[Eeglablist] Implementing Cleanline

[周云晖]

Hi,

You have band-pass filtered your data to 1-40 Hz. Why do you intend to do Cleanline again? From your screenshot there is no 60 Hz noise in your data (because it has already been filtered out).

The help_vrect error should be due to recent MATLAB's upgrade of graphics system. In old version of Matlab, graphics object handles are some random double scalar, but now it is not. This cause some compatibility issues in old Matlab code.

At2018-10-25 00:13:06，Katie Hoemannkhoemann at gmail.comwrote:

Hi all,



We are following Makoto's preprocessing pipeline and are having some issues with the Cleanliness plugin. This is a straightforward analysis so we're not sure what's off; any help would be appreciated.


We have a 32-channel ActiCap system from BrainVision and are recording in PyCorder. There are no online filters applied as far as we are aware. The data are imported via the bva-io plugin (or BIOSIG) and bandpass filtered at .1 to 40 Hz. We are in the US and looking to clean for 60 Hz line noise.


When we run Cleanline with the default settings (or even Makoto's recommended window size == step size), we do not see any original data in the plot. We only see red "cleaned" data with no indication if anything has been removed. See attached. We're finding it hard to believe we have no line noise in the data.


If we try to inspect the plots for a single channel (e.g., 14, Cz), we receive the following error:
"help_vrect error: argument 'axesHandle' must be numeric' (error occurs in function hlp_vrect() at line 16).




Thanks,
Katie Hoemann












-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20181026/67fbceda/attachment.html>