[Eeglablist] ICA training at 0.1Hz- can it be done?

[Ahmad, Jumana]

Thank you.

I do, on the continuous data. I use continuous artefact rejection. I also reject bad channels etc. But I have a very large dataset and the data quality varies, but I consistently get a bad decomposition at 0.1.

I did think about segmenting my data first but I need long segments for TFA. Since I have short trials, if I did segment before ICA I’d have to use longer overlapping trials (for time frequency later on) - and overlapping segments isn’t ideal for ICA. I did try on ICA shorter non overlapping segments too. It still didn’t decompose well for me. I’ve tried several algorithms (tunics, binica and amica).

But, I have tried ICA to 0.1 on a dataset recorded with fewer electrodes and this seems to work better.

Best wishes,
Jumana

On 12 Jun 2018, at 19:15, Eric Fields <eric.fields at bc.edu<mailto:eric.fields at bc.edu>> wrote:

I have done several studies running ICA on 0.1 Hz filtered data and got good ICA results. Do you clean your data prior to running ICA by removing epochs/segments with bad drift and idiosyncratic artifact? I've found this is important especially for relatively messy subjects.

Eric

-----
Eric Fields, Ph.D.
Postdoctoral Fellow
Cognitive and Affective Neuroscience Laboratory<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww2.bc.edu%2Felizabeth-kensinger%2F&data=01%7C01%7Cjumana.ahmad%40kcl.ac.uk%7Cd583ac01c17840ba325708d5d0907881%7C8370cf1416f34c16b83c724071654356%7C0&sdata=CYTBiowip3dGk9VVl1x1byKJ6j%2BOJJO3yChyvcThUWM%3D&reserved=0>, Boston College
Aging, Culture, and Cognition Laboratory<https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.brandeis.edu%2Fgutchess%2F&data=01%7C01%7Cjumana.ahmad%40kcl.ac.uk%7Cd583ac01c17840ba325708d5d0907881%7C8370cf1416f34c16b83c724071654356%7C0&sdata=RI1Gw487V%2BMCLRzQ08SZdEjq0OkwP0fk1SD6IPFlk6I%3D&reserved=0>, Brandeis University
eric.fields at bc.edu<mailto:eric.fields at bc.edu>

On Mon, Jun 11, 2018 at 5:17 AM, Ahmad, Jumana <jumana.ahmad at kcl.ac.uk<mailto:jumana.ahmad at kcl.ac.uk>> wrote:
Hi all,
I ran ICA with either a 0.1Hz or 1Hz filter. 0.1 does not give stable components- multiple blinks, and fuzzy looking brain maps.

However, I wanted to avoid transferring the weights from 1Hz to 0.1, incase the 0.1Hz, which does not get fully characterised within the training, subsequently gets removed when we exclude eye blinks etc. Although this is fairly unlikely, I did wonder if there would be an alternative to help my decomposition at 0.1Hz?  Something that would also not distort the ERP.

I wondered about de-trending, but again, I am unsure how this would distort the ERP and I am not sure if this has been tested?

Best wishes,
Jumana

------------------------------------------
Jumana Ahmad
Post-Doctoral Research Worker in Cognitive Neuroscience
EU-AIMS Longitudinal European Autism Project (LEAP)

Room M1.09. Department of Forensic and Neurodevelopmental Sciences (PO 23) | Institute of Psychiatry, Psychology & Neuroscience | King’s College London | 16 De Crespigny Park | London SE5 8AF

Phone: 0207848 0260| Email: jumana.ahmad at kcl.ac.uk<mailto:antonia.sanjose at kcl.ac.uk> | Website: www.eu-aims.eu<https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.eu-aims.eu%2F&data=01%7C01%7Cjumana.ahmad%40kcl.ac.uk%7Cd583ac01c17840ba325708d5d0907881%7C8370cf1416f34c16b83c724071654356%7C0&sdata=POHMhdP8ZqM3J1PTE0bwdkh0SgTKvshcgTCQ5ktR7o0%3D&reserved=0> | Facebook: www.facebook.com/euaims<https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.facebook.com%2Feuaims&data=01%7C01%7Cjumana.ahmad%40kcl.ac.uk%7Cd583ac01c17840ba325708d5d0907881%7C8370cf1416f34c16b83c724071654356%7C0&sdata=k044vmcjOo40Boq%2F17Cp5tNKuNH9CxaoYS2iZuoUzCg%3D&reserved=0>


_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html<https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fsccn.ucsd.edu%2Feeglab%2Feeglabmail.html&data=01%7C01%7Cjumana.ahmad%40kcl.ac.uk%7Cd583ac01c17840ba325708d5d0907881%7C8370cf1416f34c16b83c724071654356%7C0&sdata=T0UHVqZOjrIjcdrl0N3YU7on7RFPa1qfpA2LQtSVgm4%3D&reserved=0>
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu<mailto:eeglablist-unsubscribe at sccn.ucsd.edu>
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu<mailto:eeglablist-request at sccn.ucsd.edu>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20180612/6dec1424/attachment.html>