TUH EEG Corpus

Overview | Downloads | Publications | FAQ


The goal of this project is to enable deep learning research in neuroscience by releasing the largest publicly available database of EEG recordings. This ongoing project currently includes over 30,000 EEGs spanning the years from 2002 to present.

EEG Sensors

In addition to the raw EEG data, patient medical histories and physician interpretations of the data are available. The EEG data consists of a minimum of 24 channels of 16-bit data sampled at 250 Hz. Along with the sampled data, we have annotations generated by physicians for a small subset of the data.

Several important subsets of the data are being released, including data to support seizure detection and normal/abnormal classification.
  What's New:
  • (20170923) TUH EEG Corpus (v1.0.0): This is the official version of the TUH EEG Corpus. This release contains sessions recorded between 2002 and 2015. There are 13,500 patients and 23,218 sessions with paired EEG reports.

  • (20170920) The TUH EEG Slowing Corpus (v1.0.0): A small data set that can be used to study the difference between seizures and slowing in EEGs.

  • (20170913) The TUH Abnormal EEG Corpus (v1.1.2): A bug fix release in which age information was corrected for 38 files.
Overview: Review the status of the EEG project and learn more about the history of this project.

Downloads: Download the latest version of the database and view earlier versions.

Publications: Access background information related to this project.


Up | Home | Courses | Projects | Proposals | Publications
Please direct questions or comments to joseph.picone@gmail.com