NEDC TUH EEG Artifact Corpus
收藏资源简介:
NEDC TUH EEG Artifact Corpus是由IBM研究院澳大利亚分院创建的全球最大的开放源EEG伪影识别数据集,包含来自213名患者的259个EEG会话数据。数据集主要包含五种伪影事件:眼动、咀嚼、颤抖、电极弹跳和肌肉运动,以及非伪影的空类。创建过程涉及将原始EEG信号转换为22通道的TCP系统,并应用FFT技术提取特征。该数据集旨在通过自动化识别和移除EEG伪影,提高EEG信号分析的效率和准确性,从而辅助临床诊断。
NEDC TUH EEG Artifact Corpus, developed by the IBM Research Australia division, is the world's largest open-source EEG artifact recognition dataset, comprising 259 EEG session datasets from 213 patients. The dataset primarily covers five types of artifact events: eye movement, mastication, tremor, electrode pop, and muscle movement, as well as a non-artifact empty class. During its construction, raw EEG signals were converted into a 22-channel TCP system, and Fast Fourier Transform (FFT) was utilized to extract features. This dataset aims to improve the efficiency and accuracy of EEG signal analysis via automated identification and removal of EEG artifacts, thereby assisting clinical diagnosis.




