Four-class motor imagery dataset (MI4)

该数据集包含了9名受试者在两个不同日子进行的两次试验中收集的信息，分为四类：左手、右手、脚和舌头想象动作。数据采用22通道的脑电图信号，以256赫兹的采样率获取，每个想象提示后提取[0, 4]秒的数据，并使用[8, 32]赫兹的带通滤波进行处理。每个受试者每个类别包含144个脑电图周期。该任务旨在进行运动想象分类。

This dataset collects information from 9 subjects across two experimental sessions held on two separate days, with the data categorized into four classes: motor imagery of the left hand, right hand, foot, and tongue. The data comprises 22-channel electroencephalogram (EEG) signals acquired at a sampling rate of 256 Hz. Following each imagery cue, data segments spanning 0 to 4 seconds were extracted and processed via a band-pass filter with a passband of 8 to 32 Hz. Each subject has 144 EEG trials per class. The primary task of this dataset is motor imagery classification.