Multivariate pattern analysis strategies in detection of remitted major depressive disorder using resting state functional connectivity

Abstract

Understanding abnormal resting-state functional connectivity of distributed brain networks may aid in probing and targeting mechanisms involved in major depressive disorder (MDD). To date, few studies have used resting state functional magnetic resonance imaging (rs-fMRI) to attempt to discriminate individuals with MDD from individuals without MDD, and to our knowledge no investigations have examined a remitted (r) population. In this study, we examined the efficiency of support vector machine (SVM) classifier to successfully discriminate rMDD individuals from healthy controls (HCs) in a narrow early-adult age range. We empirically evaluated four feature selection methods including multivariate Least Absolute Shrinkage and Selection Operator (LASSO) and Elastic Net feature selection algorithms. Our results showed that SVM classification with Elastic Net feature selection achieved the highest classification accuracy of 76.1% (sensitivity of 81.5% and specificity of 68.9%) by leave-one-out cross-validation across subjects from a dataset consisting of 38 rMDD individuals and 29 healthy controls. The highest discriminating functional connections were between the left amygdala, left posterior cingulate cortex, bilateral dorso-lateral prefrontal cortex, and right ventral striatum. These appear to be key nodes in the etiopathophysiology of MDD, within and between default mode, salience and cognitive control networks. This technique demonstrates early promise for using rs-fMRI connectivity as a putative neurobiological marker capable of distinguishing between individuals with and without rMDD. These methods may be extended to periods of risk prior to illness onset, thereby allowing for earlier diagnosis, prevention, and intervention.

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

“…Studies involving a combination of MRI and pattern recognition techniques to explore biomarkers of depression have grown substantially in recent years. Such methods can accurately discriminate depressed subjects from healthy controls and predict treatment response . In our survey, there are more studies focused on classification (53 studies) than treatment response prediction (10 studies).…”

Affective traits and history of depression are related to ventral striatum connectivity

“…Second, despite mixed findings on SEN functioning in active MDD, we hypothesized increased connectivity between the VS and SEN in rMDD compared to HC individuals. In a smaller subset of the present sample, our group recently found that greater amygdala and VSs connectivity predicted rMDD status (Bhaumik et al, 2016) and we designed our second hypothesis to follow up on this finding. We also predicted hyperconnectivity from the VS to the DMN in the rMDD group relative to HCs, given the association of DMN hyperconnectivity with key clinical features of rMDD (Jacobs et al, 2014).…”

“…posterior cingulate cortex [PCC] and subgenual cingulate) with lateral, parietal, and frontal regions (Jacobs et al, 2014). In fact, a machine-learning algorithm predicted rMDD versus HC status based on increased rsFC between left PCC and DLPFC (Bhaumik et al, 2016). Young adults with rMDD also showed left amygdala hyperconnectivity with the right medial frontal gyrus, medial parietal lobe, rostral ACC, and left parahippocampal gyrus, which suggests that SEN connectivity may be increased in individuals with rMDD compared to HCs (Jacobs et al, 2014).…”

“…Young adults with rMDD also showed left amygdala hyperconnectivity with the right medial frontal gyrus, medial parietal lobe, rostral ACC, and left parahippocampal gyrus, which suggests that SEN connectivity may be increased in individuals with rMDD compared to HCs (Jacobs et al, 2014). In contrast, greater amygdala connectivity with the superior VS predicted rMDD versus HC classification (Bhaumik et al, 2016). Hyperconnectivity from the caudate to the amygdala and hippocampus has been reported in adults with rMDD compared to HCs (Admon et al, 2015).…”

Integrated cross-network connectivity of amygdala, insula, and subgenual cingulate associated with facial emotion perception in healthy controls and remitted major depressive disorder

“…Seeds from within the salience and emotion network (SEN) were consistent with those used by our group previously (Bhaumik et al, 2016; Jacobs et al, 2014) and were derived based on previous literature examining resting state connectivity of the amygdala (McCabe & Mishor, 2011; Pannekoek et al, 2013), sgACC (Kelly et al, 2009; Margulies et al, 2007) and anterior insula (Horn et al, 2010; Sridharan et al, 2008). The following coordinates were used: amygdala +/- 23, -5, -19mm, sgACC +/- 4, 21, -8, anterior insula +/- 36, 13, 5mm.…”