A multicenter ROI-level SLE neuroimaging dataset of rs-fMRI time series and DTI connectivity matrices from 631
Peng Ding1, Yifan Yang1, Bailing Tian2
1Department of Rheumatology and Immunology, First Affiliated Hospital of Kunming Medical University, Kunming, China.
Abstract:
Systemic lupus erythematosus (SLE) is a chronic autoimmune disease that frequently affects the central nervous system. Publicly available SLE neuroimaging datasets remain limited in scale and modality. Here, we present a cross-sectional multicenter dataset comprising resting-state functional MRI (rs-fMRI) and diffusion tensor imaging (DTI) data from 631 participants (396 SLE patients, 235 healthy controls) across four Chinese clinical centers. The released dataset does not include individual-level raw DICOM or NIfTI images; instead, all shared imaging files are de-identified, processed ROI-level derivatives, including AAL-based BOLD time series and DTI-derived structural connectivity matrices weighted by fractional anisotropy and fiber number. ROI-level BOLD time series and DTI-derived structural connectivity matrices are paired in three centers. Data underwent rigorous quality control, including head-motion screening and outlier detection; cross-center comparability was further evaluated using ComBat harmonization of selected derived validation features, including regional ALFF and regional mean FA strength, for which no significant inter-site differences remained after harmonization. Clinical and neuropsychological assessments are also available. This dataset supports diverse computational neuroimaging analyses, including functional characterization, structural connectivity modeling, and structure-function coupling studies.
