[Eeglablist] EEGLABTOOL PREPROCESSING

[Cedric Cannard]

Just reject the non-brain components with at least 95% confidence (for muscle and eye blinks) and 99% for other non-brain components, to preserve as much brain information that was not classified with high confidence.

FYI: you can use the PICARD algorithm, it's much faster and gives good results, similar to infomax. See https://urldefense.com/v3/__https://dl.acm.org/doi/abs/10.1109/ICASSP.2018.8461662__;!!Mih3wA!Ga_hOVFF5eca7KoEkATBeLjrDpKsbsY874pb1wIYa3ujKSVM5BHmPVF-oxyiuJe-pHxXUjx33jrWMeKQbyW6zsAXEQ$  and https://urldefense.com/v3/__https://arxiv.org/pdf/1706.08171.pdf__;!!Mih3wA!Ga_hOVFF5eca7KoEkATBeLjrDpKsbsY874pb1wIYa3ujKSVM5BHmPVF-oxyiuJe-pHxXUjx33jrWMeKQbyUpbQ7fmA$  for more details.

> "can you also suggest me some new classification model to classify eeg signals which has not been done so far."

What do you mean? Classify what?

Cedric

Sent with [Proton Mail](https://urldefense.com/v3/__https://proton.me/__;!!Mih3wA!Ga_hOVFF5eca7KoEkATBeLjrDpKsbsY874pb1wIYa3ujKSVM5BHmPVF-oxyiuJe-pHxXUjx33jrWMeKQbyXC25KISg$ ) secure email.

------- Original Message -------
On Friday, June 16th, 2023 at 3:25 AM, Soniya Usgaonkar <soniya at gec.ac.in> wrote:

> Thanks for the advice sir .but i have already rejected all other components .because when we use ica decompose it gives me ica components like component 1 brain 90percent..component two other 60percent etc.what i have done is i have rejected many other since it showed me brain activity 20 percent.have i done it correctly sir?kindly help.i have 64 channel eeg data.i have to preprocess total of 24 subject and it is taking lot of time to run ica on each subject .i have already finished with 12 subjects.sir can you also suggest me some new classification model to classify eeg signals which has not been done so far.
>
> Regards and Thanks
> Soniya Usgaonkar,
> Assistant Professor,
> Department Of Information Technology,
> Goa college Of Engineering,
> Farmagudi-Goa
>
> On Thu, Jun 15, 2023 at 9:40 PM Cedric Cannard <ccannard at protonmail.com> wrote:
>
>> Dear Soniya,
>>
>> You can, but I wouldn't reject anything below 95% confidence (maybe even 99% for heart and noise components). The current default is to reject only eye and muscle components with at least 90% confidence. I don't recommend rejecting the "Other" category as it may contain some meaningful brain information.
>>
>> Cedric Cannard
>>
>> Sent with [Proton Mail](https://urldefense.com/v3/__https://proton.me/__;!!Mih3wA!Ga_hOVFF5eca7KoEkATBeLjrDpKsbsY874pb1wIYa3ujKSVM5BHmPVF-oxyiuJe-pHxXUjx33jrWMeKQbyXC25KISg$ ) secure email.
>>
>> ------- Original Message -------
>> On Thursday, June 15th, 2023 at 3:07 AM, Soniya Usgaonkar <soniya at gec.ac.in> wrote:
>>
>>> Thanks, I'll check it out! .i had one more query when i use ica decomposition option of eeg lab tool .i get ica components having brain activity and eye and channel and others .can i retain only brain activity and reject all other?
>>>
>>> Regards and Thanks
>>> Soniya Usgaonkar,
>>> Assistant Professor,
>>> Department Of Information Technology,
>>> Goa college Of Engineering,
>>> Farmagudi-Goa
>>>
>>> On Tue, Jun 13, 2023 at 11:02 PM Cedric Cannard <ccannard at protonmail.com> wrote:
>>>
>>>> Dear Second,
>>>>
>>>> This is Biosemi data right? You have alphabetic labeling format and the default BEM channel template is expecting a different labeling format (e.g., A1 = Fp1).
>>>>
>>>> You can try to load this location template instead: standard_alphabetic.elc in the same folder (eeglab\plugins\dipfit5.2\standard_BEM\elec)
>>>>
>>>> Or you can convert them to the right labeling format, which I recommend as it facilitates the interpretation of results (F for frontal, P for parietal, etc.).
>>>>
>>>> See here to convert them (if you have 64-channel Biosemi):
>>>> https://urldefense.com/v3/__https://www.biosemi.com/pics/cap_64_layout_medium.jpg__;!!Mih3wA!Ga_hOVFF5eca7KoEkATBeLjrDpKsbsY874pb1wIYa3ujKSVM5BHmPVF-oxyiuJe-pHxXUjx33jrWMeKQbyXobg3Mgw$ 
>>>>
>>>> Here are the other montages if you need: https://urldefense.com/v3/__https://www.biosemi.com/headcap.htm__;!!Mih3wA!Ga_hOVFF5eca7KoEkATBeLjrDpKsbsY874pb1wIYa3ujKSVM5BHmPVF-oxyiuJe-pHxXUjx33jrWMeKQbyUKfUxPNA$ 
>>>>
>>>> Cedric Cannard
>>>>
>>>> Sent with Proton Mail secure email.
>>>>
>>>> ------- Original Message -------
>>>> On Monday, June 12th, 2023 at 9:42 PM, Soniya Usgaonkar via eeglablist <eeglablist at sccn.ucsd.edu> wrote:
>>>>
>>>>> Hello Sir.
>>>>> I am using compiled version of EEGTOOL for preprocessing my dataset of EEG
>>>>> signals.its reading and importing the dataset . When i have to read channel
>>>>> locations i am giving channel .txt file as an input i am getting following
>>>>> error
>>>>> readlocs(): 'txt' format assumed from file extension
>>>>> 3 channel with the same label found and imported
>>>>> Channel lookup: no location for
>>>>> A1,A2,A3,A4,A5,A6,A7,A8,A9,A10,A11,A12,A13,A14,A15,A16,A17,A18,A19,A20,A21,A22,A23,A24,A25,A26,A27,A28,A29,A30,A31,A32,B1,B2,B3,
>>>>> B4,B5,B6,B7,B8,B9,B10,B11,B12,B13,B14,B15,B16,B17,B18,B19,B20,B21,B22,B23,B24,B25,B26,B27,B28,B29,B30,B31,B32,EXG1,EXG2,EXG3,EXG4,EXG5,EXG6,EXG7,EXG8,GSR1,GSR2,Erg1,Erg2,Resp,Plet,TempRemoving
>>>>> 15 channel(s)...
>>>>> i am not able to find out why this error . Can you please help sir?
>>>>> Regards and Thanks
>>>>> Second Usgaonkar,
>>>>> Assistant Professor,
>>>>> Department Of Information Technology,
>>>>> Goa college Of Engineering,
>>>>> Farmagudi-Goa
>>>>> _______________________________________________
>>>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>>>> To unsubscribe, send an empty email to 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