Hi,<br><br>Using mri_glmfit, I found significant between-group difference in cortical thickness in some regions. However, there are a lot regions with very small vertices surviving the given threshold. What I need to do is displaying large clusters (e.g. having 20 vertices at least) assumed as &quot;true&quot; difference. I don&#39;t know how to make it by a selected threshold of cluster size . Any suggestions would be appreciated.<br>
<br>Liang<br clear="all"><br>-- <br>Liang Wang, PhD<br>Postdoctoral Fellow<br>Woodward Lab<br>Department of Psychiatry<br>University of British Columbia<br>BC Mental Health &amp; Addiction Services<br>938 West 28th Avenue<br>
Vancouver BC V5Z 4H4<br>Telephone: 1-604-875-2000 (ext. 4735)<br>Fax: 1-604-875-3871<br>Email: <a href="mailto:wanglbit@gmail.com">wanglbit@gmail.com</a><br>