[FieldTrip] relative power estimation

Mon Sep 14 17:25:00 CEST 2020

Hi Marta,

Thanks for the clarification. ft_freqbaseline is what you are looking for
then.
Unless I am missing something, averaging then baseline correction should be
(mathematically) equivalent to the other way around, but if it was me I
would just compare them once to ease my mind :-)

It behooves me to mention the alternative, namely using slepian tapers
(tapers = 'dpss') in freqanalysis to extract a more clearly defined
frequency smoothing around the frequency bands of interest, i.e. specifying
center frequency and frequency-smoothing.

cfg.taper       = 'dpss';         % Multitapers based on Slepian sequences
cfg.tapsmofrq   = 2;              % Smoothing +/- 2 Hz
cfg.foi         = 10;

The frequency analysis tutorials on the wiki page go into some depth on
this topic - as well as how to do baseline correction -, as well as most if
not all of the FieldTrip youtube videos, which are very consistent and very
good.

Happy fieldtripping,
Stephen

Op ma 14 sep. 2020 om 17:03 schreef Marta Andujar <marta.andujar at uniroma1.it
>:

> Hi  Stephen,
> thank you for the quick reply, having a data structure which is baseline
> corrected is exactly what I want.
> The other question concerns the frequency-band specific
> baseline-correction. As a first step I computed TFR representation of the
> entire frequency spectrum (i.e. from 8 HZ to 180 HZ), since I am interested
> in inspecting the data in this spectrum (In this case applying the
> baseline-correction just before plotting is ok). From this data structure
> I subsequently extract different frequency-bands by cutting the data
> (TFRwavedata) at different frequencies (e.g 12 Hz, 30 Hz and so on) in
> order to obtain four different matrices of interest each one representing a
> band-specific power spectrum. Finally, I average the power over the
> frequencies of each band in order to obtain the estimation of the power in
> that band. In the latter case, according to my understanding, I should
> apply the baseline-correction after computing the average power of each
> frequency-band, so that the relative "band specific power" in each time bin
> is the ratio between the power averaged over all the frequencies belonging
> to that band and the power averaged over all the same frequencies in the
> baseline period. I would like to know if the procedure is correct and how
> to implement it in fieldtrip. I am going to check ft_freqbaseline as you
> kindly suggested me.
> Thank for the help
> Best
> Marta
>
> Il giorno lun 14 set 2020 alle ore 16:00 Stephen Whitmarsh <
> stephen.whitmarsh at gmail.com> ha scritto:
>
>> Hi Marta,
>>
>> It's not clear from your question whether you do or do not want to
>> baseline correct power before plotting - with which I assume you mean not
>> doing it in the plotting functions but instead creating a data structure
>> which has had baseline correction applied. If so, ft_freqbaseline is what
>> you are looking for. If not, then I am afraid I do not understand that part
>> of the question.
>> Regarding the question whether or not you should apply a baseline
>> correction, that would depend on what you want to do. "My final goal is to
>> extract different frequency-bands" is not specific enough to really give an
>> informed response to that.
>>
>> Best,
>> Stephen
>>
>> Op ma 14 sep. 2020 om 15:32 schreef Marta Andujar <
>> marta.andujar at uniroma1.it>:
>>
>>> Dear all,
>>> I am performing a time-frequency decomposition via Morlet's Wavelet. My
>>> question is very basic and it concerns the estimation of the power. I
>>> went through different tutorials, and I am a bit confused about the
>>> normalization. In every tutorial I saw, the normalization is implemented
>>> before plotting the results.I am using the 'relative' option, so that for
>>> each frequency of interest and for each time-bin, the power is expressed as
>>> the ratio between the raw power and the power averaged over the baseline
>>> period. However, I would like to have my TRF results expressed as relative
>>> power before plotting. Is there a way to get what I am looking for? Or
>>> should I write my own code in order to get the relative power? My final
>>> goal is to extract different frequency-bands, and I guess that for each of
>>> these I should normalize against the power averaged both over the baseline
>>> period and over the frequencies defining the band of interest. Is it
>>> correct?
>>> Thank you in advance
>>> Marta
>>>
>>> ________________________________________________________
>>> Le informazioni contenute in questo messaggio di posta elettronica sono
>>> strettamente riservate e indirizzate esclusivamente al destinatario. Si
>>> prega di non leggere, fare copia, inoltrare a terzi o conservare tale
>>> messaggio se non si è il legittimo destinatario dello stesso. Qualora tale
>>> messaggio sia stato ricevuto per errore, si prega di restituirlo al
>>> mittente e di cancellarlo permanentemente dal proprio computer.
>>> The information contained in this e mail message is strictly
>>> confidential and intended for the use of the addressee only.  If you are
>>> not the intended recipient, please do not read, copy, forward or store it
>>> on your computer. If you have received the message in error, please forward
>>> it back to the sender and delete it permanently from your computer system.
>>> ------------------------------
>>>
>>> _______________________________________________
>>> fieldtrip mailing list
>>> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>> https://doi.org/10.1371/journal.pcbi.1002202
>>>
>> _______________________________________________
>> fieldtrip mailing list
>> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>> https://doi.org/10.1371/journal.pcbi.1002202
>>
>
> ________________________________________________________
> Le informazioni contenute in questo messaggio di posta elettronica sono
> strettamente riservate e indirizzate esclusivamente al destinatario. Si
> prega di non leggere, fare copia, inoltrare a terzi o conservare tale
> messaggio se non si è il legittimo destinatario dello stesso. Qualora tale
> messaggio sia stato ricevuto per errore, si prega di restituirlo al
> mittente e di cancellarlo permanentemente dal proprio computer.
> The information contained in this e mail message is strictly confidential
> and intended for the use of the addressee only.  If you are not the
> intended recipient, please do not read, copy, forward or store it on your
> computer. If you have received the message in error, please forward it back
> to the sender and delete it permanently from your computer system.
> ------------------------------
>
> _______________________________________________
> fieldtrip mailing list
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> https://doi.org/10.1371/journal.pcbi.1002202
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20200914/06463c39/attachment.htm>