slideshow 00 slideshow 01 slideshow 02 slideshow 03 slideshow 04 slideshow 05 slideshow 06 slideshow 07 slideshow 08 slideshow 09

Feature Of The Day

(20170923) v1.0.0 of the TUH EEG Corpus was finally released. See TUH EEG Corpus for more details.

(20170920) v1.0.0 of the TUH EEG Slowing Corpus has been released. See TUH EEG Slowing Corpus for more details.

Institute for Signal and Information Processing
Joseph Picone
on September 23, 2017 17:27
Release Announcement: The TUH EEG Corpus (v1.0.0)

After far too long a time... TUH EEG v1.0.0 is now available from our web site:

https://www.isip.piconepress.com/projects/tuh_eeg/downloads/tuh_eeg/v1.0.0/

The total release is 634G in tar format, and 980G uncompressed. That is too much to download over the web. Send us a disk including return postage and we will be happy to copy the data for you.

All of the EDF files should have valid headers and be viewable using some of the open source tools like EDFBrowser. Let us know if you find files that have problems.

There is a tar file containing only the reports here:

https://www.isip.piconepress.com/projects/tuh_eeg/downloads/tuh_eeg/v1.0.0/reports.tar.gz

We haven't released the spreadsheet yet describing the database. It will be similar to the one we released for the seizure corpus, describing the type of EEG. We also haven't updated the README file with some basic statistics. I should have the README file shortly and the spreadsheet in a few weeks at the latest.

Those of you who sent us disks should receive those disks by the end of the week. We will copy them on Monday and get them in the mail as soon as possible.

As always, let us know if you have questions.

Thanks to Lilly Veloso and James Riley McHugh for making this release happen. Cleaning up this amount of data was quite challenging given the noisy nature of the source data.

Looking towards the future, we will probably release v2.0.0, which will contain date through at least 2016, in about 6 months. We will also update the seizure corpus with more annotated data at this time.

-Joe Picone, on behalf of our Data and Resources Team

Joseph Picone
on September 22, 2017 18:41
stat command

So you learn something every day in this course... which is why I like teaching it.

Here is my new favorite command:

Mac:
isip027_[1]: stat -f "%N %z" x4.dat
x4.dat 3636406

Linux:
nedc_000_[1]: stat -c "%n %s" x5.dat
x5.dat 4191112

A nice way to particular information on a file. You can also use ls and awk together.

-Joe

Joseph Picone
on September 20, 2017 23:11
Release Announcement: The TUH EEG Slowing Corpus (v1.0.0)

Please see:

https://www.isip.piconepress.com/projects/tuh_eeg/html/downloads

for details about this release.

This is a database described in an upcoming publication:

von Weltin, E., Ahsan, T., Shah, V., Jamshed, D., Golmohammadi, M.,
Obeid, I., & Picone, J. (2017). Electroencephalographic Slowing: A
Source of Error in Automatic Seizure Detection. In J. Picone &
I. Obeid (Eds.), Proceedings of the IEEE Signal Processing in
Medicine and Biology Symposium (pp. 1–5). Philadelphia, Pennsylvania,
USA: IEEE.

A preprint is available at:

https://www.isip.piconepress.com/publications/unpublished/conferences/2017/ieee_spmb/slowing/paper_v13.docx

This is a small database we are using to study the problem of detection of slowing in EEGs.

Also, we are finally uploading v1.0.0 of TUH EEG to the web server. I hope we will have that online by the weekend.

As always, let us know if you have questions.

Joseph Picone
on September 13, 2017 13:50
Release Announcement: The TUH Abnormal EEG Corpus (v1.1.2)

Please see:

https://www.isip.piconepress.com/projects/tuh_eeg/downloads/tuh_eeg_abnormal/v1.1.2/_AAREADME.txt

for details about this release.

v1.1.2 contains the following enhancements:

- age information was corrected for 38 files

Thanks to one of our collaborators who pointed out this problem with v1.1.1.

As always, let us know if you have questions.

-Joe Picone

P.S. We are uploading v1.0.0 of the TUH EEG Corpus as we speak :)

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer