EEGLAB function pop_clean_rawdata[ LineNoiseCriterion ]

Hi, I would like to know how the LineNoiseCriterion works. I was told it works by averaging the channels and then using a pre-set value (such as 4 in my code), it uses this value as a threshold to reject channels that go below -4 or above 4. Is this correct?

[Dani]

to add, is this function working on all 128 channels or is it working with 1 channel to do this

[Walter Roberson]

https://github.com/sccn/clean_rawdata/blob/master/pop_clean_rawdata.m has the source code

Answer 1

[Garmit Pant]

Hello Dani,

I understand that you want to understand how the input argument “LineNoiseaCriterion” of the EEGLAB function “pop_clean_rawdata” works and how many channels it is working on.

To address your queries:

• “LineNoiseCriterion” sets the threshold for the channels to be removed based on the noise relative to their signal value. The default value is 4. By default, any channel that has noise relative to its signal above 4 standard deviations, that channel will be removed.
• The channel is removed based on the z-scored level of EM noise-to-signal ratio for each channel. The score is calculated using the standard deviation based on all channels’ signal.
• The function works on all channels.

For further understanding on “pop_clean_rawdata”, you can refer to the following external links:

1. Refer to ‘Reject bad channels’ section - https://github.com/sccn/clean_rawdata

I hope you find the above explanation and suggestions useful!