<table cellspacing="0" cellpadding="0" border="0" ><tr><td valign="top" style="font: inherit;">Dear Panagiotis,<br>There is a simple way for automated detection of artifactual components.<br>Firstly, you can calculate the correlation of each component with your ECG, EOG channel.<br>Secondly, you can employ high-order statistics applied to the time course of indepedent components like kurtosis (see Nadia Mammone, Francesco Carlo Morabito "Independent Component Analysis and High-Order Statistics for Automatic Artifact Rejection").<br><br>Another practical tip is to use the following criterion in order to get an optimal number of ICAs.To find <i>N</i>stable components (from N-channel data) typically requires <i>more than</i> <i>kN^2</i>
data sample points (at each channel), where N^2 is the number of
weights in the unmixing matrix that ICA is trying to learn and <i>k</i> is a multiplier. A practical criterion is points, k/N^2 > 30 pts/weight
points. So it is better to apply first PCA in order to meet the above requirements.<br>Finally, it is better to follow the above unsupervised technique as a guide and not as a automatic procedure.<br><br>Best regards<br>Dimitriadis Stavros<br>PhD candidate in NeuroInformatics, Dept. of Physics, University of Patras<br><br><p class="style3"><a href="http://www.ellab.physics.upatras.gr/content/view/198/191/lang,english/">1)Electronics Laboratory</a><a href="http://www.ellab.physics.upatras.gr/index.php?lang=english">, </a><a href="http://www.physics.upatras.gr/">Department of Physics</a><a href="http://www.ellab.physics.upatras.gr/index.php?lang=english">, </a><a href="http://www.upatras.gr/">University of Patras</a></p>
<p class="style3"><span class="style10"><a href="http://poseidon.csd.auth.gr/EN/">2)Artificial Intelligence Information Analysis lab </a></span><a href="http://www.csd.auth.gr/">Department of Informatics</a></p>
<span class="style3"><u><a href="http://www.auth.gr/home/">Aristotle University of Thessaloniki</a><br><br>http://users.auth.gr/stdimitr/<br></u></span><br><br><br><br></td></tr></table>