Hello all,
I’m really a novice in the field and just started to learn to use the EEGlab. I’m writing because I’d like to run a power/frequency analysis on the data I have – and I’ve got a question.
The data was collected from 9-month-old infants using the EGI system, HydroCel Geodesic Sensor Net with 128 channels. We’re interested in brain activities occurring during a 5-sec long interval. Each infant did 7 to 9 trials (i.e. each infant gave us 7-9 intervals to analyse), which seems very small as a number of trials but it’s not very unusual for infant EEG studies + the study is a pilot.
I’ve been using a NetStation to pre-process EEG so I thought it’d be easier to segment the data, do the artefact detection and replace bad channels identified, and then import these epoched data into the EEG lab to do the FFT (or other transformation methods for the power analysis). But I was advised a few times to use the EEG lab from the beginning of the preprocessing (i.e. epoching, artefact detection and interpolation). Is there any difference between these two? If so, what is the difference and what may be the pros and cons of each method?
I hope this makes sense – but please get in touch if you can help me with this and/or need more information.
Thank you very much in advance, and I hope you have a lovely day.
Kind regards,
Saya
*********
Sayaka Fujita
Marie Skłodowska-Curie Early Career Researcher (MOTION)
Department of Psychology
Lancaster University
Hello Sayaka!! Yoroshiku :)
First of all good luck and best wishes for the begining of your endeavour with EEG devices and biomedical signal processing in general!
To get to the point.... Most people tell you to do everything straight in EEGlab because it is considered a very well organized plugin with many tools available to help you do anyy eeg analysis you desire...
You could do everything first on NetStation(i dont know it but i guess..) of cource and then pass everything to eeg to resume your process but you migh come across some connectivity issues considering the anotations you recorded during EEG recording. EEGlab ofcource offers solutions to these problems and ofcource any post on the forum will be answered but the main point is saving time. Its better to have to deal with one platform instead of multiple. I'll be glad to help with anything i can. Just to inform you, I have worked with eeg devices and analysis for 3 projects so far... I know i am considered a novice too but any help i could provide please dont hesitate to ask :)
Panos Bonotis Thank you very much for your answer, and how nice to see you using Japanese :) I'm currently using the EGI system which I'm more familiar with but will surely try to learn the EEG lab as well in the near future. I've been reading and watching written and video tutorials and it doesn't seem to be unmanageable.
And thank you so much for your kind offer to help, I appreciate it!
Hi Sayaka,
I would recommend importing the raw EEG data into EEGLAB first and then reject the data manually or by using the automatic rejection method provided in the EEGLAB.
It would be good if you plot the raw data and just scroll across it, looking for the bad channels or noisy segments. This toolbox has already great features specially designed for tackling EEG data, so I would prefer using it for pre-processing and analysis steps.
Also, try to do most pre-processing steps before segmenting the data into epochs, especially filtering.
Hope it helps..!!
Best Wishes,
Riti
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS