Aberrant dynamic functional and effective connectivity changes of the primary visual cortex in patients with retinal detachment via machine learning

Previous neuroimaging studies have identified significant alterations in brain functional activity in retinal detachment (RD) patients, these investigations predominantly concentrated on local functional activity changes. The potential directional alterations in functional connectivity within the pr...

Full description

Saved in:
Bibliographic Details
Published in:Neuroreport Vol. 35; no. 17; p. 1071
Main Authors: Ji, Yu, Wang, Yuan-Yuan, Cheng, Qi, Fu, Wen-Wen, Shu, Ben-Liang, Wei, Bin, Huang, Qin-Yi, Wu, Xiao-Rong
Format: Journal Article
Language:English
Published: England 04.12.2024
Subjects:
Adult
Connectome - methods
Female
Humans
Machine Learning
Magnetic Resonance Imaging - methods
Male
Middle Aged
Nerve Net - diagnostic imaging
Nerve Net - physiopathology
Primary Visual Cortex - diagnostic imaging
Primary Visual Cortex - physiopathology
Retinal Detachment - diagnostic imaging
Retinal Detachment - physiopathology
Support Vector Machine
Young Adult
ISSN:1473-558X, 1473-558X
Online Access:Get more information
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Previous neuroimaging studies have identified significant alterations in brain functional activity in retinal detachment (RD) patients, these investigations predominantly concentrated on local functional activity changes. The potential directional alterations in functional connectivity within the primary visual cortex (V1) in RD patients remain to be elucidated. In this study, we employed seed-based functional connectivity analysis along with Granger causality analysis to examine the directional alterations in dynamic functional connectivity (dFC) within the V1 region of patients diagnosed with RD. Finally, a support vector machine algorithm was utilized to classify patients with RD and healthy controls (HCs). RD patients exhibited heightened dynamic functional connectivity (dFC) and dynamic effective connectivity (dEC) between the Visual Network (VN) and default mode network (DMN), as well as within the VN, compared to HCs. Conversely, dFC between VN and auditory network (AN) decreased, and dEC between VN and sensorimotor network (SMN) significantly reduced. In state 4, RD patients had higher frequency. Notably, variations in dFC originating from the left V1 region proved diagnostically effective, achieving an AUC of 0.786. This study reveals significant alterations in the connectivity between the VN and the default mode network in patients with RD. These changes may disrupt visual information processing and higher cognitive integration in RD patients. Additionally, alterations in the left V1 region and whole-brain dFC show promising potential in aiding the diagnosis of RD. These findings offer valuable insights into the neural mechanisms underlying visual and cognitive impairments associated with RD.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1473-558X
1473-558X
DOI:10.1097/WNR.0000000000002100