Multivariate Classification of Major Depressive Disorder Using the Effective Connectivity and Functional Connectivity

Abstract

Major depressive disorder (MDD) is a mental disorder characterized by at least 2 weeks of low mood, which is present across most situations. Diagnosis of MDD using rest-state functional magnetic resonance imaging (fMRI) data faces many challenges due to the high dimensionality, small samples, noisy and individual variability. To our best knowledge, no studies aim at classification with effective connectivity and functional connectivity measures between MDD patients and healthy controls. In this study, we performed a data-driving classification analysis using the whole brain connectivity measures which included the functional connectivity from two brain templates and effective connectivity measures created by the default mode network (DMN), dorsal attention network (DAN), frontal-parietal network (FPN), and silence network (SN). Effective connectivity measures were extracted using spectral Dynamic Causal Modeling (spDCM) and transformed into a vectorial feature space. Linear Support Vector Machine (linear SVM), non-linear SVM, k-Nearest Neighbor (KNN), and Logistic Regression (LR) were used as the classifiers to identify the differences between MDD patients and healthy controls. Our results showed that the highest accuracy achieved 91.67% (p < 0.0001) when using 19 effective connections and 89.36% when using 6,650 functional connections. The functional connections with high discriminative power were mainly located within or across the whole brain resting-state networks while the discriminative effective connections located in several specific regions, such as posterior cingulate cortex (PCC), ventromedial prefrontal cortex (vmPFC), dorsal cingulate cortex (dACC), and inferior parietal lobes (IPL). To further compare the discriminative power of functional connections and effective connections, a classification analysis only using the functional connections from those four networks was conducted and the highest accuracy achieved 78.33% (p < 0.0001). Our study demonstrated that the effective connectivity measures might play a more important role than functional connectivity in exploring the alterations between patients and health controls and afford a better mechanistic interpretability. Moreover, our results showed a diagnostic potential of the effective connectivity for the diagnosis of MDD patients with high accuracies allowing for earlier prevention or intervention.

Abstract Representations of Emotions Perceived From the Face, Body, and Whole-Person Expressions in the Left Postcentral Gyrus

“…The GLM was constructed to model the data for each participant and the subsequent analysis was conducted on each of the first three runs, generating nine activation patterns in total. The sources of nuisance regressors along with their time derivatives were removed through the linear regression, including six head motion correction parameters, and averaged signals from the white matter and CSF ( Xu et al, 2017 ; Geng et al, 2018 ). The main analytical steps included in this study were shown in Figure 2 .…”

Neural Network Modeling and Correlation Analysis of Brain Plasticity Mechanisms in Stroke Patients

“…Authors studied discriminative brain areas that contribute to the classification of depression disorders, which may help understand the pathogenesis of depression disorders. In the paper [10] it is mentioned that major depressive disorder (MDD) is a mental disorder characterized by at least 2 weeks of low mood, which is present across most situations. Diagnosis of MDD using resting-state fMRI data faces many challenges due to the high dimensionality, small samples, noisy, and individual variability.…”

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

“…Specifically, resting‐state functional magnetic resonance imaging (rs‐fMRI) has been widely used for the diagnosis of MDD by investigating altered functional networks while a subject is at rest (Anand et al, 2005; Craddock, Holtzheimer, Hu, & Mayberg, 2009; Greicius et al, 2007). In the meantime, more recently, the investigation of dynamic changes between connections beyond simple correlations has been attracting increasing interest (Geng, Xu, Liu, & Shi, 2018; Rolls et al, 2018). The notion of effective connectivity (EC) describes the influence of one neural system on another (Friston, Ungerleider, Jezzard, & Turner, 1994), in contrast to functional connectivity (FC) that denotes intrinsic correlations.…”

“…Specifically, (Schlösser et al, 2008) found that adolescents suffering from MDD exhibited a significant difference in EC between the amygdala and subgenual anterior cingulate cortex (ACC) during an emotion‐relevant task. In addition, Geng et al (2018) directly utilized both FC and EC measures as features for the diagnosis of MDD and established that the discriminative power of EC features is higher than that of FC features. More recently, using a large sample size (336 patients with MDD and 350 control subjects), Rolls et al (2018) identified significantly altered EC measures in MDD, such as reduced connectivity from temporal lobe areas to the medial orbitofrontal cortex.…”

“…Several approaches such as dynamic causal modeling (DCM) (Park & Friston, 2013) and Granger causality (GC) (Granger, 1969) have been suggested for estimating EC. DCM is a commonly used approach; however, it requires the selection of seed regions of interest (ROIs) that are widely known as discriminant biomarkers in relevant literature rather than the whole brain connectivity due to computational complexity (Geng et al, 2018). GC, owing to its simplicity and ease of implementation, has been widely used to estimate the EC (Hamilton, Chen, Thomason, Schwartz, & Gotlib, 2011; Liao et al, 2011; Wu & Marinazzo, 2015).…”