[MINC-users] Classify
Jon Erik Ween
jween at klaru-baycrest.on.ca
Wed Apr 1 10:55:03 EDT 2009
Dear list
Sorry about the pics. They can be found at: http://www.klaru-baycrest.on.ca/index.php?section=969
I figured out that the mask option in classify specifies which voxels
you WANT to classify (for us in the lesion world, we generally think
of a mask for those things you DON'T want to include, ie: The stuff
that will mess up the algorithm). So, the 3DT1 volume I want to
classify (with lesion mask overlaid and in tal space) is given in
Picture1. (You'll notice that the registration is not great, I get
some of the lesion in the ventricle! Any hints as to how to optimize
the registration with mritotal would be appreciated!).
Running classify -clobber -user_mask_value 0 -tagfile tags.tag -mask
14017_083005_combo_lbl2_tal.mnc 14017_083005_3dvol_tal.mnc
14017_083005_3dvol_cls.mnc
gives me the result you see in Picture2. Lesioned whitematter is
classified as grey matter and grey matter seems to be way over
represented.
So, I figured out one way to go around this: Subtract the lesion mask
from the volume BEORE running classify. The result of this is in
Picture3. Here, we don't have the problem with lesion being classified
as grey matter, but normal grey still seems over represented.
Questions:
1) How to optimize the classification so the grey matter is more
representative of that seen visually on the 3DT1 (particularly the
cortical ribbon)?
2) When setting tag points in your tag file to feed classify, setting
paint label value in "Display" to "0" (for background) you don't seem
to record anything. One work-around I did was to set the value to "4"
and then edit the tagfile back to "0". Is this normal behavior or am I
doing something wrong?
3) Saving the tagfile in "Display" gives you a bit different format
(which classify doesn't like) compared to the tagfile in ../mni/share/
classify/ntags_1000_bg.tag
Display "save labels .tag": -82 -109 108 1 1 1
/mni/share/classify/ntags_1000_bg.tag: -82 -109 108 "1"
Am I using the wrong function here?
4) Once I get a classified volume (and I have flair lesion and 3DT1
lesion in separate masks) how do I calculate volumes? I can't seem to
find the functions that do this, though suspect they are in minmath or
minccalc somewhere.
Thanks for all the help!
Cheers
Jon
Soli Deo Gloria
Jon Erik Ween, MD, MS
Scientist, Kunin-Lunenfeld Applied Research Unit
Director, Stroke Clinic, Brain Health Clinic, Baycrest Centre
Assistant Professor, Dept. of Medicine, Div. of Neurology
University of Toronto Faculty of Medicine
Kimel Family Building, 6th Floor, Room 644
Baycrest Centre
3560 Bathurst Street
Toronto, Ontario M6A 2E1
Canada
Phone: 416-785-2500 x3648
Fax: 416-785-2484
Email: jween at klaru-baycrest.on.ca
Confidential: This communication and any attachment(s) may contain
confidential or privileged information and is intended solely for the
address(es) or the entity representing the recipient(s). If you have
received this information in error, you are hereby advised to destroy
the document and any attachment(s), make no copies of same and inform
the sender immediately of the error. Any unauthorized use or
disclosure of this information is strictly prohibited.
On 1-Apr-09, at 7:45 AM, Andrew Janke wrote:
> 2009/3/31 Jon Erik Ween <jween at klaru-baycrest.on.ca>:
>> May I trouble the list with another question? (Guess I'll do so
>> anyhow...)
>
> Hi Jon,
>
> No images... (the list wont accept them) Can you put them up
> somewhere on the web instead?
>
>
> --
> Andrew Janke
> (a.janke at gmail.com || http://a.janke.googlepages.com/)
> Canberra->Australia +61 (402) 700 883
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www2.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>
More information about the MINC-users
mailing list