Liu2025 — Lower limb motor imagery EEG dataset for stroke patients (re-hosted from Figshare, BIDS-converted)
收藏资源简介:
Re-hosted EEG data from Liu et al. (2025): Lower limb motor imagery dataset for stroke patients, originally deposited on Figshare under a CC BY 4.0 license. Data has been converted from non-standard BIDS (EEGLAB .set) to proper BIDS format (BrainVision .vhdr/.vmrk/.eeg) using mne_bids.write_raw_bids(). Paper Y. Liu, Z. Gui, D. Yan, Z. Wang, R. Gao, N. Han, J. Chen, J. Wu, D. Ming, “Lower limb motor imagery EEG dataset based on the multi-paradigm and longitudinal-training of stroke patients,” Scientific Data, 12, 314, 2025.DOI: 10.1038/s41597-025-04618-4 Paradigm TaskGait motor imagery vs idle state (binary)Trials~40 MI + 40 idle per session (4 runs × 10+10)Timing3 s baseline → 1 s cue → 5 s MI/idle → restEventsCode 3 = gait MI onset, code 9 = idle onset Sessions ses-prePre-treatment (conventional MI, all 27 subjects)ses-misesMI + Sequential Electrical Stimulation (16 subjects)ses-miiesMI + Invariable Electrical Stimulation (16 subjects)ses-postPost-treatment assessment (all 27 subjects)ses-followFollow-up assessment (all 27 subjects) Participants 27 stroke patients in recovery phase Age: 33–68 years (mean 51.5, SD 9.31) Sex: 23 male, 4 female Affected side: 9 left hemiplegia, 18 right hemiplegia Paradigm group (16 subj): all 5 sessions Longitudinal group (11 subj): pre/post/follow only Recording setup ParameterValueSystemNeuSen W (Neuracle, Inc.)Channels60 EEG + 1 ECG + 2 HEOG + 2 VEOG = 64 + 1 triggerSampling rate1000 HzReferenceCommon averageGroundAFzMontage10-20 system BIDS conversion Source: EEGLAB .set files (non-standard BIDS naming) Output: BrainVision format (.vhdr/.vmrk/.eeg) in proper BIDS Tool: mne_bids.write_raw_bids() Channel types: ECG/EOG correctly labelled in channels.tsv Events: preserved as BrainVision markers Reading the data import mne_bids bids_path = mne_bids.BIDSPath( subject='01', session='pre', task='MotorImagery', run='01', datatype='eeg', root='path/to/bids_root' ) raw = mne_bids.read_raw_bids(bids_path) License CC BY 4.0 as per the original paper and Figshare dataset.



