TUH Abnormal EEG Corpus

该数据集是从Temple大学医院（TUH）的脑电图（EEG）语料库中选取的一个子集，它平衡地代表了临床正常和异常的脑电图记录。数据集包含了多样化的人口统计学特征，并经过了降采样、分割和Z标准化等预处理步骤。规模上，该数据集包含2,993个以欧洲数据格式（EDF）存储的脑电图文件，其中2,717个用于训练，276个用于评估。其任务是进行脑电图异常检测。

This dataset is a curated subset extracted from the electroencephalogram (EEG) corpus of Temple University Hospital (TUH), which provides a balanced representation of clinically normal and abnormal EEG recordings. It includes diverse demographic characteristics and has undergone preprocessing procedures including downsampling, segmentation, and Z-score normalization. In terms of scale, the dataset comprises 2,993 EEG files stored in the European Data Format (EDF), with 2,717 allocated for training and 276 for evaluation. The core task of this dataset is EEG abnormality detection.