder (MDD) are scarce, and the neuropathology of these disorders is poorly understood. This study investigated source-level cortical functional networks using resting-state electroencephalography (EEG) in patients with BD and MDD. EEG was recorded in 35 patients with BD, 39 patients with MDD, and 42 healthy controls (HCs). Graph theory-based source-level weighted functional networks were assessed via strength, clustering coefficient (CC), and path length (PL) in six frequency bands. At the global level, patients with BD and MDD showed higher strength and CC, and lower PL in the high beta band, compared to HCs. At the nodal level, compared to HCs, patients with BD showed higher high beta band nodal CCs in the right precuneus, left isthmus cingulate, bilateral paracentral, and left superior frontal; however, patients with MDD showed higher nodal CC only in the right precuneus compared to HCs. Although both MDD and BD patients had similar global level network changes, they had different nodal level network changes compared to HCs. Our findings might suggest more altered cortical functional network in patients with BD than in those with MDD.

Altered cerebellar gray matter and cerebellar-cortex resting-state functional connectivity in patients with bipolar disorder Ⅰ

BACKGROUND

Bipolar disorder (BP) is a common psychiatric disorder characterized by extreme fluctuations in mood. Recent studies have indicated the involvement of cerebellum in the pathogenesis of BP. However, no study has focused on the precise role of cerebellum exclusively in patients with bipolar I disorder (BP-I).

METHODS

Forty-five patients with BP-I and 40 healthy controls were recruited. All subjects underwent clinical evaluation and Magnetic Resonance diffusion Tension Imaging scans. For structural images, we used a spatially unbiased infratentorial template toolbox to isolate the cerebellum and then preformed voxel-based morphometry (VBM) analyses to assess the difference in cerebellar gray matter volume (GMV) between the two groups. For the functional images, we chose the clusters that survived from VBM analysis as seeds and performed functional connectivity (FC) analysis. Between-group differences were assessed using the independent Students t test or the nonparametric Mann-Whitney U Test. For multiple comparisons, the results were further corrected with Gaussian random field (GRF) approach (voxel-level P < 0.001, cluster-level P < 0.05).

RESULTS

Compared with healthy controls, BP-I patients showed significantly decreased GMV in left lobule V and left lobule VI (P < 0.05, GRF corrected). The FC of cerebellum with bilateral superior temporal gyrus, bilateral insula, bilateral rolandic operculum, right putamen, and left precentral gyrus was disrupted in BP-I patients (P < 0.05, GRF corrected).

CONCLUSIONS

BP-I patients showed decreased cerebellar GMV and disrupted cerebellar-cortex resting-state FC. This suggests that cerebellar abnormalities may play an important role in the pathogenesis of BP-I.

Shared and specific patterns of dynamic functional connectivity variability of striato-cortical circuitry in unmedicated bipolar and major depressive disorders

BACKGROUND

Accumulating studies have found structural and functional abnormalities of the striatum in bipolar disorder (BD) and major depressive disorder (MDD). However, changes in intrinsic brain functional connectivity dynamics of striato-cortical circuitry have not been investigated in BD and MDD. This study aimed to investigate the shared and specific patterns of dynamic functional connectivity (dFC) variability of striato-cortical circuitry in BD and MDD.

METHODS

Brain resting-state functional magnetic resonance imaging data were acquired from 128 patients with unmedicated BD II (current episode depressed), 140 patients with unmedicated MDD, and 132 healthy controls (HCs). Six pairs of striatum seed regions were selected: the ventral striatum inferior (VSi) and the ventral striatum superior (VSs), the dorsal-caudal putamen (DCP), the dorsal-rostral putamen (DRP), and the dorsal caudate and the ventral-rostral putamen (VRP). The sliding-window analysis was used to evaluate dFC for each seed.

RESULTS

Both BD II and MDD exhibited increased dFC variability between the left DRP and the left supplementary motor area, and between the right VRP and the right inferior parietal lobule. The BD II had specific increased dFC variability between the right DCP and the left precentral gyrus compared with MDD and HCs. The MDD had increased dFC variability between the left VSi and the left medial prefrontal cortex compared with BD II and HCs.

CONCLUSIONS

The patients with BD and MDD shared common dFC alteration in the dorsal striatal-sensorimotor and ventral striatal-cognitive circuitries. The patients with MDD had specific dFC alteration in the ventral striatal-affective circuitry.

Cortical thickness abnormalities in patients with bipolar disorder: A systematic review and meta-analysis

BACKGROUND

An increasing number of neuroimaging studies report alterations of cortical thickness (CT) related to the neuropathology of bipolar disorder (BD). We provide here a whole-brain vertex-wise meta-analysis, which may help improve the spatial precision of these identifications.

METHODS

A comprehensive meta-analysis was performed to investigate the differences in CT between patients with BD and healthy controls (HCs) by using a newly developed mask for CT analysis in seed-based d mapping (SDM) meta-analytic software. We used meta-regression to explore the effects of demographics and clinical characteristics on CT. This meta-review was conducted in accordance with PRISMA guideline.

RESULTS

We identified 21 studies meeting criteria for the systematic review, of which 11 were eligible for meta-analysis. The meta-analysis comprising 649 BD patients and 818 HCs showed significant cortical thinning in the left insula extending to left Rolandic operculum and Heschl gyrus, the orbital part of left inferior frontal gyrus (IFG), the medial part of left superior frontal gyrus (SFG) as well as bilateral anterior cingulate cortex (ACC) in BD. In meta-regression analyses, mean patient age was negatively correlated with reduced CT in the left insula.

LIMITATIONS

All enrolled studies were cross-sectional; we could not explore the potential effects of medication and mood states due to the limited data.

CONCLUSIONS

Our results suggest that BD patients have significantly thinner frontoinsular cortex than HCs, and the results may be helpful in revealing specific neuroimaging biomarkers of BD patients.

Machine learning approaches for prediction of bipolar disorder based on biological, clinical and neuropsychological markers: a systematic review and meta-analysis

Applying machine learning (ML) to objective markers may overcome prognosis uncertainty due to the subjective nature of the diagnosis of bipolar disorder (BD). This PRISMA-compliant meta-analysis provides new systematic evidence of the BD classification accuracy reached by different markers and ML algorithms. We focused on neuroimaging, electrophysiological techniques, peripheral biomarkers, genetic data, neuropsychological or clinical measures, and multimodal approaches. PubMed, Embase and Scopus were searched through 3^rd December 2020. Meta-analyses were performed using random-effect models. Overall, 81 studies were included in this systematic review and 65 in the meta-analysis (11,336 participants, 3,903 BD). The overall pooled classification accuracy was 0.77 (95%CI[0.75;0.80]). Despite subgroup analyses for diagnostic comparison group, psychiatric disorders, marker, ML algorithm, and validation procedure were not significant, linear discriminant analysis significantly outperformed support vector machine for peripheral biomarkers (p=0.03). Sample size was inversely related to accuracy. Evidence of publication bias was detected. Ultimately, although ML reached a high accuracy in differentiating BD from other psychiatric disorders, best practices in methodology are needed for the advancement of future studies.

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 degree centrality in first-episode medication-free adolescent depression at rest: A functional magnetic resonance imaging study and support vector machine analysis

BACKGROUND

Depression in adolescents is more heterogeneous and less often diagnosed than depression in adults. At present, reliable approaches to differentiating between adolescents who are and are not affected by depression are lacking. This study was designed to assess voxel-level whole-brain functional connectivity changes associated with adolescent depression in an effort to define an imaging-based biomarker associated with this condition.

MATERIALS AND METHODS

In total, 71 adolescents affected by major depressive disorder (MDD) and 71 age-, sex-, and education level-matched healthy controls were subjected to resting-state functional magnetic resonance imaging (rs-fMRI) based analyses of brain voxel-wise degree centrality (DC), with a support vector machine (SVM) being used for pattern classification analyses.

RESULTS

DC patterns derived from 16-min rs-fMRI analyses were able to effectively differentiate between adolescent MDD patients and healthy controls with 95.1% accuracy (136/143), and with respective sensitivity and specificity values of 92.1% (70/76) and 98.5% (66/67) based upon DC abnormalities detected in the right cerebellum. Specifically, increased DC was evident in the bilateral insula and left lingual area of MDD patients, together with reductions in the DC values in the right cerebellum and bilateral superior parietal lobe. DC values were not significantly correlated with disease severity or duration in these patients following correction for multiple comparisons.

CONCLUSION

These results suggest that whole-brain network centrality abnormalities may be present in many brain regions in adolescent depression patients. Accordingly, these DC maps may hold value as candidate neuroimaging biomarkers capable of differentiating between adolescents who are and are not affected by MDD, although further validation of these results will be critical.

Practical Advice for Primary Care Clinicians on the Safe and Effective Use of Vortioxetine for Patients with Major Depressive Disorder (MDD)

Primary care clinicians have a vital role to play in the diagnosis and management of patients with major depressive disorder (MDD). This includes screening for MDD as well as identifying other possible psychiatric disorders including bipolar disorder and/or other comorbidities. Once MDD is confirmed, partnering with patients in the shared decision-making process while considering different treatment options and best management of MDD over the course of their illness is recommended. Vortioxetine has been approved for the treatment of adults with MDD since 2013, and subsequent US label updates indicate that vortioxetine may be particularly beneficial for specific populations of patients with MDD, including those with treatment-emergent sexual dysfunction and patients experiencing certain cognitive symptoms. Given these recent label updates, this prescribing guide for vortioxetine aims to provide clear and practical guidance for primary care clinicians on the safe and effective use of vortioxetine for the treatment of MDD, including how to identify appropriate patients for treatment.

Distinguish bipolar and major depressive disorder in adolescents based on multimodal neuroimaging: Results from the Adolescent Brain Cognitive Development study®

Background

Major depressive disorder and bipolar disorder in adolescents are prevalent and are associated with cognitive impairment, executive dysfunction, and increased mortality. Early intervention in the initial stages of major depressive disorder and bipolar disorder can significantly improve personal health.

Methods

We collected 309 samples from the Adolescent Brain Cognitive Development study, including 116 adolescents with bipolar disorder, 64 adolescents with major depressive disorder, and 129 healthy adolescents, and employed a support vector machine to develop classification models for identification. We developed a multimodal model, which combined functional connectivity of resting-state functional magnetic resonance imaging and four anatomical measures of structural magnetic resonance imaging (cortical thickness, area, volume, and sulcal depth). We measured the performances of both multimodal and single modality classifiers.

Results

The multimodal classifiers showed outstanding performance compared with all five single modalities, and they are 100% for major depressive disorder versus healthy controls, 100% for bipolar disorder versus healthy control, 98.5% (95% CI: 95.4–100%) for major depressive disorder versus bipolar disorder, 100% for major depressive disorder versus depressed bipolar disorder and the leave-one-site-out analysis results are 77.4%, 63.3%, 79.4%, and 81.7%, separately.

Conclusions

The study shows that multimodal classifiers show high classification performances. Moreover, cuneus may be a potential biomarker to differentiate major depressive disorder, bipolar disorder, and healthy adolescents. Overall, this study can form multimodal diagnostic prediction workflows for clinically feasible to make more precise diagnose at the early stage and potentially reduce loss of personal pain and public society.

Cortical Thickness and Surface Area Abnormalities in Bipolar I and II Disorders

OBJECTIVE

Although bipolar II disorder (BD II) is not simply a mitigated form of bipolar I disorder (BD I), their neurobiological differences have not been elucidated. The present study aimed to explore cortical thickness (CT) and surface area (SA) in patients with BD I and BD II and healthy controls (HCs) to investigate the shared and unique neurobiological mechanisms of BD subtypes.

METHODS

We enrolled 30 and 44 patients with BD I and BD II, respectively, and 100 HCs. We evaluated CT and SA using FreeSurfer and estimated differences in CT and SA among the three groups (BD I vs. BD II vs. HC). We adjusted for age, sex, educational level, and intracranial volume as confounding factors.

RESULTS

We found widespread cortical thinning in the bilateral frontal, temporal, and occipital regions; cingulate gyrus; and insula in patients with BD. Alterations in SA, including increased SA of the pars triangularis and decreased SA of the insula, were noted in patients with BD. Overall, we found BD II patients demonstrated decreased SA in the right long insula compared to BD I patients.

CONCLUSION

Our results suggest that decreased SA in the right long insula is crucial for differentiating BD subtypes.

Inflammation is correlated with abnormal functional connectivity in unmedicated bipolar depression: an independent component analysis study of resting-state fMRI

BACKGROUND

Inflammation might play a role in bipolar disorder (BD), but it remains unclear the relationship between inflammation and brain structural and functional abnormalities in patients with BD. In this study, we focused on the alterations of functional connectivity (FC), peripheral pro-inflammatory cytokines and their correlations to investigate the role of inflammation in FC in BD depression.

METHODS

In this study, 42 unmedicated patients with BD II depression and 62 healthy controls (HCs) were enrolled. Resting-state-functional magnetic resonance imaging was performed in all participants and independent component analysis was used. Serum levels of Interleukin-6 (IL-6) and Interleukin-8 (IL-8) were measured in all participants. Correlation between FC values and IL-6 and IL-8 levels in BD was calculated.

RESULTS

Compared with the HCs, BD II patients showed decreased FC in the left orbitofrontal cortex (OFC) implicating the limbic network and the right precentral gyrus implicating the somatomotor network. BD II showed increased IL-6 (p = 0.039), IL-8 (p = 0.002) levels. Moreover, abnormal FC in the right precentral gyrus were inversely correlated with the IL-8 (r = -0.458, p = 0.004) levels in BD II. No significant correlation was found between FC in the left OFC and cytokines levels.

CONCLUSIONS

Our findings that serum IL-8 levels are associated with impaired FC in the right precentral gyrus in BD II patients suggest that inflammation might play a crucial role in brain functional abnormalities in BD.

Examining the relationship between gray matter volume and a continuous measure of bipolarity in unmedicated unipolar and bipolar depression

BACKGROUND

It has been argued that unipolar major depressive disorder (MDD) and bipolar disorder (BD) exist on a continuous spectrum, given their overlapping symptomatology and genetic diatheses. The Bipolarity Index (BI) is a scale that considers bipolarity as a continuous construct and was developed to assess confidence in bipolar diagnosis. Here we investigated whether BI scores correlate with gray matter volume (GMV) in a sample of unmedicated unipolar and bipolar depressed individuals.

METHODS

158 subjects (139 with MDD, 19 with BD) in a major depressive episode at time of scan were assigned BI scores. T1-weighted Magnetic Resonance Imaging scans were obtained and processed with Voxel-Based Morphometry using SPM12 (CAT12 toolbox) to assess GMV. Regression was performed at the voxel level to identify clusters of voxels whose GMV was associated with BI score, (p<0.001, family-wise error-corrected cluster-level p<0.05), with age, sex and total intracranial volume as covariates.

RESULTS

GMV was inversely correlated with BI score in four clusters located in left lateral occipital cortex, bilateral angular gyri and right frontal pole. Clusters were no longer significant after controlling for diagnosis. GMV was not correlated with BI score within the MDD cohort alone.

LIMITATIONS

Incomplete clinical data required use of a modified BI scale.

CONCLUSION

BI scores were inversely correlated with GMV in unmedicated subjects with MDD and BD, but these correlations appeared driven by categorical diagnosis. Future work will examine other imaging modalities and focus on elements of the BI scale most likely to be related to brain structure and function.

Shared and specific dynamics of brain segregation and integration in bipolar disorder and major depressive disorder: A resting-state functional magnetic resonance imaging study

BACKGROUND

When bipolar disorder (BD) presents as the depressive state, it is often misdiagnosed as major depressive disorder (MDD). However, few studies have focused on dynamic differences in local brain activity and connectivity between BD and MDD. Therefore, the present study explored shared and specific patterns of abnormal dynamic brain segregation and integration in BD and MDD patients.

METHODS

BD Patients (n = 106), MDD patients (n = 114), and 130 healthy controls (HCs) underwent resting state functional magnetic resonance imaging (fMRI). We first used a sliding window analysis to evaluate the dynamic amplitude of low-frequency fluctuations (dALFF) and, based on the altered dALFF, further analyzed the dynamic functional connectivity (dFC) using a seed-based approach.

RESULTS

Both the BD and MDD groups showed decreased temporal variability of the dALFF (less dynamic segregation) in the bilateral posterior cingulate cortex (PCC)/precuneus compared with the HCs. The MDD group showed increased temporal variability of the dALFF (more dynamic segregation) in the left putamen compared with the controls, but there was no significant difference between the BD and HCs. The dFC analysis also showed that both the BD and MDD groups had reduced dFC (less dynamic integration) between the bilateral PCC/ precuneus and the left inferior parietal lobule compared with the HCs.

LIMITATIONS

This study was cross-sectional and did not examine data from remitted BD and MDD patients.

CONCLUSION

Our findings indicated disrupted dynamic balance between segregation and integration within the default mode network in both BD and MDD. Moreover, we found MDD-specific abnormal brain dynamics in the putamen.

Structural and functional alterations in untreated patients with major depressive disorder and bipolar disorder experiencing first depressive episode: A magnetic resonance imaging study combined with follow-up

BACKGROUND

Magnetic resonance imaging (MRI) could assist in identifying objective biomarkers and follow-up study could effectively improve subjective diagnostic accuracy. By combining MRI with follow-up, this study aims to determine the shared and distinct alterations between major depressive disorder (MDD) and bipolar disorder (BD).

METHODS

Untreated patients with MDD experiencing the first episode were subjected to MRI and subsequent follow-up. Fifteen patients with mania or hypomania were regrouped into BD group. Twenty patients were still grouped as MDD after an average of 37.95 months follow-up. Thirty healthy controls (HCs) were recruited to match the patients. Gray matter volume (GMV) and amygdala-seed functional connectivity (FC) in the whole brain were detected and compared among the three groups.

RESULTS

GMV analysis revealed that the MDD and BD groups presented reduced GMV predominantly in the parietal, occipital, and frontal regions in the bilateral cerebrum compared with the HCs. The BD group had reduced GMV predominantly in the parietal, temporal, insular regions and the Rolandic operculum in the right-side cerebrum compared with MDD and HC groups. FC analysis revealed that the MDD and BD patients displayed increased FC values mainly in the bilateral parietal, and left occipital regions. Only the BD group displayed increased FC values in the temporal, occipital, parietal and limbic regions in the right-side cerebrum relative to HCs.

LIMITATIONS

The main limitation is the relatively small sample size.

CONCLUSIONS

Alterations in the cortical regions and cortico-limbic neural system may provide the scientific basis for differential diagnosis in affective disorders.

Specific Gray Matter Volume Changes of the Brain in Unipolar and Bipolar Depression

To identify the common and specific structural basis of bipolar depression (BD) and unipolar depression (UD) is crucial for clinical diagnosis. In this study, a total of 85 participants, including 22 BD patients, 36 UD patients, and 27 healthy controls, were enrolled. A voxel-based morphology method was used to identify the common and specific changes of the gray matter volume (GMV) to determine the structural basis. Significant differences in GMV were found among the three groups. Compared with healthy controls, UD patients showed decreased GMV in the orbital part of the left inferior frontal gyrus, whereas BD patients showed decreased GMV in the orbital part of the left middle frontal gyrus. Compared with BD, UD patients have increased GMV in the left supramarginal gyrus and middle temporal gyrus. Our results revealed different structural changes in UD and BD patients suggesting BD and UD have different neurophysiological underpinnings. Our study contributes toward the biological determination of morphometric changes, which could help to discriminate between UD and BD.

Cost-effectiveness of neuroimaging technologies in management of psychiatric and insomnia disorders: A meta-analysis and prospective cost analysis

BACKGROUND

The optimal diagnostic strategy for patients with psychiatric and insomnia disorders has not been established yet.

PURPOSE

The purpose of this study was to perform cost-effectiveness analysis of six neuroimaging technologies in diagnosis of patients with psychiatric and insomnia disorders.

METHODS

An economic evaluation study was conducted in three parts, including a systematic review for determining diagnostic accuracy, a descriptive cross-sectional study with Activity-Based Costing (ABC) technique for tracing resource consumption, and a cost-effectiveness analysis using a short-term decision-analytic model.

RESULTS

In the first phase, 93 diagnostic accuracy studies were included in the systematic review. The accumulated results (meta-analysis) showed that the highest diagnostic accuracy for psychiatric and insomnia disorders was attributed to PET (sensitivity of 90% and specificity of 80%) and MRI (sensitivity of 76% and specificity of 78%) respectively. In the second phase of the study, we calculated the cost of each technology. The results showed that MRI has the lowest cost. Based on the results in the model of cost-effectiveness sMRI ($ 50.08 per accurate diagnosis) and MRI ($ 58.54 per accurate diagnosis) were more cost-effective neuroimaging technologies.

CONCLUSION

In psychiatric disorders, no single strategy was characterized by both low cost and high accuracy. However, MRI and PET scan had lower cost and higher accuracy for psychiatric disorders, respectively. MRI was the least costly with the highest diagnostic accuracy in insomnia disorders. Based on our model, sMRI in psychiatric disorders and MRI in insomnia disorders were the most cost-effective technologies.

Neurobiology of bipolar disorders: a review of genetic components, signaling pathways, biochemical changes, and neuroimaging findings

Bipolar disorder (BD) is a chronic mental illness characterized by changes in mood that alternate between mania and hypomania or between depression and mixed states, often associated with functional impairment. Although effective pharmacological and non-pharmacological treatments are available, several patients with BD remain symptomatic. The advance in the understanding of the neurobiology underlying BD could help in the identification of new therapeutic targets as well as biomarkers for early detection, prognosis, and response to treatment in BD. In this review, we discuss genetic, epigenetic, molecular, physiological and neuroimaging findings associated with the neurobiology of BD. Despite the advances in the pathophysiological knowledge of BD, the diagnosis and management of the disease are still essentially clinical. Given the complexity of the brain and the close relationship between environmental exposure and brain function, initiatives that incorporate genetic, epigenetic, molecular, physiological, clinical, environmental data, and brain imaging are necessary to produce information that can be translated into prevention and better outcomes for patients with BD.

Volumetric brain differences in clinical depression in association with anxiety: a systematic review with meta-analysis

BACKGROUND

Structural differences associated with depression have not been confirmed in brain regions apart from the hippocampus. Comorbid anxiety has been inconsistently assessed, and may explain discrepancies in previous findings. We investigated the link between depression, comorbid anxiety and brain structure.

METHODS

We followed Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines (PROSPERO CRD42018089286). We searched the Cochrane Library, MEDLINE, PsycInfo, PubMed and Scopus, from database inception to Sept. 13, 2018, for MRI case-control studies that reported brain volumes in healthy adults and adults with clinical depression. We summarized mean volumetric differences using meta-analyses, and we assessed demographics, depression factors and segmentation procedure as moderators using meta-regressions.

RESULTS

We included 112 studies in the meta-analyses, assessing 4911 healthy participants and 5934 participants with depression (mean age 49.8 yr, 68.2% female). Volume effects were greater in late-onset depression and in multiple episodes of depression. Adults with depression and no comorbidity showed significantly lower volumes in the putamen, pallidum and thalamus, as well as significantly lower grey matter volume and intracranial volume; the largest effects were in the hippocampus (6.8%, p < 0.001). Adults with depression and comorbid anxiety showed significantly higher volumes in the amygdala (3.6%, p < 0.001). Comorbid anxiety lowered depression effects by 3% on average. Sex moderated reductions in intracranial volume.

LIMITATIONS

High heterogeneity in hippocampus effects could not be accounted for by any moderator. Data on symptom severity and medication were sparse, but other factors likely made significant contributions.

CONCLUSION

Depression-related differences in brain structure were modulated by comorbid anxiety, chronicity of symptoms and onset of illness. Early diagnosis of anxiety symptomatology will prove crucial to ensuring effective, tailored treatments for improving long-term mental health and mitigating cognitive problems, given the effects in the hippocampus.

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.

Cortical Volumetric Correlates of Childhood Trauma, Anxiety, and Impulsivity in Bipolar Disorder

OBJECTIVE

More recently, attention has turned to the linkage between childhood trauma and emotional dysregulation, but the evidence in bipolar disorder (BD) is limited. To determine neurobiological relationships between childhood trauma, current anxiety, and impulsivity, we investigated cortical volumetric correlates of these clinical factors in BD.

METHODS

We studied 36 patients with DSM-5 BD and 29 healthy controls. Childhood trauma, coexisting anxiety, and impulsivity were evaluated with the Korean version-Childhood Trauma Questionnaire (CTQ), the Korean version-Beck Anxiety Inventory (BAI), and the Korean version-Barratt Impulsiveness Scale (BIS). Voxel-based morphometry (VBM) was used to assess gray matter volume (GMV) alterations on the brain magnetic resonance imaging (MRI). Partial correlation analyses were conducted to examine associations between the GMV and each scale in the BD group.

RESULTS

Childhood trauma, anxiety, and impulsivity were interrelated in BD. BD patients revealed significant inverse correlations between the GMV in the right precentral gyrus and CTQ scores (r=-0.609, p<0.0003); between the GMV in the left middle frontal gyrus and BAI scores (r=-0.363, p=0.044). Moreover, patients showed similar tendency of negative correlations between the GMV in the right precentral gyrus and BIS scores; between the GMV in the left middle frontal gyrus and CTQ scores.

CONCLUSION

The present study provides evidence for a neural basis between childhood trauma and affect regulations in BD. The GMV alterations in multiple frontal lobe areas may represent neurobiological markers for anticipating the course of BD.

Will machine learning applied to neuroimaging in bipolar disorder help the clinician? A critical review and methodological suggestions

OBJECTIVES

The existence of anatomofunctional brain abnormalities in bipolar disorder (BD) is now well established by magnetic resonance imaging (MRI) studies. To create diagnostic and prognostic tools, as well as identifying biologically valid subtypes of BD, research has recently turned towards the use of machine learning (ML) techniques. We assessed both supervised ML and unsupervised ML studies in BD to evaluate their robustness, reproducibility and the potential need for improvement.

METHOD

We systematically searched for studies using ML algorithms based on MRI data of patients with BD until February 2019.

RESULT

We identified 47 studies, 45 using supervised ML techniques and 2 including unsupervised ML analyses. Among supervised studies, 43 focused on diagnostic classification. The reported accuracies for classification of BD ranged between (a) 57% and 100%, for BD vs healthy controls; (b) 49.5% and 93.1% for BD vs patients with major depressive disorder; and (c) 50% and 96.2% for BD vs patients with schizophrenia. Reported accuracies for discriminating subjects genetically at risk for BD (either from control or from patients with BD) ranged between 64.3% and 88.93%.

CONCLUSIONS

Although there are strong methodological limitations in previous studies and an important need for replication in large multicentric samples, the conclusions of our review bring hope of future computer-aided diagnosis of BD and pave the way for other applications, such as treatment response prediction. To reinforce the reliability of future results we provide methodological suggestions for good practice in conducting and reporting MRI-based ML studies in BD.

Predicting differential diagnosis between bipolar and unipolar depression with multiple kernel learning on multimodal structural neuroimaging

One of the greatest challenges in providing early effective treatment in mood disorders is the early differential diagnosis between major depression (MDD) and bipolar disorder (BD). A remarkable need exists to identify reliable biomarkers for these disorders. We integrate structural neuroimaging techniques (i.e. Tract-based Spatial Statistics, TBSS, and Voxel-based morphometry) in a multiple kernel learning procedure in order to define a predictive function of BD against MDD diagnosis in a sample of 148 patients. We achieved a balanced accuracy of 73.65% with a sensitivity for BD of 74.32% and specificity for MDD of 72.97%. Mass-univariates analyses showed reduced grey matter volume in right hippocampus, amygdala, parahippocampal, fusiform gyrus, insula, rolandic and frontal operculum and cerebellum, in BD compared to MDD. Volumes in these regions and in anterior cingulate cortex were also reduced in BD compared to healthy controls (n = 74). TBSS analyses revealed widespread significant effects of diagnosis on fractional anisotropy, axial, radial, and mean diffusivity in several white matter tracts, suggesting disruption of white matter microstructure in depressed patients compared to healthy controls, with worse pattern for MDD. To best of our knowledge, this is the first study combining grey matter and diffusion tensor imaging in predicting BD and MDD diagnosis. Our results prompt brain quantitative biomarkers and multiple kernel learning as promising tool for personalized treatment in mood disorders.

A proof of concept machine learning analysis using multimodal neuroimaging and neurocognitive measures as predictive biomarker in bipolar disorder

BACKGROUND

Concomitant use of complementary, multimodal imaging measures and neurocognitive measures is reported to have higher accuracy as a biomarker in Alzheimer's dementia. However, such an approach has not been examined to differentiate healthy individuals from Bipolar disorder. In this study, we examined the utility of support vector machine (SVM) technique to differentiate bipolar disorder patients and healthy using structural, functional and diffusion tensor images of brain and neurocognitive measures.

METHODS

30 patients with Bipolar disorder-I and 30 age, sex matched individuals participated in the study. Structural MRI, resting state functional MRI and diffusion tensor images were obtained using a 1.5 T scanner. All participants were administered neuropsychological tests to measure executive functions. SVM, a supervised machine learning technique was applied to differentiate patients and healthy individuals with k-fold cross validation over 10 trials.

RESULTS

The composite marker consisting of both neuroimaging and neuropsychological measures, had an accuracy of 87.60 %, sensitivity of 82.3 % and specificity of 92.7 %. The performance of composite marker was better compared to that of individual markers on classificatory.

CONCLUSIONS

We were able to achieve a high accuracy for machine learning technique in distinguishing BD from HV using a combination of multimodal neuroimaging and neurocognitive measures. Findings of this proof of concept study, if replicated in larger samples, could have potential clinical applications.

Differential biomarker signatures in unipolar and bipolar depression: A machine learning approach

OBJECTIVE

This study used machine learning techniques combined with peripheral biomarker measurements to build signatures to help differentiating (1) patients with bipolar depression from patients with unipolar depression, and (2) patients with bipolar depression or unipolar depression from healthy controls.

METHODS

We assessed serum levels of interleukin-2, interleukin-4, interleukin-6, interleukin-10, tumor necrosis factor-α, interferon-γ, interleukin-17A, brain-derived neurotrophic factor, lipid peroxidation and oxidative protein damage in 54 outpatients with bipolar depression, 54 outpatients with unipolar depression and 54 healthy controls, matched by sex and age. Depressive symptoms were assessed using the Hamilton Depression Rating Scale. Variable selection was performed with recursive feature elimination with a linear support vector machine kernel, and the leave-one-out cross-validation method was used to test and validate our model.

RESULTS

Bipolar vs unipolar depression classification achieved an area under the receiver operating characteristics (ROC) curve (AUC) of 0.69, with 0.62 sensitivity and 0.66 specificity using three selected biomarkers (interleukin-4, thiobarbituric acid reactive substances and interleukin-10). For the comparison of bipolar depression vs healthy controls, the model retained five variables (interleukin-6, interleukin-4, thiobarbituric acid reactive substances, carbonyl and interleukin-17A), with an AUC of 0.70, 0.62 sensitivity and 0.7 specificity. Finally, unipolar depression vs healthy controls comparison retained seven variables (interleukin-6, Carbonyl, brain-derived neurotrophic factor, interleukin-10, interleukin-17A, interleukin-4 and tumor necrosis factor-α), with an AUC of 0.74, a sensitivity of 0.68 and 0.70 specificity.

CONCLUSION

Our findings show the potential of machine learning models to aid in clinical practice, leading to more objective assessment. Future studies will examine the possibility of combining peripheral blood biomarker data with other biological data to develop more accurate signatures.

Using structural MRI to identify bipolar disorders - 13 site machine learning study in 3020 individuals from the ENIGMA Bipolar Disorders Working Group

Bipolar disorders (BDs) are among the leading causes of morbidity and disability. Objective biological markers, such as those based on brain imaging, could aid in clinical management of BD. Machine learning (ML) brings neuroimaging analyses to individual subject level and may potentially allow for their diagnostic use. However, fair and optimal application of ML requires large, multi-site datasets. We applied ML (support vector machines) to MRI data (regional cortical thickness, surface area, subcortical volumes) from 853 BD and 2167 control participants from 13 cohorts in the ENIGMA consortium. We attempted to differentiate BD from control participants, investigated different data handling strategies and studied the neuroimaging/clinical features most important for classification. Individual site accuracies ranged from 45.23% to 81.07%. Aggregate subject-level analyses yielded the highest accuracy (65.23%, 95% CI = 63.47-67.00, ROC-AUC = 71.49%, 95% CI = 69.39-73.59), followed by leave-one-site-out cross-validation (accuracy = 58.67%, 95% CI = 56.70-60.63). Meta-analysis of individual site accuracies did not provide above chance results. There was substantial agreement between the regions that contributed to identification of BD participants in the best performing site and in the aggregate dataset (Cohen's Kappa = 0.83, 95% CI = 0.829-0.831). Treatment with anticonvulsants and age were associated with greater odds of correct classification. Although short of the 80% clinically relevant accuracy threshold, the results are promising and provide a fair and realistic estimate of classification performance, which can be achieved in a large, ecologically valid, multi-site sample of BD participants based on regional neurostructural measures. Furthermore, the significant classification in different samples was based on plausible and similar neuroanatomical features. Future multi-site studies should move towards sharing of raw/voxelwise neuroimaging data.

Common and distinct neural activities in frontoparietal network in first-episode bipolar disorder and major depressive disorder: Preliminary findings from a follow-up resting state fMRI study

BACKGROUND

It is difficult to distinguish bipolar disorder (BD) from major depressive disorder (MDD), especially with the initial depressive episode. In this study, we compared neural activities of BD and MDD patients during the first-episode (FE) to investigate common and distinct neural activities and further explore predictive indicators in the two diseases.

METHODS

FE-MDD patients were performed resting state functional magnetic resonance imaging and followed up after scanning. After follow-up, FE-MDD patients were regrouped into FE-BD and FE-MDD patients. The study included 24 FE-BD patients, 28 FE-MDD patients, and 30 age- and sex-matched healthy controls (HC) to investigate neural activities with regional homogeneity (ReHo) analysis among the 3 groups.

RESULTS

Compared to HC, FE-BD patients displayed significantly higher ReHo values in the superior frontal gyrus, the medial superior frontal gyrus within right-side cerebral hemisphere than FE-MDD patients and HC. Compared to HC, FE-BD and FE-MDD patients displayed significant decreased ReHo values in the paracentral lobule, the precuneus and the median cingulate and paracingulate gyrus within bilateral cerebral hemisphere, and the postcentral gyrus and the precentral gyrus within the right-side. FE-BD displayed significant lower ReHo values than FE-MDD patients in these regions.

LIMITATIONS

The potential effects of medicine, age, course of disease and handedness on results could not be ignored.

CONCLUSIONS

Abnormal neural activities of frontoparietal network may provide common and distinct markers to affective disorders and scientific basis for further prediction researches of affective disorders.

Automated cortical thickness and skewness feature selection in bipolar disorder using a semi-supervised learning method

BACKGROUND

Bipolar disorder (BD) broadly affects brain structure, in particular areas involved in emotion processing and cognition. In the last years, the psychiatric field's interest in machine learning approaches has been steadily growing, thanks to the potentiality of automatically discriminating patients from healthy controls.

METHODS

In this work, we employed cortical thickness of 58 regions of interest obtained from magnetic resonance imaging scans of 41 BD patients and 34 healthy controls, to automatically identify the regions which are mostly involved with the disease. We used a semi-supervised method, addressing the criticisms on supervised methods, related to the fact that the diagnosis is not unaffected by uncertainty.

RESULTS

Our results confirm findings in previous studies, with a classification accuracy of about 75% when mean thickness and skewness of up to five regions are considered. We obtained that the parietal lobe and some areas in the temporal sulcus were the regions which were the most involved with BD.

LIMITATIONS

The major limitation of our work is the limited size or our dataset, but in line with other recent machine learning works in the field. Moreover, we considered chronic patients, whose brain characteristics may thus be affected.

CONCLUSIONS

The automatic selection of the brain regions most involved in BD may be of great importance when dealing with the pathogenesis of the disorder. Our method selected regions which are known to be involved with BD, indicating that damage to the identified areas can be considered as a marker of disease.

Explicit emotional memory biases in mood disorders: A systematic review

Research suggests that major depressive disorder (MDD) and bipolar disorder (BD) are both associated with unique emotional memory (EM) biases. To better elucidate the EM phenotypes of these disorders, we systematically reviewed the literature on non-autobiographical explicit EM biases in individuals with MDD and BD compared to healthy controls. The following databases were searched: Cochrane, Embase, HAPI, LILACs, Medline, PsycInfo and Web of Science. Grey literature and hand searches were also performed. Fourteen studies met full eligibility criteria. Eleven studies included data from an MDD sample (10 during acute depression, 1 during euthymia) and 3 studies included data from a BD sample (2 during acute mood episodes, 1 during euthymia). Only 3 of the studies in acute depression revealed a negative explicit EM bias. One study in MDD during euthymia revealed an EM deficit for negative stimuli. One of the two studies in BD (type I; BD-I) during an acute mood episode revealed a positive explicit EM bias, while the other showed no bias. One study in BD during euthymia showed an EM deficit for negative stimuli. Overall, this review concludes that current empirical evidence does not readily support the existence of an explicit EM bias in MDD during acute depression. The identification and implications of potential moderating factors on explicit EM performance in MDD and BD during both illness stages are discussed.

Machine learning in mental health: a scoping review of methods and applications

BACKGROUND

This paper aims to synthesise the literature on machine learning (ML) and big data applications for mental health, highlighting current research and applications in practice.

METHODS

We employed a scoping review methodology to rapidly map the field of ML in mental health. Eight health and information technology research databases were searched for papers covering this domain. Articles were assessed by two reviewers, and data were extracted on the article's mental health application, ML technique, data type, and study results. Articles were then synthesised via narrative review.

RESULTS

Three hundred papers focusing on the application of ML to mental health were identified. Four main application domains emerged in the literature, including: (i) detection and diagnosis; (ii) prognosis, treatment and support; (iii) public health, and; (iv) research and clinical administration. The most common mental health conditions addressed included depression, schizophrenia, and Alzheimer's disease. ML techniques used included support vector machines, decision trees, neural networks, latent Dirichlet allocation, and clustering.

CONCLUSIONS

Overall, the application of ML to mental health has demonstrated a range of benefits across the areas of diagnosis, treatment and support, research, and clinical administration. With the majority of studies identified focusing on the detection and diagnosis of mental health conditions, it is evident that there is significant room for the application of ML to other areas of psychology and mental health. The challenges of using ML techniques are discussed, as well as opportunities to improve and advance the field.

Transdiagnostic neuroimaging in psychiatry: A review

Transdiagnostic approach has a long history in neuroimaging, predating its recent ascendance as a paradigm for new psychiatric nosology. Various psychiatric disorders have been compared for commonalities and differences in neuroanatomical features and activation patterns, with different aims and rationales. This review covers both structural and functional neuroimaging publications with direct comparison of different psychiatric disorders, including schizophrenia, bipolar disorder, major depressive disorder, autism spectrum disorder, obsessive-compulsive disorder, attention-deficit/hyperactivity disorder, conduct disorder, anorexia nervosa, and bulimia nervosa. Major findings are systematically presented along with specific rationales for each comparison.

Image Based Brain Segmentation: From Multi-Atlas Fusion to Deep Learning

Background:

This review aims to identify the development of the algorithms for brain tissue and structure segmentation in MRI images.

Discussion:

Starting from the results of the Grand Challenges on brain tissue and structure segmentation held in Medical Image Computing and Computer-Assisted Intervention (MICCAI), this review analyses the development of the algorithms and discusses the tendency from multi-atlas label fusion to deep learning. The intrinsic characteristics of the winners’ algorithms on the Grand Challenges from the year 2012 to 2018 are analyzed and the results are compared carefully.

Conclusion:

Although deep learning has got higher rankings in the challenge, it has not yet met the expectations in terms of accuracy. More effective and specialized work should be done in the future.

Larger right inferior frontal gyrus volume and surface area in participants at genetic risk for bipolar disorders

BACKGROUND

Larger grey matter volume of the inferior frontal gyrus (IFG) is among the most replicated biomarkers of genetic risk for bipolar disorders (BD). However, the IFG is a heterogeneous prefrontal region, and volumetric findings can be attributable to changes in cortical thickness (CT), surface area (SA) or gyrification. Here, we investigated the morphometry of IFG in participants at genetic risk for BD.

METHODS

We quantified the IFG cortical grey matter volume in 29 affected, 32 unaffected relatives of BD probands, and 42 controls. We then examined SA, CT, and cortical folding in subregions of the IFG.

RESULTS

We found volumetric group differences in the right IFG, with the largest volumes in unaffected high-risk and smallest in control participants (F2,192 = 3.07, p = 0.01). The volume alterations were localized to the pars triangularis of the IFG (F2,97 = 4.05, p = 0.02), with no differences in pars opercularis or pars orbitalis. Pars triangularis volume was highly correlated with its SA [Pearson r(101) = 0.88, p &lt; 0.001], which significantly differed between the groups (F2,97 = 4.45, p = 0.01). As with volume, the mean SA of the pars triangularis was greater in unaffected (corrected p = 0.02) and affected relatives (corrected p = 0.05) compared with controls. We did not find group differences in pars triangularis CT or gyrification.

CONCLUSIONS

These findings strengthen prior knowledge about the volumetric findings in this region and provide a new insight into the localization and topology of IFG alterations. The unique nature of rIFG morphology in BD, with larger volume and SA early in the course of illness, could have practical implications for detection of participants at risk for BD.

Differentiating between bipolar and unipolar depression in functional and structural MRI studies

Distinguishing depression in bipolar disorder (BD) from unipolar depression (UD) solely based on clinical clues is difficult, which has led to the exploration of promising neural markers in neuroimaging measures for discriminating between BD depression and UD. In this article, we review structural and functional magnetic resonance imaging (MRI) studies that directly compare UD and BD depression based on neuroimaging modalities including functional MRI studies on regional brain activation or functional connectivity, structural MRI on gray or white matter morphology, and pattern classification analyses using a machine learning approach. Numerous studies have reported distinct functional and structural alterations in emotion- or reward-processing neural circuits between BD depression and UD. Different activation patterns in neural networks including the amygdala, anterior cingulate cortex (ACC), prefrontal cortex (PFC), and striatum during emotion-, reward-, or cognition-related tasks have been reported between BD and UD. A stronger functional connectivity pattern in BD was pronounced in default mode and in frontoparietal networks and brain regions including the PFC, ACC, parietal and temporal regions, and thalamus compared to UD. Gray matter volume differences in the ACC, hippocampus, amygdala, and dorsolateral prefrontal cortex (DLPFC) have been reported between BD and UD, along with a thinner DLPFC in BD compared to UD. BD showed reduced integrity in the anterior part of the corpus callosum and posterior cingulum compared to UD. Several studies performed pattern classification analysis using structural and functional MRI data to distinguish between UD and BD depression using a supervised machine learning approach, which yielded a moderate level of accuracy in classification.

Cortical thickness and surface area as an endophenotype in bipolar disorder type I patients and their first-degree relatives

OBJECTIVES

So far, few studies have investigated cortical thickness (CT) and surface area (SA) measures in bipolar disorder type I (BDI) in comparison to a high genetic risk group such as first-degree relatives (FR). This study aimed to examine CT and SA differences between BDI, FR and healthy controls (HC).

METHODS

3D T1 magnetic resonance images were acquired from 27 euthymic BDI patients, 24 unaffected FR and 29 HC. CT and SA measures were obtained with FreeSurfer version 5.3.0. Generalized estimating equations were used to compare CT and SA between groups. Group comparisons were repeated with restricting the FR group to 17 siblings (FR-SB) only.

RESULTS

\Mean age in years was 36.3 ± 9.5 for BDI, 32.1 ± 10.9 for FR, 34.7 ± 9.8 for FR-SB and 33.1 ± 9.0 for HC group respectively. BDI patients revealed larger SA of left pars triangularis (LPT) compared to HC (p = .001). In addition, increased SA in superior temporal cortex (STC) in FR-SB group compared to HC was identified (p = .0001).

CONCLUSIONS

Our result of increased SA in LPT of BDI could be a disease marker and increased SA in STC of FR-SB could be a marker related with resilience to illness.

Cortical thickness in major depressive disorder: A systematic review and meta-analysis

Neuroimaging studies assessing neurobiological differences between patients with major depressive disorder (MDD) and healthy controls (HC) are often hindered by small sample sizes and heterogeneity of the patient sample. We performed a comprehensive literature search for studies assessing cortical thickness between patient and control groups, including studies investigating treatment effects on cortical thickness. We identified 34 studies meeting criteria for the systematic review and used Seed-based d Mapping to meta-analyze 24 of those that met additional criteria. Analysis of the full sample of subjects (MDD = 1073; HC = 936) revealed significant thinning in the MDD group in the bilateral orbitofrontal gyrus (BA 11), left pars opercularis (BA 45) and left calcarine fissure/lingual gyrus (BA 17), as well as an area of significant thickening in the left supramarginal gyrus (BA 40). These results support other imaging modalities that report disruptions in various frontal and temporal areas in MDD and identify additional areas in all major cerebral lobes likely to be significant when parsing for biomarkers of treatment or relapse.

Cortico-limbic connectivity as a possible biomarker for bipolar disorder: where are we now?

INTRODUCTION

The fronto-limbic network has been suggested as a key circuitry in the pathophysiology and maintenance of bipolar disorder. In the past decade, a disrupted connectivity within prefrontal-limbic structures was identified as a promising candidate biomarker for the disorder. Areas Covered: In this review, the authors examine current literature in terms of the structural, functional and effective connectivity in bipolar disorder, integrating recent findings of imaging genetics and machine learning. This paper profiles the current knowledge and identifies future perspectives to provide reliable and usable neuroimaging biomarkers for bipolar psychopathology in clinical practice. Expert Opinion: The replication and the translation of acquired knowledge into useful and usable tools represents one of the current greatest challenges in biomarker research applied to psychiatry.

Illness, at-risk and resilience neural markers of early-stage bipolar disorder

BACKGROUND

Current knowledge on objective and specific neural markers for bipolar risk and resilience-related processes is lacking, partly due to not subdividing high-risk individuals manifesting different levels of subclinical symptoms who possibly possess different levels of resilience.

METHODS

We delineated grey matter markers for bipolar illness, genetic high risk (endophenotype) and resilience, through comparing across 42 young non-comorbid bipolar patients, 42 healthy controls, and 72 diagnosis-free, medication-naive high-genetic-risk individuals subdivided into a combined-high-risk group who additionally manifested bipolar risk-relevant subsyndromes (N = 38), and an asymptomatic high-risk group (N = 34). Complementary analyses assessed the additional predictive and classification values of grey matter markers beyond those of clinical scores, through using logistic regression and support vector machine analyses.

RESULTS

Illness-related effects manifested as reduced grey matter volumes of bilateral temporal limbic-striatal and cerebellar regions, which significantly differentiated bipolar patients from healthy controls and improved clinical classification specificity by 20%. Reduced bilateral cerebellar grey matter volume emerged as a potential endophenotype and (along with parieto-occipital grey matter changes) separated combined-high-risk individuals from healthy and high-risk individuals, and increased clinical classification specificity by approximately 10% and 27%, respectively, while the relatively normalized cerebellar grey matter volumes in the high-risk sample may confer resilience.

LIMITATIONS

The cross-validation procedure was not performed on an independent sample using independently-derived features. The BD group had different age and sex distributions than some other groups which may not be fully addressable statistically.

CONCLUSIONS

Our framework can be applied in other measurement domains to derive complete profiles for bipolar patients and at-risk individuals, towards forming strategies for promoting resilience and preclinical intervention.