A new subset of the popular open source electroencephalogram (EEG) corpus – TUH EEG: - The Temple University Artifact Corpus (TUAR) consists of high yield artifact files annotated using a five-way classification system: 1. Chewing (CHEW): An artifact resulting from the tensing and relaxing of the jaw muscles. 2. Electrode (ELEC): An artifact that encompasses various electrode related phenomena. 3. Eye Movement (EYEM): A spike-like waveform created during patient eye movement. 4. Muscle (MUSC): A common artifact with high frequency, sharp waves corresponding to patient movement. 5. Shiver (SHIV): A specific and sustained sharp wave artifact that occurs when a patient shivers. - EEG artifacts are waveforms that are not of cerebral origin and may have been affected by several external and physiological factors. - These artifacts cause false alarms in seizure prediction machine learning systems. This corpus was developed to support research and evaluation of artifact suppression technology.
Paper | Code | Results | Date | Stars |
---|