<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META http-equiv=Content-Type content="text/html; charset=iso-8859-1"> <META content="MSHTML 6.00.5700.6" name=GENERATOR> <STYLE></STYLE> </HEAD> <BODY bgColor=#ffffff> <DIV><FONT face=Arial size=2> <DIV><FONT face=Arial size=2>Dear all,</FONT></DIV> <DIV><FONT face=Arial size=2></FONT> </DIV> <DIV><FONT face=Arial size=2>I am trying to simulate NEC curves with GATE but cannot achieve good agreement with measurements. </FONT><FONT face=Arial size=2>The reason for that seems to be that <U>% losses of trues, scatters and randoms coincidences due to dead-time are not the same in measurements</U>, whereas the standard techniques of dead-time simulation with para and non-para models give exactly the same % dead-time for trues, scatters and randoms (easy to prove).</FONT></DIV> <DIV><FONT face=Arial size=2></FONT> </DIV> <DIV><FONT face=Arial size=2>I have seen number of GATE papers addressing the challenge of modeling PET scanners at high count rates, so does anyone could share his/her expertise on how to model realistic count rates at high activity? I would be in particular interrested in knowing how to model different % dead-time losses for trues, scatters and randoms with GATE.</FONT></DIV> <DIV><FONT face=Arial size=2></FONT> </DIV> <DIV><FONT face=Arial size=2>Many thanks,</FONT></DIV> <DIV><FONT face=Arial size=2></FONT> </DIV> <DIV><FONT face=Arial size=2>Bastien G.<BR>-------------------------------<BR>Bastien P Guerin<BR>PhD candidate, University of Paris VI<BR>Visiting Fellow, Harvard University<BR>t: 617 817 2762<BR>f: 617 732 6695</FONT></DIV></FONT></DIV></BODY></HTML>