Viz problem

[arnodelorme]

[from Scott]

I'm looking at EEG VIZ for ds004574/sub-018/eeg from https://nemar.org/dataexplorer/detail?dataset_id=ds004574

The IC activations scrolling plot clearly shows IC32 collects all the Line Noise The IC activations plot purports to show IC13-IC47 -- it should show IC01-IC35. However, the IC topoplot shows the Line Noise IC as #29 -- not #32. Why??

Here's the ZIX screenshot. Note that IC29 is Line Noise per ICLabel, but in the IC scrolling plot the line noise IC is clearly IC32...

[arnodelorme]

Another example

This is ds004502/sub-007/eeg Note: Here ICA decomp was run on 7M frames! (k = 1781) ... gives a 'good' decomposition. BUT the Line Noise IC is shown as IC60 -- but in the topolot it is #34 (bottom, next to right). So it appears that the ICs shown may also be IC27-IC61 ?? I don't think so, as IC1 is eyeblinks, as expected... So the IC #s in the IC activations plot are probably wrong. (Here, IC27-61 --> IC1-35 ??)

[arnodelorme]

Interesting indeed. @asalazar4 can you check the ds004502/sub-007/eeg subject and component IC60?

[arnodelorme]

@asalazar4 did you get a chance to look at it? Tx

[asalazar4]

@arnodelorme The line noise in both ds004502/sub-007/eeg and ds004574/sub-018/eeg EEG vis, seem to be correct. For ds004502/sub-007/eeg the line noise is displayed in both visuals as being line 34, and ds004574/sub-18/eeg is visualized as line 20.

[dungscout96]

close for inactivity