Effective brain network analysis with resting-state EEG data: a comparison between heroin abstinent and non-addicted subjects.

Effective brain network analysis with resting-state EEG data: a comparison between heroin abstinent and non-addicted subjects.

J Neural Eng. 2017 Apr 11;:

Authors: Hu B, Dong Q, Hao Y, Zhao Q, Shen J, Zheng F

Abstract
Abstract-Objective. Neuro-electrophysiological tools have been widely used in heroin addiction studies. Previous studies indicated that chronic heroin abuse would result in abnormal functional organization of the brain. While few heroin addiction studies apply the effective connectivity tool to analyze the brain functional system (BFS) alterations induced by heroin abuse. The present study aims to identify the abnormality of resting-state heroin abstinent BFS using source decomposition and effective connectivity tools.
APPROACH: The resting-state electroencephalograph (EEG) signals were acquired from 15 male heroin abstinent (HA) subjects and 14 male non-addicted (NA) controls. Multivariate autoregressive models combined independent component analysis (MVARICA) approach was applied for blind source decomposition. Generalized partial directed coherence (GPDC) was applied for effective brain connectivity analysis. Effective brain networks of HA and NA groups were constructed respectively. The two groups of effective cortical networks were compared by the bootstrap method. Abnormal causal interactions between decomposed source regions were estimated in the frequency domain (1-45 Hz).
MAIN RESULTS: This work suggested: (a) there were clear effective network alterations in heroin abstinent subject groups; (b) the parietal region was a dominant hub of the abnormally weaker causal pathways, and the left occipital region was a dominant hub of the abnormally stronger causal pathways.
SIGNIFICANCE: These findings provided direct evidence that chronic heroin abuse induced brain functional abnormalities. The potential value of combining effective connectivity analysis and brain sources decomposition methods in exploring brain alterations of heroin addicts was also implied. (Some illustrations presenting rely on the color.).

PMID: 28397708 [PubMed - as supplied by publisher]

http://ift.tt/2ookIv8