A study on MRI dynamic functional connectivity based on independent component analysis in patients with neuropsychiatric systemic lupus erythematosus
10.3760/cma.j.cn112149-20191230-01017
- VernacularTitle:基于独立成分分析的神经精神性红斑狼疮患者MRI动态功能连接研究
- Author:
Man XU
1
;
Kangkang XUE
;
Junying CHENG
;
Yong ZHANG
;
Jingliang CHENG
Author Information
1. 郑州大学第一附属医院磁共振科 450052
- From:
Chinese Journal of Radiology
2020;54(11):1066-1072
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To investigate the abnormal dynamic characteristics of brain functional connectivity in patients with neuropsychiatric systemic lupus erythematosus (NPSLE) and its correlation with clinical indicators by using dynamic functional connectivity (dFC) analysis based on independent component analysis (ICA).Methods:The clinical and imaging data of female NPSLE patients diagnosed in the First Affiliated Hospital of Zhengzhou University from July 2018 to September 2019 were prospectively collected. The levels of complement C3, C4, CH50, glucocorticoid prednisone dosage, systemic lupus erythematosus disease activity index (SLEDAI) score and Systemic Lupus International Collaborating Clinics (SLICC)/American College of Rheumatology (ACR) damage index (SDI) score were recorded. Age and sex matched healthy controls (HC) were enrolled at the same time. All subjects underwent resting state functional MRI (rs-fMRI). The spatial group independent component analysis was performed on MRI data using GIFT software, and 29 independent components (IC) were selected as internal connectivity network; five functional connectivity states and three dFC indexes (fraction time, mean dwell time, number of transitions) were obtained by using sliding time window technology. Two independent sample t test was used to calculate the difference of functional connectivity in different states. Mann-Whitney U test was used to calculate the difference of dFC indexes between groups. Spearman correlation analysis was used to calculate the correlation between dFC index and clinical data in NPSLE group. Results:A total of 45 NPSLE patients and 35 HC patients were enrolled. There was no significant difference in age and education level between the two groups ( t=-0.327, -0.460, P>0.05). Compared with HC group, NPSLE group had higher fraction time and longer mean dwell time ( Z=-2.496, -2.462, P<0.05); in state3 strong connection, compared with HC group, functional connectivity (FC) between posterior cerebellar lobe (IC39) and basal ganglia (IC10) was enhanced ( t=-5.201, P<0.05); FC was found decreased between posterior cerebellar lobe (IC39) and temporal lobe (IC5), temporal lobe (IC7), superior parietal lobe (IC65) ( t=4.212, 5.572, 4.415, P<0.05), as well as between paracentral lobular region (IC12) and posterior cingulate gyrus (IC15) ( t=3.893, P<0.05) in NPSLE group. The SDI score of NPSLE patients was negatively correlated with the fraction time and mean dwell time of state1 and state3 strong connection state ( P<0.05), and the SLEDAI score was negatively correlated with the fraction time and mean dwell time of state1 and state2 ( P<0.05). The SDI and SLEDAI scores were positively correlated with the fraction time and mean dwell time of state4 weak connection state, respectively ( P<0.05). The levels of serum complement C3, C4 and CH50 in NPSLE patients were positively correlated with the number of transitions ( r=0.428, 0.354, 0.385, P<0.05), and the dosage of prednisone was negatively correlated with the number of transitions ( r=-0.466, P<0.05). The validation analysis results showed the experimental results could be effectively repeated. Conclusion:The dFC analysis method based on ICA can effectively identify the alterations of brain functional connectivity on a shorter time scale, which may provide a new perspective for further exploration of the neuroimaging mechanism of cognitive impairment in NPSLE.
