Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Dev #1

Closed
wants to merge 329 commits into from
Closed

Dev #1

wants to merge 329 commits into from

Conversation

zkaufman
Copy link
Member

No description provided.

Z K added 30 commits August 12, 2016 14:17
    /autofs/cluster/freesurfer/subjects/atlases/DKTatlas
    to the atlas.acfb40.noaparc.i12.2016-08-02 tif used to
    define fsaverage for V6. See reg-to-v6 in that folder
    -greve
    /autofs/cluster/freesurfer/subjects/atlases/christophe_atlas
    to the atlas.acfb40.noaparc.i12.2016-08-02 tif used to
    define fsaverage for V6. See reg-to-v6 in that folder
     -greve
…generate

the ?h.aparc.annot. This atlas was generated using the registration to
folding.atlas.acfb40.noaparc.i12.2016-08-02.tif (which is what sphere.reg is
registered to in V6). This makes it consistent.

    cd /autofs/cluster/freesurfer/subjects/atlases/aparc_atlas.V6
    setenv SUBJECTS_DIR /autofs/cluster/freesurfer/subjects/atlases/aparc_atlas.V6
    mris_ca_train -t
    /average/colortable_desikan_killiany.txt
    lh avgsubj.acfb40.noaparc.i12.sphere.reg aparc_edited 15_626 vc1265
    vc604 16_vc660 19_vc681 20_vc700 21_vc716 vc722 23_vc740 vc747 vc764
    31_vc783 18_vc792 vc799 32_vc809 33_vc876 35_vc891 vc922 34_vc1024
    vc1172 vc1249 vc1289 vc1337 vc1379 24_vc1401 25_vc1420 vc1423
    27_vc1425 vc1440 vc1456 vc1463 vc1465 vc1479 vc763 vc6126 vc1493 vc852
    vc1474 vc803 lh.DKaparc.atlas.acfb40.noaparc.i12.2016-08-02.gcs -greve
…009+aseg.mgz by adding --a2009s to the command line for this output.
…use aseg (allows backwards compatibility with 5.3 cross dirs). Added long-hippocampal-subfields help. -mreuter -zkaufman
…bsequent commands were changed to access this file. The ?h.white surface is now the result of the second call to mris_make_surfaces. -greve
…isting segmentation useing a mask. -fischl
…pful for tracking down the source of mislabelings. -fischl
@zkaufman zkaufman closed this Feb 24, 2017
@inobrevi inobrevi mentioned this pull request Feb 13, 2019
htregidgo added a commit to htregidgo/freesurfer that referenced this pull request Nov 30, 2021
htregidgo added a commit to htregidgo/freesurfer that referenced this pull request Nov 30, 2021
Adds the flag --photo-seg to the run_samseg python script. The flag
currently just disables the affine template registration as the photo
volumes are missing the cerebelum and require separate registration. As
the registration is disabled the photo-seg flag requires the filepath
of a registered template as input.

The plan is now to change the behaviour of the bias field correction
when the photo-seg flag is active. This requires changing the bias
field basis function generation and handling the downsampling so that
there is no-downsampling in the z direction.
htregidgo added a commit to htregidgo/freesurfer that referenced this pull request Nov 30, 2021
Adds brightness handling through bias field correction basis functions.
The photo-seg flag now causes a change in the specification of the bias
field basis functions so that the component vectors are full vectors of
ones for the in plane directions and are vector elements of the
identity in the out of plane direction.

It remains to handle the downsampling so that the out of plane
directions aren't adversely affected.
htregidgo added a commit to htregidgo/freesurfer that referenced this pull request Nov 30, 2021
Modifies samsegment and estimateModelParameters to switch downsampling
behaviour when photos have been specified as an input. This is achieved
by skipping the out of plane resampling in estimateModelParameters and
allows uniform in plane basis functions to be used.
htregidgo added a commit to htregidgo/freesurfer that referenced this pull request Nov 30, 2021
Modifies the run_samseg script to add a flag that will bypass the
affine template registration and use the one that has been provided as
an accompanying input.
htregidgo added a commit to htregidgo/freesurfer that referenced this pull request Nov 30, 2021
Modifies the samsegment file to generate bias field basis functions
that are bilinear in plane and characteristic out of plane. current
behaviour is to run this when the photoseg flag has been provided.
redundant constant basis functions are still pressent within the code.
htregidgo added a commit to htregidgo/freesurfer that referenced this pull request Nov 30, 2021
This commit adds a regularisation parameter for fitting the brightness
bassis values when used on photos.

At first a slight mistake meant that I specified too many linear basis
funcions to be composed into the bilinear basis functions for the
brightness field. Initially I didn't notice as I was getting quite good
results, however eventually the fact that there was linear dependence
between my basis functions meant that in some cases the system matrix
would become singular and the segmentation would fail.

The first solution we tried to this was to correctly specify the basis
functions. However, this method was taking an excessive amount of time
to converge. Eventually we decided to go back to the additional basis
functions but add a regularisation penalty so that the system matrix
didn't become singular.
yhuang43 pushed a commit that referenced this pull request May 2, 2023
Update LUT with OB and hypothalamus labels
yhuang43 pushed a commit that referenced this pull request May 2, 2023
OB and hypothalamus LUT updates
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Development

Successfully merging this pull request may close these issues.

None yet

5 participants