[Eeglablist] Extracting features from bigdata

[Tarik S Bel-Bahar]

Hello, just
​some ​
extra notes below for you
​. ​
​Good luck!



​*******************************************************************************​

​
​
"Feature-extraction" can mean many different things. Also, remember that
Google and Google scholar can help you find a lot of what you need.
Try reviewing a dozen or so recent articles in developmental or
neuroscience journals on your EEG topic. There you can find the best
metrics, and also contact the authors for the metrics.​
Beginner books like Luck's Introduction to ERPs and Mike Cohen's
Time-Frequency books are certainly also useful if you've not read them in
full yet.

TAPEEG may have some source estimation features which are cutting-edge but
have not been activated yet.
​ ​
In many cases you have to build the tools yourself to compute the metrics.
One can learn a lot
​ by doing so. ​


You can
​search
 on Google the many for-pay software/metrics available from companies like
EGI, Brain Vision, Brain Voyager, and other EEG hardware, software, or
analysis companies.

You can search for
​and closely review ​
many
​free and for-pay ​
EEG tools here:
​ ​
https://www.nitrc.org/

​Y​
ou can learn about most available public/aceadadmic EEG softwares here at
the full issue on
Academic Software Applications for Electromagnetic Brain Mapping Using MEG
and EEG
http://www.hindawi.com/journals/cin/2011/972050/

​Several companies sell suites you can buy, like Neuroguide from Applied
Neuroscience that sort of give you a report full of metrics from EEG.​ See
also for example the B-Alert metrics.
​​
You can also try the Hermes toolbox, which computes at least a dozen
different EEG-based connectivity metrics.
You can also try the NBT toolbox and also the Connectivity Toolbox (many
metrics) from Sporns. For machine learning-based feature extraction
consider the features computed in BCI toolbox.

You might
​ ​
also
​ ​
want to
​ thoroughly review the
 QEEG literature,
​ ​
or the machine learning literature,
or the traditional EEG/ERP literature (e.g., power, topography, erps).
You may also want to review recent papers about connectivity measures
​, or
the TBI literature
​.

More recent or advanced measures will tend not to have toolboxes or
published methods.
Please
​ review
 ​
review the lists of measures in  each of the articles
​I've listed below for you​
, all findable Google Scholar.

Feature extraction of electroencephalogram (EEG) signal-A review
Aligning strategies for using EEG as a surrogate biomarker: a review of
preclinical and clinical research
Signal processing techniques applied to human sleep EEG signals—A review
Interpreting EEG alpha activity
Opportunities and methodological challenges in EEG and MEG resting state
functional brain network research
Research Review: use of EEG biomarkers in child psychiatry research–current
state and future directions

​
​*******************************************************************************​
​







On Mon, Oct 19, 2015 at 4:42 AM, Md. Abdul Awal <awalece04 at yahoo.com> wrote:
>
> Thanks for your useful suggestions. I went through the TAPEEG and it is
also useful. However, this software doesn't extract much features.
>
> I need to extract different features from these neonatal EEG to correlate
with underdevelopmental outcome.
> Do respected EEGLABListers know any software or related materials for
extracting different features to predict outcome.
>
> Thanks in advance.
>
> with regards,
> ------------------------
> Md. Abdul Awal
> The University of Queensland, Australia
>
> ________________________________
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20151028/7f871287/attachment.html>