test annotation data download

[crazyhottommy]

Hi Chris, Where can I download the annotation files for your test data(half brain tif)? I only found the full brain tif from https://idisco.info/clearmap-2/

AtlasFile      = os.path.join(PathReg, 'half_template_25_right_fullWD.tif');
AnnotationFile = os.path.join(PathReg, 'annotation_25_right_fullWD.tif');

Thanks!

[vzickus]

At the bottom of that webpage, look for

Download the ClearMap ressources for the Mouse brain (48Mb, zip archive).

[crazyhottommy]

thanks @vzickus I downloaded that, but it is for full brain not for half brain.

[crazyhottommy]

@vzickus if you are using clearmap successfully. Do you mind if I ask questions? Thanks!

[vzickus]

Ah, you can just split it yourself (e.g. using ImageJ or otherwise).

[vzickus]

@crazyhottommy I'm afraid I can't say I have done it successfully! It would be interesting to compare notes if you ever get round to validating the cell counting (localization accuracy of humans vs machine).

[crazyhottommy]

Happy to share my notes. it is here https://hackmd.io/IKmM1BeVSyG4LDf_Q1dG0Q?view I now only want to finish the running first... as for benchmarking, I will consider later. One needs to have a ground truth to compare.(human counting)..

BTW, the elastix version matters or not?

I got this

Installing all components.
InstallingComponents was successful.

ELASTIX version: 4.900
Command line options from ElastixBase:
-in       unspecified, so no input image specified
-out      /tmp/elastix_output/
-threads  unspecified, so all available threads are used
-tp       /n/holylfs/LABS/informatics/mtang/projects/clearmap_test/test_data/haloperidol/1267/elastix_cfos_to_auto/TransformParameters.0.txt
-def      /tmp/elastix_input.txt
-jac      unspecified, so no det(dT/dx) computed
-jacmat   unspecified, so no dT/dx computed
Calling all ReadFromFile()'s ...
  Calling all ReadFromFile()'s took 0.000295 s
Transforming points ...

[vzickus]

4.9 worked fine for me. It's been a while since I used the whole thing, so don't remember the output (but it does not seem to be crashing?). I am mostly focusing on optimizing cell counting at the moment. I feel there might be some bugs left in the code (I tried following the tutorial from the supplementary materials from their 2016 paper, but couldn't complete it because of some issue that I did not yet manage to track down). Although my "debugging" is inserting print "I AM HERE" followed by raw_input("OK") so not very efficient!

[crazyhottommy]

It is not crashing, just no images were processed

Reading the elastix parameters from file ...

  Calling all ReadFromFile()'s took 0.028959 s
Transforming points ...
  The transform is evaluated on some points, specified in the input point file.
  Reading input point file: /tmp/elastix_input.txt
  Input points are specified as image indices.
  Number of specified input points: 6043
  The input points are transformed.
  The transformed points are saved in: /tmp/elastix_output/outputpoints.txt
  Transforming points done, it took 0.12s
Compute determinant of spatial Jacobian ...
  The command-line option "-jac" is not used, so no det(dT/dx) computed.
  Computing determinant of spatial Jacobian done, it took 0.00s
Compute spatial Jacobian (full matrix) ...
  The command-line option "-jacmat" is not used, so no dT/dx computed.
  Computing spatial Jacobian done, it took 0.00s

transformix has finished at Thu Jan 24 10:59:00 2019.
Total time elapsed: 0.3s.

Processed 0/6043

Processed 0/6043

hmm...so this package is not maintained anymore by the author? but if there are enough interests, we can do together to revive this library? I am very interested in hearing more on optimizing the counting.

[crazyhottommy]

@vzickus BTW this is Ming Tang, I am working for Catherin Dulac lab on some imaging data. Nice to meet you. I can be reached at tangming2005ATgmail.com

[vzickus]

It is not crashing, just no images were processed

Reading the elastix parameters from file ...

  Calling all ReadFromFile()'s took 0.028959 s
Transforming points ...
  The transform is evaluated on some points, specified in the input point file.
  Reading input point file: /tmp/elastix_input.txt
  Input points are specified as image indices.
  Number of specified input points: 6043
  The input points are transformed.
  The transformed points are saved in: /tmp/elastix_output/outputpoints.txt
  Transforming points done, it took 0.12s
Compute determinant of spatial Jacobian ...
  The command-line option "-jac" is not used, so no det(dT/dx) computed.
  Computing determinant of spatial Jacobian done, it took 0.00s
Compute spatial Jacobian (full matrix) ...
  The command-line option "-jacmat" is not used, so no dT/dx computed.
  Computing spatial Jacobian done, it took 0.00s

transformix has finished at Thu Jan 24 10:59:00 2019.
Total time elapsed: 0.3s.

Processed 0/6043

Processed 0/6043

hmm...so this package is not maintained anymore by the author? but if there are enough interests, we can do together to revive this library? I am very interested in hearing more on optimizing the counting.

Actually, I seem to recall something like that. If you test elastix by itself from terminal, it should work no problem. I think the issue was with paths.

EDIT: Actually, I had raised an issue about something similar, check the closed #12 issue.

[vzickus]

@vzickus BTW this is Ming Tang, I am working for Catherin Dulac lab on some imaging data. Nice to meet you. I can be reached at tangming2005ATgmail.com

Thanks, Ming. I've e-mailed you to avoid spamming here.

[crazyhottommy]

I had something similar to https://github.com/ChristophKirst/ClearMap/issues/12 before, and it seems the folder name has space (the folder was downloaded from the link you gave), I later removed the space and it worked.

I guess this error is the same here because the image tif files(downloaded from osf) have spaces as well.

[aaubry89]

Can either off you provide some help on how to split the annotation file in Image J? Feel free to contact me at Antonio.aubry@mssm.edu. Thanks!

[vzickus]

@aaubry89 I think you want the function "Stack splitter".

[aaubry89]

Thanks, do I also need to split the .nrrd annotation file or just the .tif annotation file?