[Eeglablist] Please: need help with EEGLAB

[Makoto Miyakoshi]

Dear Nahla,

Let's say I have 100 EPI scans. This means that you need 100 point
data. So the first thing you have to do is downsampling your EEG data.
Then, you may want to convolve hrf. If your TR is 3 sec, spm_hrf(3)
creates it. You may want to discard both ends (both EEG and EPI). You
finally input it as 'multiple regressor' in SPM.

If you have further questions let me know.

Makoto


2012/10/11 Nahla <nahla.faizo at uqconnect.edu.au>:
> Hi again Makoto,
>
> I was wondering how to make an EEG information (for example spikes) an input
> function for the SPM analysis basically and technically?
>
>
> Regards
> Nahla
>
> -----Original Message-----
> From: Makoto Miyakoshi [mailto:mmiyakoshi at ucsd.edu]
> Sent: Friday, 14 September 2012 8:08 AM
> To: nahla.faizo at uqconnect.edu.au
> Cc: eeglab at sccn.ucsd.edu
> Subject: Re: Please: need help with EEGLAB
>
> Dear Nahla,
>
> Plot-Channel data (scroll)
> You can choose a portion of data to discard interactively.
>
> You may find our EEGLAB wiki useful.
> http://sccn.ucsd.edu/wiki/EEGLAB
>
> Makoto
>
> 2012/9/13  <nahla.faizo at uqconnect.edu.au>:
>> Hi Makoto
>>
>> As processing EEG data is requiring long time, I want to create a small
> data set from my own data, which is saved as one file as .eeg.
>>
>> Is that possible and how?
>>
>> I think it is much more easier to use small dataset to play with the
> setting.
>>
>> Cheers
>> Nahla
>> Sent via BlackBerryR from Vodafone
>>
>> -----Original Message-----
>> From: Makoto Miyakoshi <mmiyakoshi at ucsd.edu>
>> Date: Thu, 13 Sep 2012 11:06:50
>> To: Ms Nahla Lutfi A Faizo<nahla.faizo at uqconnect.edu.au>
>> Reply-To: <mmiyakoshi at ucsd.edu>
>> Cc: EEGLAB List<eeglablist at sccn.ucsd.edu>
>> Subject: Re: Please: need help with EEGLAB
>>
>> Dear Nahla,
>>
>>> Yes, they are from the same dataset.
>>
>> You mean the same dataset showed different three results? If that's
>> the case, you surely used different parameters. Anyway, I still don't
>> see your point. If the differences you showed are not from individual
>> differences (including differences in noise), what made them
>> different?
>>
>>> I have also another question. In the FMRIB plug-in for EEGLAB tutorial,
> the 7th step is to remove noise. it is saying to click on tools--> filter
> the data and enter [45 55] as the notch filter range. However, when I click
> on tools in the eeglab GUI, it gives 4 options. I choose the basic FIR which
> gives a window with lower and high frequency pass band and other cells. I
> tried different ways to enter the 45 55 for notch filter but it keep giving
> me an error "d MATLAB expression".
>>
>> I would rather use simple a low-pass filter below your line noise
>> frequency (is it 50 Hz or 60 Hz?).  Make sure you use fir1 option when
>> applying the filter.
>>
>>> So, what is the proper way to do the settings to the 50 Hz notch filter?
>>
>> If you still want to apply notch, enter 45 for a lower edge and 55 for
>> a higher edge, and check 'notch filter the band...' to run 45-55
>> bandstop.
>>
>> Makoto
>>
>>
>>
>>
>> 2012/9/12 Ms Nahla Lutfi A Faizo <nahla.faizo at uqconnect.edu.au>:
>>> Yes, they are from the same dataset.
>>> What other options do you mean? Is there another way to remove fMRI
> gradient artefact?
>>> I have also another question. In the FMRIB plug-in for EEGLAB tutorial,
> the 7th step is to remove noise. it is saying to click on tools--> filter
> the data and enter [45 55] as the notch filter range. However, when I click
> on tools in the eeglab GUI, it gives 4 options. I choose the basic FIR which
> gives a window with lower and high frequency pass band and other cells. I
> tried different ways to enter the 45 55 for notch filter but it keep giving
> me an error "d MATLAB expression".
>>>
>>> So, what is the proper way to do the settings to the 50 Hz notch filter?
>>>
>>> Nahla
>>>
>>> ________________________________________
>>> From: Makoto Miyakoshi [mmiyakoshi at ucsd.edu]
>>> Sent: Thursday, 13 September 2012 12:26 PM
>>> To: Ms Nahla Lutfi A Faizo
>>> Cc: EEGLAB List
>>> Subject: Re: Please: need help with EEGLAB
>>>
>>> Dear Nahla,
>>>
>>> Are they from the same dataset? If not, the differences should come
>>> from variation of the data quality from dataset to dataset.
>>>
>>> If you are not satisfied with the performance (after studying the
>>> options deeply), you may want to use other method than this one.
>>>
>>> Makoto
>>>
>>> 2012/9/12 Ms Nahla Lutfi A Faizo <nahla.faizo at uqconnect.edu.au>:
>>>> Soory, I forgot to mention that I can still see the gradient artefacts
> on some segments of the recording.
>>>>
>>>> I attached 3 figure. The first one with the gradient artifact. The
> second fiure shows another segment without artefact. The final one shows
> some artifact, which I ma not sure what it is.
>>>>
>>>> So, why is this variation in the data??
>>>>
>>>> Nahla
>>>> ________________________________________
>>>> From: Ms Nahla Lutfi A Faizo
>>>> Sent: Thursday, 13 September 2012 11:11 AM
>>>> To: mmiyakoshi at ucsd.edu
>>>> Subject: RE: Please: need help with EEGLAB
>>>>
>>>> Hi Makoto
>>>>
>>>> I removed the ECG channel and started the Fastr processing. It finished
> in 6 and half hours.
>>>> Now, when I viewed the data from plot--> channel data (scroll), still I
> can see the gradient artefact.
>>>> I make sure that the data I opened is the new set that I've saved after
> the fastr was done.
>>>>
>>>> Help please !!
>>>>
>>>> Cheers
>>>> Nahla
>>>>
>>>> ________________________________________
>>>> From: Makoto Miyakoshi [mmiyakoshi at ucsd.edu]
>>>> Sent: Wednesday, 12 September 2012 11:49 AM
>>>> To: Ms Nahla Lutfi A Faizo
>>>> Cc: EEGLAB List
>>>> Subject: Re: Please: need help with EEGLAB
>>>>
>>>> Dear Nahla,
>>>>
>>>> As a part of proprocessing, could you remove the channel from GUI?
>>>>
>>>> Makoto
>>>>
>>>> 2012/9/11 Ms Nahla Lutfi A Faizo <nahla.faizo at uqconnect.edu.au>:
>>>>> Hi Makoto
>>>>>
>>>>> Channel 31 is the POz. The ECG channel is number 32.
>>>>> When I looked at the channel info for the ECG, all cells are clear.
> There is no information about this channel except its number (32).
>>>>>
>>>>> So, I guess the problem that when the processing reaches the ECG
> channel, which is ofcourse not similar the other EEG channels, it gives the
> error.
>>>>>
>>>>> So, how to fix it?
>>>>>
>>>>> Cheers
>>>>> Nahla
>>>>>
>>>>> ________________________________________
>>>>> From: Makoto Miyakoshi [mmiyakoshi at ucsd.edu]
>>>>> Sent: Wednesday, 12 September 2012 11:11 AM
>>>>> To: Ms Nahla Lutfi A Faizo
>>>>> Cc: EEGLAB List
>>>>> Subject: Re: Please: need help with EEGLAB
>>>>>
>>>>> Dear Nahla,
>>>>>
>>>>>> I've checked the bug list and I found one (819) that have the same
> error but I am not sure if is due to the same problem that I have.
>>>>>
>>>>> Very good.
>>>>> What is your channel 31? EOG or ECG? Check if Ch31 has the same
>>>>> structure as other EEG channels do.
>>>>>
>>>>> Makoto
>>>>>
>>>>> 2012/9/11 Ms Nahla Lutfi A Faizo <nahla.faizo at uqconnect.edu.au>:
>>>>>> Dear Makoto
>>>>>>
>>>>>> Thank you for your promp responses.
>>>>>> You are right, I checked the data and it looks fine, its about the
> scaling as you have said.
>>>>>>
>>>>>> Now, I've done the fastr processing to remove gradient artifacts, the
> process started fine, until it reached electrode 31 (which took about 3
> hours), then a message error came up "matrix dimensions must agree".
>>>>>>
>>>>>> I've checked the bug list and I found one (819) that have the same
> error but I am not sure if is due to the same problem that I have.
>>>>>> How can I deal with this please?
>>>>>> I am using the latest version of eeglab 11.0.3.1b.
>>>>>>
>>>>>> Cheers
>>>>>> Nahla
>>>>>> ________________________________________
>>>>>> From: Makoto Miyakoshi [mmiyakoshi at ucsd.edu]
>>>>>> Sent: Tuesday, 11 September 2012 1:37 AM
>>>>>> To: cvandermaas at hotmail.com
>>>>>> Cc: Ms Nahla Lutfi A Faizo
>>>>>> Subject: Fwd: Please: need help with EEGLAB
>>>>>>
>>>>>> Dear Christian,
>>>>>>
>>>>>> Thank you for your offer. I told him that I don't see any problem
>>>>>> in his data. What do you think?
>>>>>>
>>>>>> Makoto
>>>>>>
>>>>>>
>>>>>> ---------- Forwarded message ----------
>>>>>> From: Ms Nahla Lutfi A Faizo <nahla.faizo at uqconnect.edu.au>
>>>>>> Date: 2012/9/6
>>>>>> Subject: RE: Please: need help with EEGLAB
>>>>>> To: "mmiyakoshi at ucsd.edu" <mmiyakoshi at ucsd.edu>
>>>>>>
>>>>>>
>>>>>> Dear Makoto,
>>>>>>
>>>>>> I downloaded the latest version of eeglab which is eeglab11_0_3_1b.
>>>>>> When I imported the EEG data that I acquired during fMRI, it
>>>>>> looked weird (attached figure).
>>>>>>
>>>>>> What could be the problem now? Bear in mind I did import EEG data
>>>>>> from an earlier experiment and il looked different but more
>>>>>> realistic(attached figure).
>>>>>>
>>>>>> Can you help me with this please.
>>>>>>
>>>>>> Cheers
>>>>>> Nahla
>>>>>>
>>>>>> ________________________________________
>>>>>> From: Makoto Miyakoshi [mmiyakoshi at ucsd.edu]
>>>>>> Sent: Friday, 7 September 2012 8:27 AM
>>>>>> To: Ms Nahla Lutfi A Faizo
>>>>>> Cc: eeglab at sccn.ucsd.edu
>>>>>> Subject: Re: Please: need help with EEGLAB
>>>>>>
>>>>>> Dear Nahla,
>>>>>>
>>>>>> Would you please make sure that you are using the latest version
>>>>>> of EEGLAB? Which version are you using right now?
>>>>>>
>>>>>> Makoto
>>>>>>
>>>>>>
>>>>>> Dear Sir/Madam,
>>>>>>
>>>>>> I am doing EEG-fMRI for my research project.
>>>>>> Now, after I imported the data using FILE-IO toolbox on eeglab, as
>>>>>> described in the tutorial on EEG-fMRI data, i went to
>>>>>> tools-->FMRIB
>>>>>> tools-->FASTR to remove the gradient artifacts. I've used the
>>>>>> tools-->default
>>>>>> numbers that the tutorial suggested which were: low pass filter
>>>>>> 70, interpolation 10, averaging window length 30, artefact timing
>>>>>> event slice, non-EEG channels 32 (which is the EEG), but it keeps
>>>>>> giving me an error message that subscipt indices must either be
>>>>>> real positive integers or logicals.
>>>>>>
>>>>>> What does this mean and do you know how to deal with this problem?
>>>>>>
>>>>>>
>>>>>> Regards
>>>>>> Nahla Faizo
>>>>>> PhD student, Centre for Adbanced Imaging The University of
>>>>>> Queensland, Australia
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Makoto Miyakoshi
>>>>>> JSPS Postdoctral Fellow for Research Abroad Swartz Center for
>>>>>> Computational Neuroscience Institute for Neural Computation,
>>>>>> University of California San Diego
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Makoto Miyakoshi
>>>>> JSPS Postdoctral Fellow for Research Abroad Swartz Center for
>>>>> Computational Neuroscience Institute for Neural Computation,
>>>>> University of California San Diego
>>>>
>>>>
>>>>
>>>> --
>>>> Makoto Miyakoshi
>>>> JSPS Postdoctral Fellow for Research Abroad Swartz Center for
>>>> Computational Neuroscience Institute for Neural Computation,
>>>> University of California San Diego
>>>
>>>
>>>
>>> --
>>> Makoto Miyakoshi
>>> JSPS Postdoctral Fellow for Research Abroad Swartz Center for
>>> Computational Neuroscience Institute for Neural Computation,
>>> University of California San Diego
>>
>>
>>
>> --
>> Makoto Miyakoshi
>> JSPS Postdoctral Fellow for Research Abroad Swartz Center for
>> Computational Neuroscience Institute for Neural Computation,
>> University of California San Diego
>>
>
>
>
> --
> Makoto Miyakoshi
> JSPS Postdoctral Fellow for Research Abroad Swartz Center for Computational
> Neuroscience Institute for Neural Computation, University of California San
> Diego
>
>



-- 
Makoto Miyakoshi
JSPS Postdoctral Fellow for Research Abroad
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego