基于独立成分分析的锰暴露焊工脑动态功能连接网络研究OA

Objective:To investigate alterations in dynamic functional connectivity network(dFNC)and its temporal properties among welders with occupational manganese(Mn)exposure.Materials and Methods:We conducted a study on 25 Mn-exposed welders and 29 healthy control group matched with welders.Both groups underwent resting-state functional magnetic resonance imaging(rs-fMRI)scan on a 3.0 T MRI scanner.Following standard preprocessing,group independent component analysis(ICA)was performed to identify resting-state networks.dFNC was computed using a sliding window correlation approach,and the resulting connectivity matrices were clustered into recurring brain states via the K-means algorithm.Group differences in state-specific dFNC patterns and temporal metrics were then evaluated.Furthermore,correlations between aberrant dFNC features and clinical indicators were examined in the welder group.Results:Compared with HCs,Mn-exposed welders showed significantly reduced dynamic functional connectivity between the salience network(SN)-executive control network(ECN),the default mode network(DMN)-language network(LAN),and the SN-DMN(t=-3.18、-3.31、-3.11,false discovery rate correction,P＜0.05).However,no significant between-group differences were found in other dFNC features(P＞0.05).Conclusions:The dFNC between the SN,ECN and DMN of Mn-exposed welders were significantly reduced,which provided evidence value for understanding the neuropathological mechanisms related to Mn overexposure and early prevention of Mn poisoning.