Hi Mohan,
The short answer is that you can not use the instructions you mention for the hippocampal subfields.
The long answer is that the instructions only work for segmentations that are already crisp, such as the ones contained in the aseg.mgz file: each intensity level in aseg.mgz corresponds to a certain structure, and voxels having that intensity are assigned exclusively to the structure. The hippocampal subfield segmentation results are stored in a different way that tells you something about the uncertainty with which such assigned can be made, so you'll need to make those exclusive assignments to a single structure manually...
Hope this helps,
Koen
On Thu, Aug 11, 2011 at 10:06 AM, mohan boddu bmohan.boddu@gmail.com wrote:
---------- Forwarded message ---------- From: mohan boddu bmohan.boddu@gmail.com Date: Tue, Aug 9, 2011 at 6:17 AM Subject: Re: [Freesurfer] Segmentation of hippocampal subfilelds To: Koen Van Leemput koen@nmr.mgh.harvard.edu
Hello,
I would like to thank you for the information, but the problem is, I am a CS student working on this and I dont have much idea about it. Basically my task is to convert those images into ROI mask. For this I have to follow the instructions in the website stated below:
http://brainybehavior.com/neuroimaging/2010/06/saving-individual-labels-from...
But the process stated in the above website works for label maps, and if the images generated in the "segmentation of hippocampal subfields" (http://www.freesurfer.net/fswiki/HippocampalSubfieldSegmentation) (8 images are generated for each side) stage are probability maps then what should be given as upper threshold(-uthr) and lower threshold(-thr) values. And as the last step says to divide it by its label number to normalize it. Is it necessary for the probability maps, if it is, then what would be the number that we have to divide?
Sorry, if I am not clear and thank you very much for helping me out.
On Tue, Aug 9, 2011 at 10:41 AM, Koen Van Leemput koen@nmr.mgh.harvard.edu wrote:
Hi Mohan,
The segmentation results are given as probability maps: a (unsigned 8-bit) value of 0 means probability 0 , 255 means probability 1, and everything in between scales linearly.
The proper way to obtain "crisp" label maps from these images is to assign each voxel to the label map that has the highest probability, so you'd need to read in all probabilty maps, visit each voxel and compare the probabilities in the voxel for all structures.
Generating such "scrisp", winner-takes-all label maps is not recommended to calculate e.g., volumes etc, though...
Hope this helps,
Koen
On Mon, Aug 8, 2011 at 2:26 PM, mohan boddu bmohan.boddu@gmail.com wrote:
I have completed the segmentation of hippocampal subfileds, but now I would like to know the generated images are in probability map or label map? If they are in label map what would be the best upper and lower thresholds to generate the ROI mask images. Could you please also tell me the way to do so.
-- --Mohan Boddu-- Research Assistant Department of Radiology and Imaging Research II Building Indiana University School Of Medicine Indiana University-Purdue University Indianapolis Indianapolis, IN 46202
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
-- --Mohan Boddu-- Research Assistant Department of Radiology and Imaging Research II Building Indiana University School Of Medicine Indiana University-Purdue University Indianapolis Indianapolis, IN 46202
-- --Mohan Boddu-- Research Assistant Department of Radiology and Imaging Research II Building Indiana University School Of Medicine Indiana University-Purdue University Indianapolis Indianapolis, IN 46202
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.