Strange behavior of mne.preprocessing.current source density

✉️ Mailing List Archive (read-only)
1

External Email - Use Caution

Hi,
The current source density implementation in mne.python seems to compute incorrect CSD estimates. The attached file illustrates the issue. The first row shows topoplots of monopolar data (average reference). The second row shows topoplots of CSD estimates of these data, computed using the mne.preprocessing.courrent_cource_density function using default parameters (lambda2=1e-05, stiffness=4, n_legendre_terms=50). the third row shows topoplots of CSD estimates of these data, computed using Mike Cohen's code and similar parameters ( lambda2=1e-05, stiffness=4, n_legendre_terms=50).
Do you have any idea of what's going on? Note that for rows 2 and 3, I used the 'biosemi64' electrode coordinates provided by mne.

Best,
Mat

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20200602/ff600ad0/attachment-0001.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Results.png
Type: image/png
Size: 3622881 bytes
Desc: not available
Url : http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20200602/ff600ad0/attachment-0001.png

2

External Email - Use Caution

hi Mat,

please open an issue here https://github.com/mne-tools/mne-python/issues
with enough information so we can reproduce.

I would start here by testing that the matlab code gives different results
on the data from this example:

https://mne.tools/dev/auto_examples/preprocessing/plot_eeg_csd.html
<https://mne.tools/dev/auto_examples/preprocessing/plot_eeg_csd.html#sphx-glr-auto-examples-preprocessing-plot-eeg-csd-py>

Alex