<html><head><style type="text/css"><!-- DIV {margin:0px;} --></style></head><body><div style="font-family:times new roman,new york,times,serif;font-size:10pt"><div>Hi all,<br><br>I'm working on an automatic landmark localisation algorithm in MR images. Im trying to<br>locate the nasion (saddle point between the eyes) and two points on the ears.<br>My algorithm is based on registration to an atlas. First I manually locate the landmark<br>and then I register to the atlas (icbm_152). I use two images to define the landmark in the atlas and<br>use the third image to measure the accuracy of the algorithm by applying the inverse transformation<br> to the landmark in the atlas.<br>I've tried a two step registration (rigid + affine) and the euclidean distance between manual and automatic<br>localization is about 10mm. <br>Currently I'm trying the BSplineDeformableTransform but the results get worse to about 20mm. I use
the<br>InverseDeformationFieldImageFilter and everything seems to be working fine but I don't understand why<br>the result get worse. I've tried several setting but am at loss at the moment.<br>Anybody any tips for fine-tuning a deformable transform? The algorithm takes a long time, especially inverting<br>the deformation field (for data info see below). How can I speed up the algorithm? Any help or ideas will be appreciated.<br>regards,<br>Lars<br><br>fixed<br>ObjectType = Image<br>NDims = 3<br>BinaryData = True<br>BinaryDataByteOrderMSB = True<br>ElementSpacing = 1 1 1<br>DimSize = 181 217 181<br>Position = -90 -126 -72<br>ElementNumberOfChannels = 1<br>ElementType = MET_UCHAR<br>ElementDataFile = icbm_avg_152_t1_tal_lin.raw<br><br>moving<br>ObjectType = Image<br>NDims = 3<br>BinaryData = True<br>BinaryDataByteOrderMSB = True<br>ElementSpacing = 1 1 1<br>DimSize = 256 256 180<br>Position = -127 -112.92 -82.4405<br>ElementNumberOfChannels =
1<br>ElementType = MET_UCHAR<br>ElementDataFile = n474_mra.raw<br></div></div><br>__________________________________________________<br>Bruker du Yahoo!?<br>Lei av spam? Yahoo! Mail har den beste spambeskyttelsen <br>http://no.mail.yahoo.com </body></html>