Predictive brain networks for major depression in a semi-multimodal fusion hierarchical feature reduction framework

Diagnosis of Major Depressive Disorder Using Machine Learning Based on Multisequence MRI Neuroimaging Features

BACKGROUND

Previous studies have found qualitative structural and functional brain changes in major depressive disorder (MDD) patients. However, most studies ignored the complementarity of multisequence MRI neuroimaging features and cannot determine accurate biomarkers.

PURPOSE

To evaluate machine-learning models combined with multisequence MRI neuroimaging features to diagnose patients with MDD.

STUDY TYPE

Prospective.

SUBJECTS

A training cohort including 111 patients and 90 healthy controls (HCs) and a test cohort including 28 patients and 22 HCs.

FIELD STRENGTH/SEQUENCE

A 3.0 T/T1-weighted imaging, resting-state functional MRI with echo-planar sequence, and single-shot echo-planar diffusion tensor imaging.

ASSESSMENT

Recruitment and integration were used to reflect the dynamic changes of functional networks, while gray matter volume and fractional anisotropy were used to reflect the changes in the morphological and anatomical network. We then fused features with significant differences in functional, morphological, and anatomical networks to evaluate a random forest (RF) classifier to diagnose patients with MDD. Furthermore, a support vector machine (SVM) classifier was used to verify the stability of neuroimaging features. Linear regression analyses were conducted to investigate the relationships among multisequence neuroimaging features and the suicide risk of patients.

STATISTICAL TESTS

The comparison of functional network attributes between patients and controls by two-sample t-test. Network-based statistical analysis was used to identify structural and anatomical connectivity changes between MDD and HCs. The performance of the model was evaluated by receiver operating characteristic (ROC) curves.

RESULTS

The performance of the RF model integrating multisequence neuroimaging features in the diagnosis of depression was significantly improved, with an AUC of 93.6%. In addition, we found that multisequence neuroimaging features could accurately predict suicide risk in patients with MDD (r = 0.691).

DATA CONCLUSION

The RF model fusing functional, morphological, and anatomical network features performed well in diagnosing patients with MDD and provided important insights into the pathological mechanisms of MDD.

EVIDENCE LEVEL

1.

TECHNICAL EFFICACY

Stage 2.

Functional connectivity signatures of major depressive disorder: machine learning analysis of two multicenter neuroimaging studies

The promise of machine learning has fueled the hope for developing diagnostic tools for psychiatry. Initial studies showed high accuracy for the identification of major depressive disorder (MDD) with resting-state connectivity, but progress has been hampered by the absence of large datasets. Here we used regular machine learning and advanced deep learning algorithms to differentiate patients with MDD from healthy controls and identify neurophysiological signatures of depression in two of the largest resting-state datasets for MDD. We obtained resting-state functional magnetic resonance imaging data from the REST-meta-MDD (N = 2338) and PsyMRI (N = 1039) consortia. Classification of functional connectivity matrices was done using support vector machines (SVM) and graph convolutional neural networks (GCN), and performance was evaluated using 5-fold cross-validation. Features were visualized using GCN-Explainer, an ablation study and univariate t-testing. The results showed a mean classification accuracy of 61% for MDD versus controls. Mean accuracy for classifying (non-)medicated subgroups was 62%. Sex classification accuracy was substantially better across datasets (73-81%). Visualization of the results showed that classifications were driven by stronger thalamic connections in both datasets, while nearly all other connections were weaker with small univariate effect sizes. These results suggest that whole brain resting-state connectivity is a reliable though poor biomarker for MDD, presumably due to disease heterogeneity as further supported by the higher accuracy for sex classification using the same methods. Deep learning revealed thalamic hyperconnectivity as a prominent neurophysiological signature of depression in both multicenter studies, which may guide the development of biomarkers in future studies.

Creating sparser prediction models of treatment outcome in depression: a proof-of-concept study using simultaneous feature selection and hyperparameter tuning

BACKGROUND

Predicting treatment outcome in major depressive disorder (MDD) remains an essential challenge for precision psychiatry. Clinical prediction models (CPMs) based on supervised machine learning have been a promising approach for this endeavor. However, only few CPMs have focused on model sparsity even though sparser models might facilitate the translation into clinical practice and lower the expenses of their application.

METHODS

In this study, we developed a predictive modeling pipeline that combines hyperparameter tuning and recursive feature elimination in a nested cross-validation framework. We applied this pipeline to a real-world clinical data set on MDD treatment response and to a second simulated data set using three different classification algorithms. Performance was evaluated by permutation testing and comparison to a reference pipeline without nested feature selection.

RESULTS

Across all models, the proposed pipeline led to sparser CPMs compared to the reference pipeline. Except for one comparison, the proposed pipeline resulted in equally or more accurate predictions. For MDD treatment response, balanced accuracy scores ranged between 61 and 71% when models were applied to hold-out validation data.

CONCLUSIONS

The resulting models might be particularly interesting for clinical applications as they could reduce expenses for clinical institutions and stress for patients.

Dynamical regional activity in putamen distinguishes bipolar type I depression and unipolar depression

OBJECTIVES

Intrinsic human brain activity is time-varying and dynamic. However, there is still a lack of knowledge about the dynamic regional activity differences between unipolar depression (UD) and bipolar type I depression (BD-I), and whether their differential pattern can help to distinguish these two patient groups who are prone to misdiagnosis in clinical practice.

METHOD

In this study, we used the dynamical fractional amplitude of low-frequency fluctuations (dfALFF) to examine the resting-state dynamical regional activity in 40 BD-I, 42 UD, and 44 healthy controls (HCs). Analysis of covariance was applied to explore the shared and distinct dfALFF pattern among three groups, and machine-learning methods were conducted to classify BD-I from UD by using the detected distinct dfALFF pattern.

RESULTS

Compared with HCs, both BD-I and UD exhibited decreased dfALFF temporal variability in the left inferior temporal gyrus. The BD-I showed significantly decreased dfALFF temporal variability in the left putamen compared to UD. By using the dfALFF variability pattern of the left putamen as features, we achieved the 75.61% accuracy and 0.756 area under curve in classifying BD-I from UD.

LIMITATIONS

The small sample size of the current study may limit the generalizability of the findings.

CONCLUSIONS

The current study demonstrated that the dfALFF temporal variability pattern in the putamen may show a promise as future diagnostic aids for BD-I and UD.

Decreased Cortical Folding of the Fusiform Gyrus and Its Hypoconnectivity with Sensorimotor Areas in Major Depressive Disorder

BACKGROUND

Neuroimaging studies have revealed abnormal cortical folding pattern and disruptive functional connectivity in major depressive disorder (MDD). Combining structure and function in the same population may further our understanding of the neuropathological mechanisms of MDD.

METHOD

Sixty-two patients with MDD and 61 healthy controls (HCs) underwent structural and resting-state functional magnetic resonance imaging (MRI). Group differences in the cortical folding (measured by local gyrification index (LGI)) were analyzed in FreeSurfer. Taking the brain regions with significant group differences in LGI as seed regions, the resting-state functional connectivity analysis was further conducted to explore the corresponding functional connectivity alterations.

RESULTS

Comparing with HCs, patients with MDD showed significantly decreased LGI in the right fusiform gyrus (cohen's d = 0.70). In the seed-based functional connectivity analysis, we found that compared with HCs, patients with MDD showed decreased functional connections between the right fusiform gyrus with sensorimotor areas (precentral and postcentral gyrus) (cohen's d = 1.32) and right superior temporal gyrus (cohen's d = 0.94).

LIMITATIONS

Main limitations are the relatively small sample size and the cross-sectional study design.

CONCLUSION

Decreased LGI in the right fusiform gyrus, as well as decreased functional connectivity between the right fusiform gyrus and the sensorimotor area and right superior temporal gyrus, appears to play a role in the pathophysiology of MDD.

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.

Multimodal fusion of structural and functional brain imaging in depression using linked independent component analysis

Previous structural and functional neuroimaging studies have implicated distributed brain regions and networks in depression. However, there are no robust imaging biomarkers that are specific to depression, which may be due to clinical heterogeneity and neurobiological complexity. A dimensional approach and fusion of imaging modalities may yield a more coherent view of the neuronal correlates of depression. We used linked independent component analysis to fuse cortical macrostructure (thickness, area, gray matter density), white matter diffusion properties and resting-state functional magnetic resonance imaging default mode network amplitude in patients with a history of depression (n = 170) and controls (n = 71). We used univariate and machine learning approaches to assess the relationship between age, sex, case-control status, and symptom loads for depression and anxiety with the resulting brain components. Univariate analyses revealed strong associations between age and sex with mainly global but also regional specific brain components, with varying degrees of multimodal involvement. In contrast, there were no significant associations with case-control status, nor symptom loads for depression and anxiety with the brain components, nor any interaction effects with age and sex. Machine learning revealed low model performance for classifying patients from controls and predicting symptom loads for depression and anxiety, but high age prediction accuracy. Multimodal fusion of brain imaging data alone may not be sufficient for dissecting the clinical and neurobiological heterogeneity of depression. Precise clinical stratification and methods for brain phenotyping at the individual level based on large training samples may be needed to parse the neuroanatomy of depression.

Deep neural networks in psychiatry

Machine and deep learning methods, today's core of artificial intelligence, have been applied with increasing success and impact in many commercial and research settings. They are powerful tools for large scale data analysis, prediction and classification, especially in very data-rich environments ("big data"), and have started to find their way into medical applications. Here we will first give an overview of machine learning methods, with a focus on deep and recurrent neural networks, their relation to statistics, and the core principles behind them. We will then discuss and review directions along which (deep) neural networks can be, or already have been, applied in the context of psychiatry, and will try to delineate their future potential in this area. We will also comment on an emerging area that so far has been much less well explored: by embedding semantically interpretable computational models of brain dynamics or behavior into a statistical machine learning context, insights into dysfunction beyond mere prediction and classification may be gained. Especially this marriage of computational models with statistical inference may offer insights into neural and behavioral mechanisms that could open completely novel avenues for psychiatric treatment.

Abnormal Connectivity Within Anterior Cortical Midline Structures in Bipolar Disorder: Evidence From Integrated MRI and Functional MRI

Background: Aberrant functional and structural connectivity across multiple brain networks have been reported in bipolar disorder (BD). However, most previous studies consider the functional and structural alterations in isolation regardless of their possible integrative relationship. The present study aimed to identify the brain connectivity alterations in BD by capturing the latent nexus in multimodal neuroimaging data. Methods: Structural and resting-state images were acquired from 83 patients with BD and 94 healthy controls (HCs). Combined with univariate methods conducted to detect the dysconnectivity in BD, we also employed a semi-multimodal fusion framework fully utilizing the interrelationship between the two modalities to distinguish patients from HCs. Moreover, one-way analysis of variance was adopted to explore whether the detected dysconnectivity has differences across stages of patients with BD. Results: The semi-multimodal fusion framework distinguished patients from HCs with 81.47% accuracy, 85.42% specificity, and 74.75% sensitivity. The connection between the anterior cingulate cortex (ACC) and superior medial prefrontal cortex (sMPFC) contributed the most to BD diagnosis. Consistently, the univariate method also identified that this ACC-sMPFC functional connection significantly decreased in BD patients compared to HCs, and the significant order of the dysconnectivity is: depressive episode < HCs and remission episode < HCs. Conclusions: Our findings, by adopting univariate and multivariate analysis methods, shed light on the decoupling within the anterior midline brain in the pathophysiology of BD, and this decoupling may serve as a trait marker for this disease.

Effective Diagnosis of Alzheimer's Disease via Multimodal Fusion Analysis Framework

Alzheimer’s disease (AD) is a complex neurodegenerative disease involving a variety of pathogenic factors, and the etiology detection of this disease has been a major concern of researchers. Neuroimaging is a basic and important means to explore the problem. It is the main current scientific research direction for combining neuroimaging with other modal data to dig deep into the potential information of AD through the complementarities among multiple data points. Machine learning methods possess great potentiality and have reached some achievements in this research area. A few studies have proposed some solutions to the effects of multimodal data fusion, however, the overall analytical framework for data fusion and fusion result analysis has thus far been ignored. In this paper, we first put forward a novel multimodal data fusion method, and further present a new machine learning framework of data fusion, classification, feature selection, and disease-causing factor extraction. The real dataset of 37 AD patients and 35 normal controls (NC) with functional magnetic resonance imaging (fMRI) and genetic data was used to verify the effectiveness of the framework, which was more accurate in classification and optimal feature extraction than other methods. Furthermore, we revealed disease-causing brain regions and genes, such as the olfactory cortex, insula, posterior cingulate gyrus, lingual gyrus, CNTNAP2, LRP1B, FRMD4A, and DAB1. The results show that the machine learning framework could effectively perform multimodal data fusion analysis, providing new insights and perspectives for the diagnosis of Alzheimer’s disease.