Hi gurus, I am evaluating two segmentation routines relative to a gold standard and would like to calculate the Hausdorff distance for each routine relative to the gold standard in the left and right Amygdala, Caudate, Putamen, and Hippocampus. I know that there are 3 main options to the mri_hausdorff_dist binary, namely -b <thresh> (binarize input volumes with threshold <thresh>), -g <sigma> (blur the input image with Gaussian <sigma>) and -max (compute the max of the min distances instead of the mean) but I am not quite sure how to use them to get the desired result. Does anyone know how best to use this tool to get structure-specific Hausdorff distances from 2 aseg.mgz volumes?
use mri_extract_label to copy the label you want out, then run mri_hausdorff distance on the binary volumes. Or I just added a -l <label index> flag to mri_hausdorff distance so you that you can specify a single label (e.g. 17 for left hippo). If you go this route you'll need to get a new version.
cheers, Bruce
On Fri, 17 Jul 2009, Jared Price wrote:
Hi gurus, I am evaluating two segmentation routines relative to a gold standard and would like to calculate the Hausdorff distance for each routine relative to the gold standard in the left and right Amygdala, Caudate, Putamen, and Hippocampus. I know that there are 3 main options to the mri_hausdorff_dist binary, namely -b <thresh> (binarize input volumes with threshold <thresh>), -g <sigma> (blur the input image with Gaussian <sigma>) and -max (compute the max of the min distances instead of the mean) but I am not quite sure how to use them to get the desired result. Does anyone know how best to use this tool to get structure-specific Hausdorff distances from 2 aseg.mgz volumes? _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu