search

lcnbeapp / beapp

The Batch Electroencephalography Automated Processing Platform (BEAPP)
Other
35 stars 23 forks source link

Not able to get "Abs/raw" output from PSD #2

Closed AnnaKirkAU closed 5 years ago

AnnaKirkAU commented 5 years ago

I can get all the other excel outputs (Mean, StD, Median, Log, Log10, NormAbs), except for "Abs/raw". When I tried to run files and only selected the Abs/raw output, I got the error attached.

absolute power output error
mgm248 commented 5 years ago

You have to select at least one of these outputs: Mean, StD, Median. These variables specify how you want to report the other values (Raw, Log, Log10, NormAbs). So if you select only mean and raw, beapp will output a file called mean_Pwr_Per_Hz. Sorry, this is unclear when you're running beapp, i'll try to add some explanation.

AnnaKirkAU commented 5 years ago

Thank you, that makes sense! To be clear, the mean_Pwr_Per_Hz output is the absolute power?

On Tue, Feb 5, 2019 at 3:04 PM mgm248 notifications@github.com wrote:

You have to select at least one of these outputs: Mean, StD, Median. These variables specify how you want to report the other values (Raw, Log, Log10, NormAbs). So if you select only mean and raw, beapp will output a file called mean_Pwr_Per_Hz. Sorry, this is unclear when you're running beapp, i'll try to add some explanation.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/lcnbeapp/beapp/issues/2#issuecomment-460783328, or mute the thread https://github.com/notifications/unsubscribe-auth/ArfnsxniRrpJRzYgDTqBDdP3JtR_Tlxfks5vKeO8gaJpZM4adROt .

-- Anna Kirkland, B.A. Ph.D. Student in Behavior, Cognition, and Neuroscience American University

mgm248 commented 5 years ago

Well, it's per frequency. We take the power spectra, which has the power values at each frequency, and sum the power values at each frequency in the given frequency range. Then, because different frequency bands have different lengths, we divide by the number of frequencies in the given frequency band. So we don't report the raw absolute power, we report the power per frequency (Hz).

AnnaKirkAU commented 5 years ago

Thank you for explaining. I am new at analyzing EEG data, so I have a follow up question: can I calculate the relative power from the power per frequency output?

On Tue, Feb 5, 2019 at 4:54 PM mgm248 notifications@github.com wrote:

Well, it's per frequency. We take the power spectra, which has the power values at each frequency, and sum the power values at each frequency in the given frequency range. Then, because different frequency bands have different lengths, we divide by the number of frequencies in the given frequency band. So we don't report the raw absolute power, we report the power per frequency (Hz).

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/lcnbeapp/beapp/issues/2#issuecomment-460819004, or mute the thread https://github.com/notifications/unsubscribe-auth/ArfnszMHbVZnoaI0R8TpEEcxcuFUv7FHks5vKf2ogaJpZM4adROt .

-- Anna Kirkland, B.A. Ph.D. Student in Behavior, Cognition, and Neuroscience American University

mgm248 commented 5 years ago

Actually, our normalized power outputs (mean norm, SD norm, etc) are relative power. For the norm outputs, beapp takes the absolute power in the frequency band (NOT per Hz) and divides it by the absolute total power.

AnnaKirkAU commented 5 years ago

That helps a lot, thank you!

Sent from my iPhone

On Feb 6, 2019, at 8:56 AM, mgm248 notifications@github.com wrote:

Actually, our normalized power outputs (mean norm, SD norm, etc) are relative power. For the norm outputs, beapp takes the absolute power in the frequency band (NOT per Hz) and divides it by the absolute total power.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.