Good morning,
I'm having trouble normalizing my signals. I would like them all to be on the same baseline but can't do it.
Thank you for taking the time to answer this question.
Arthur,
Can you please explain what you mean by "normalizing" and "same "baseline"?
Please include screen captures showing the files in the database explorer, the contents of the files and the options of the process you're trying to use
Good morning,
Sorry for the delay in answering.
I would like to have the same baseline (at 0 for all the signals) for the different signals corresponding to different conditions (screenshot), to be able to observe the differences of the signals afterwards.
I'm currently trying to apply baseline normalization to each signal, but the signals have different baselines because they are calculated using different values, resulting in varying means…
Thank you for your time,
Arthur
I'm sorry, we can't help you understand what you're doing wrong with the software if you don't explain precisely what you are doing with it.
Please include screen captures showing the files in the database explorer, the contents of the files and the options of the process you're trying to use.
Good morning,
I have different signal records representing different search conditions. These signals vary differently.
I have done different signal processing, resample, band pass run ICA to get the average of the recorded signals according to the conditions. The problem is that the signals are therefore not normalized on the same basis of amplitude making the interpretation complicated or even impossible because the basis of comparison of signals is not the same.
I tried to take the signals and to normalize it by adding a base line of normalization of DC offset present in the tab of standardization but the base lines are carried out according to the average of each signal, not giving the same base line according to each signal.
I would like to have the same base line for all the signals, allowing a more correct interpretation of the signals.
Signal Treatment.
Normalization Process
Example of the results, were the signals are not really normalizing…
Thank you for your answers,
Arthur.
See this publication:
EEG is better left alone
The problem is that the signals are therefore not normalized on the same basis of amplitude making the interpretation complicated or even impossible because the basis of comparison of signals is not the same.
What is still missing here is the key information: what options do you use in this "Baseline normalization" process. From the look of your signals, I guess you used a baseline time window that includes all the epochs instead of only the baseline.
With the information you shared, I want to bring your attention to some possible misunderstanding.
The term "normalization" tends to refer to a multiplication factor applied to the signals, e.g. divinding all the signals by their maximum to get values in a normalized range between -1 and 1.
It seems that what you are expecting here is a "baseline correction" or "DC offset removal" instead, i.e. subtract the mean amplitude over a baseline from all the signal.
This procedure is explained in the introduction tutorials and the EEG tutorial, which I recommend you follow before trying to process your own recordings:
- https://neuroimage.usc.edu/brainstorm/Tutorials/Epoching#Import_in_database
- https://neuroimage.usc.edu/brainstorm/Tutorials
Note that if you already applied a band-pass filter, the "DC offset" or "slow fluctuations of the signals" are already removed from your signals, and maybe you don't need to apply any additional baseline correction: