View Issue Details

IDProjectCategoryView StatusLast Update
0003828Slicer4Module Editorpublic2014-09-10 08:07
ReporterZ o o m Assigned Topieper  
PrioritynormalSeverityminorReproducibilityalways
Status closedResolutionfixed 
Product VersionSlicer 4.3.1-2 
Target VersionFixed in VersionSlicer 4.4.0 
Summary0003828: wand tool does not work
Description

The wand tool does not allways work for me..... .
I've got 400 tifs, CT scan data; already segmented.
That label map was used to crop a smaller volume (ROI) to get a new, smaller lable map.
On that very lable map, all but one editor tools work.
Hovering with the wand over the volume and the lable map, gives 'Out of frame'
at places, which had not been covered by the ROI (There, ofcourse, none of the tools work)
But even at places without "Out of frame" the wand does not work anymore, quite frustrating .... .

Starting from scratch, wand works without a problem ... .

Hope you can help me here :)

greetings,
z o o m

TagsNo tags attached.

Activities

pieper

pieper

2014-09-08 05:30

administrator   ~0012479

First, please try the nightly build and see if the issue has already been fixed.

Also please look in the python console and error log and add any info from there to this report.

Also can you share any small dataset that can replicate the issue?

2014-09-08 08:52

 

python_interactor (5,262 bytes)
python_interactor (5,262 bytes)

2014-09-08 08:52

 

Slicer_error.jpg (185,661 bytes)
Slicer_error.jpg (185,661 bytes)

2014-09-08 08:53

 

HERR001.TIF (262,463 bytes)
Z o o m

Z o o m

2014-09-08 08:57

reporter   ~0012480

Last edited: 2014-09-08 09:08

First of all, thanks for answering that fast! :)

Ok, i gave the nightly a try:
When i start the Editor (After adding all necessary data) an error message pops up complaining about not matching master and merge volume geometry. (screenshot)
Edit:
Error:
###########################################################
Warning: Geometry of master and merge volumes do not match. Dimension mismatch at row [0] (512 != 552) Dimension mismatch at row [1] (512 != 348) Dimension mismatch at row [2] (391 != 420) IJKToRAS mismatch at [0, 0] (-0.294922 != 0.294922) IJKToRAS mismatch at [0, 3] (75.3525 != -89.5089) IJKToRAS mismatch at [1, 1] (-0.294922 != 0.294922) IJKToRAS mismatch at [1, 3] (75.3525 != -9.85254) IJKToRAS mismatch at [2, 3] (-195 != -214.053)
###########################################################
I copied the Python Interactor output, hope this is what you meant.

The *.TIF is one of 400 TIFs, but i thought one'd be enough :)

The lable file is the one, created from a ROI, crop, which was (of course) smaller than the image it self.
Any ideas?

2014-09-08 08:58

 

HERR001-label_1-unterkiefer_trennung01.nrrd (442,752 bytes)
pieper

pieper

2014-09-08 09:54

administrator   ~0012481

Ah, yes, now that makes sense. The other tools actually will only partially work on that data - anything that uses the CT, such as thresholding, would also fail. That's basically what the new dialog box is warning about.

The thing you can do is to resample it to the same size. You can use BRAINSResample for this. Create an identity transform and use NearestNeighbor resampling with the CT as the reference volume and the label as the image to 'warp'. The output should be in the space of the CT. You will probably need to change the output back to a label map (there's a checkbox for this in the Volumes module). Then you should be able to edit this new one. You can probably just reverse this process if you need the edited volume back in the ROI space.

Z o o m

Z o o m

2014-09-09 05:31

reporter   ~0012482

Ah, great you seem to know a solution for this :)

Anyway, i tried to resample it, but without success.
I've uploaded some screenshots for you to see what i did and what i got in return!

  • screen01 & 02.jpg: settings
  • screen03 & 04.jpg: error

screen03, error message:
"Resample Image (BRAINS) returned 1 which probably indicates an error."

screen04, error message:
##############################################################################
"Found SharedObject Module
ModuleType: SharedObjectModule
Resample Image (BRAINS) command line:

slicer:0x7f22a1c68130 --processinformationaddress 0x44e15f0 --inputVolume slicer:0x1a88b30#vtkMRMLScalarVolumeNode1 --referenceVolume slicer:0x1a88b30#vtkMRMLScalarVolumeNode3 --outputVolume slicer:0x1a88b30#vtkMRMLScalarVolumeNode4 --pixelType float --interpolationMode NearestNeighbor --defaultValue 0 --numberOfThreads -1

Resample Image (BRAINS) standard output:

=====================================================
Input Volume: slicer:0x1a88b30#vtkMRMLScalarVolumeNode1
Reference Volume: slicer:0x1a88b30#vtkMRMLScalarVolumeNode3
Output Volume: slicer:0x1a88b30#vtkMRMLScalarVolumeNode4
Pixel Type: float
Interpolation: NearestNeighbor
Background Value: 0

HERE /home/kitware/Dashboards/Nightly/Slicer-0-build/BRAINSTools/BRAINSResample/BRAINSResample.cxx 465

itk::ExceptionObject (0x7f227001f620)
Location: "unknown"
File: /home/kitware/Dashboards/Nightly/Slicer-0-build/ITKv4/Modules/IO/ImageBase/include/itkImageFileWriter.hxx
Line: 116
Description: itk::ERROR: ImageFileWriter(0x7f22700208b0): No input to writer!"
##############################################################################

Even tough a file is created (under 'data') this file can not be saved anywhere:
"Cannot write data file: /home/zoom/Uni/Bachelor_Arbeit/Media/Scan_Verarbeitung/33/Herrera001_Unterkiefer_LableMap_Resample.nrrd.
Do you want to continue saving?"

Sooo, what did i do wrong?

[Resampling the lable-map, 'created by' a ROI should be done automatically i guess .. :) ]

2014-09-09 05:32

 

screen01.jpg (896,186 bytes)
screen01.jpg (896,186 bytes)

2014-09-09 05:32

 

screen02.jpg (895,446 bytes)
screen02.jpg (895,446 bytes)

2014-09-09 05:32

 

screen03.jpg (734,646 bytes)
screen03.jpg (734,646 bytes)

2014-09-09 05:32

 

screen04.jpg (1,011,336 bytes)
screen04.jpg (1,011,336 bytes)
pieper

pieper

2014-09-09 08:51

administrator   ~0012484

I think you are close. First you need to make an identity transform and select it as the 'transform file'. Also you should pick the output type of 'short'.

Z o o m

Z o o m

2014-09-10 08:01

reporter   ~0012487

Great, thank you!
Even though i don't have a clue what all these transformation can do, this one did it, after checking this new file as a "label map" everything works again!

Greetings Z o o m

pieper

pieper

2014-09-10 08:07

administrator   ~0012488

Glad it worked!

Issue History

Date Modified Username Field Change
2014-09-08 02:59 Z o o m New Issue
2014-09-08 02:59 Z o o m Status new => assigned
2014-09-08 02:59 Z o o m Assigned To => pieper
2014-09-08 05:30 pieper Note Added: 0012479
2014-09-08 08:52 Z o o m File Added: python_interactor
2014-09-08 08:52 Z o o m File Added: Slicer_error.jpg
2014-09-08 08:53 Z o o m File Added: HERR001.TIF
2014-09-08 08:57 Z o o m Note Added: 0012480
2014-09-08 08:58 Z o o m File Added: HERR001-label_1-unterkiefer_trennung01.nrrd
2014-09-08 09:08 Z o o m Note Edited: 0012480
2014-09-08 09:54 pieper Note Added: 0012481
2014-09-09 05:31 Z o o m Note Added: 0012482
2014-09-09 05:32 Z o o m File Added: screen01.jpg
2014-09-09 05:32 Z o o m File Added: screen02.jpg
2014-09-09 05:32 Z o o m File Added: screen03.jpg
2014-09-09 05:32 Z o o m File Added: screen04.jpg
2014-09-09 08:51 pieper Note Added: 0012484
2014-09-10 08:01 Z o o m Note Added: 0012487
2014-09-10 08:07 pieper Note Added: 0012488
2014-09-10 08:07 pieper Status assigned => closed
2014-09-10 08:07 pieper Resolution open => fixed
2014-09-10 08:07 pieper Fixed in Version => Slicer 4.4.0