Investigation of the static and dynamic brain network mechanisms in patients with rhegmatogenous retinal detachment based on independent component analysis

Abstract

Background

Previous neuroimaging studies have identified substantial structural and functional abnormalities in the brains of patients with rhegmatogenous retinal detachment (RRD). Nonetheless, there remains a paucity of comprehensive research on the alterations in functional connectivity (FC) within large-scale static and dynamic brain networks in these patients.

Methods

This study utilized independent component analysis (ICA) to investigate alterations in large-scale brain network FC in patients with RRD. Additionally, it employed support vector machine (SVM) to classify RRD patients and healthy controls (HCs) and examined the relationship between abnormal brain regions and clinical ophthalmic parameters.

Results

The RRD patients demonstrated significantly increased FC in the default mode network (DMN) and visual network (VN) compared to the HCs, whereas the FC in the auditory network (AN) and the sensorimotor network (SMN) was significantly decreased. Analysis of dynamic functional network connectivity (dFNC) revealed that the fraction of time (FT) spent in state 4 was significantly greater in RRD patients compared to HCs. SVM analysis showed that the AUC for classification using AN and FNC features reached 0.828 and 0.819, respectively. Additionally, the FC value of the right medial superior frontal gyrus (R-SFGmed) in RRD patients was positively correlated with axial length (r = 0.401, p = 0.038).

Conclusion

This study revealed that patients with RRD exhibit both damage and adaptive remodeling in their brain functional networks. Alterations in the AN and FNC may serve as potential neuroimaging biomarkers for distinguishing RRD patients from HCs, providing crucial neuroimaging evidence for understanding the mechanisms underlying visual loss in RRD.