<div dir="ltr"><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">Dear all,</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">I am stepping in some problems when running ICA decomposition for artifact detection.</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">My dataset consists in a continous 38 channels 70 minutes long recording, sampled at 500 Hz referenced to Cz.</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">My preprocessing steps are the following:</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">- Re-reference to the average reference and adding Cz to the recording</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">-Filter with a band-pass filter set at 1 - 100 Hz</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">- Visual inspection of the recording and removal of big noisy artifacts, mainly movement artifacts, as suggested in the EEGLAB tutorials</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">After these steps my dataset consists of 1864585 data points on which I perform <i>runica</i> with the default options ('extended', '1').</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">I noticed that the process is very slow, and the algorithm needs to lowering the learning rate many times at the beginning but even so it seems that it fails to converge, since the wchange values does not decrease progressively (as they should) and it fails to reach the stop criterium (wchange <1e-07).</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">As a consequence I get a bad decomposition with uninterpretable components that prevent their use for artifact correction.</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">I am wondering if this problem is related to the amount of data points fed to the ICA, since when I preprocessed shorter recordings I have not encountered such difficulties, or I am making some mistakes during my pipeline.</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">A great thank to anyone who can help me.</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">Antonio</div>
</div>