Static and dynamic connectomics differentiate between depressed patients with and without suicidal ideation. Issue 10 (1st July 2018)

• Record Type: Journal Article
• Authors: Liao, Wei; Li, Jiao; Duan, Xujun; Cui, Qian; Chen, Heng; Chen, Huafu
• Is Part Of: Human brain mapping. Volume 39:Issue 10(2018)
• Page Start: 4105
• Page End: 4118
• Publication Date: 2018-07-01
• Subjects: diagnostic model -- dynamic connectomics -- major depression -- predictive model -- suicidal ideation -- topological dissociation; Brain mapping -- Periodicals; 611.81
• DOI: 10.1002/hbm.24235
• Languages: English
• ISSNs: 1065-9471

Abstract

Abstract: Neural circuit dysfunction underlies the biological mechanisms of suicidal ideation (SI). However, little is known about how the brain's "dynome" differentiate between depressed patients with and without SI. This study included depressed patients ( n = 48) with SI, without SI (NSI), and healthy controls (HC, n = 30). All participants underwent resting‐state functional magnetic resonance imaging. We constructed dynamic and static connectomics on 200 nodes using a sliding window and full‐length time–series correlations, respectively. Specifically, the temporal variability of dynamic connectomic was quantified using the variance of topological properties across sliding window. The overall topological properties of both static and dynamic connectomics further differentiated between SI and NSI, and also predicted the severity of SI. The SI showed decreased overall topological properties of static connectomic relative to the HC. The SI exhibited increases in overall topological properties with regard to the dynamic connectomic when compared with the HC and the NSI. Importantly, combining the overall topological properties of dynamic and static connectomics yielded mean 75% accuracy (all p < .001) with mean 71% sensitivity and mean 75% specificity in differentiating between SI and NSI. Moreover, these features may predict the severity of SI (mean r = .55, all p < .05). The findings revealed that combining static and dynamic connectomics could differentiate between SI and NSI, offering new insight into the physiopathological mechanisms underlying SI. Furthermore, combining the brain's connectome and dynome may be considered a neuromarker for diagnostic and predictive models in the study of SI.