[Eeglablist] ASR algorithm settings for movement artefact removal from EEG

[Makoto Miyakoshi]

> The segment provided as input for the algorithm was only a 9 [s] long
time-slice. Will longer (~10 min) segments provide better filtering while
using ASR?

Of course! 9 sec long is apparently suboptimal. Provide all data you have
for the best result.

Makoto

On Sat, Oct 10, 2015 at 11:20 PM, Evyatar.Arad at sheba.health.gov.il <
Evyatar.Arad at sheba.health.gov.il> wrote:

> Dear Makoto,
>
> Good week.
> The segment provided as input for the algorithm was only a 9 [s] long
> time-slice. Will longer (~10 min) segments provide better filtering while
> using ASR?
>
> All the best,
> Evyatar
>
> On Oct 8, 2015, at 8:17 PM, Makoto Miyakoshi <mmiyakoshi at ucsd.edu> wrote:
>
> Oh that's true. Are you sure that you provided full-length continuous data?
>
> On Wed, Oct 7, 2015 at 11:30 PM, Evyatar.Arad at sheba.health.gov.il <
> Evyatar.Arad at sheba.health.gov.il> wrote:
>
>> Makoto,
>>
>> It too a few (<10) seconds.
>> The entire segment cleaned is 9 [s] long.
>>
>> Best,
>> Evyatar
>>
>> On Oct 8, 2015, at 1:23 AM, Makoto Miyakoshi <mmiyakoshi at ucsd.edu> wrote:
>>
>> Dear Evyatar,
>>
>> I don't see anything strange in the log. How long did it take?
>>
>> Makoto
>>
>> On Tue, Oct 6, 2015 at 11:10 PM, Evyatar.Arad at sheba.health.gov.il <
>> Evyatar.Arad at sheba.health.gov.il> wrote:
>>
>>> Scanning for bad channels...
>>> clean_channel:   1/1, 0.0 minutes remaining.
>>> Finding a clean section of the data...
>>> Determining time window rejection thresholds...done.
>>> Keeping 45.7% (5 seconds) of the data.
>>> eeg_insertbound(): 4 boundary (break) events added.
>>> eeg_checkset note: upper time limit (xmax) adjusted so
>>> (xmax-xmin)*srate+1 = number of frames
>>> Estimating calibration statistics; this may take a while...
>>> Determining per-component thresholds...done.
>>> Now doing final post-cleanup of the output.
>>> Determining time window rejection thresholds...done.
>>> Keeping 100.0% (10 seconds) of the data.
>>> Use vis_artifacts to compare the cleaned data to the original.
>>> Done.
>>> Creating a new ALLEEG dataset 3
>>>
>>> On Oct 7, 2015, at 8:59 AM, Evyatar Arad <
>>> evyatar.arad at sheba.health.gov.il> wrote:
>>>
>>> Thanks for the quick reply :)
>>>
>>> I thought so too. Could you help me in locating the log file?
>>>
>>> Regards,
>>> Evyatar
>>>
>>> On Oct 7, 2015, at 1:50 AM, Makoto Miyakoshi <mmiyakoshi at ucsd.edu>
>>> wrote:
>>>
>>> Dear Evyatar,
>>>
>>> Well that's strange. What does the log say (how much data are left in
>>> the initial clean portion selection etc)
>>>
>>> Makoto
>>>
>>> On Sun, Oct 4, 2015 at 5:36 AM, Evyatar.Arad at sheba.health.gov.il <
>>> Evyatar.Arad at sheba.health.gov.il> wrote:
>>>
>>>> Hello Miyakoshi,
>>>>
>>>> My name is Evyatar Arad and I’m a Biomedical engineer currently working
>>>> on movement artefact removal from an 32 electrodes EEG system.
>>>> I have installed the ASR extension to Matlab and I was wondering what
>>>> are the optimal parameters for the ASR algorithm.
>>>>
>>>> My current dataset is composed of a 30 channel EEG singal, 290 [s] in
>>>> length.
>>>> Running the algorithm in default setting yields the following result:
>>>>
>>>> *Pre-ASR, normalised signal:*
>>>> <No Filt.png>
>>>>
>>>> *Post-ASR, normalised signal:*
>>>>
>>>> It seems as though at best no change has occurred.
>>>>
>>>> Your recommendations will be beneficiary.
>>>>
>>>> Thanks in advance,
>>>> Evyatar Arad
>>>> <http://www.sheba.co.il/h/> <https://yashir.sheba.co.il/>
>>>> <http://www.sheba.co.il/sheba_movies> <https://twitter.com/sheba_>
>>>> <https://www.facebook.com/SHEBA.MEDICAL?ref=ts>
>>>>
>>>
>>>
>>>
>>> --
>>> Makoto Miyakoshi
>>> Swartz Center for Computational Neuroscience
>>> Institute for Neural Computation, University of California San Diego
>>>
>>>
>>>
>>>
>>
>>
>> --
>> Makoto Miyakoshi
>> Swartz Center for Computational Neuroscience
>> Institute for Neural Computation, University of California San Diego
>>
>>
>>
>
>
> --
> Makoto Miyakoshi
> Swartz Center for Computational Neuroscience
> Institute for Neural Computation, University of California San Diego
>
>
>


-- 
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20151011/5df32685/attachment.html>