mitk-users Mailing List for MITK
Brought to you by:
mitk
Menu
▾
▴
mitk-users — Post questions about compiling and using MITK here.
You can subscribe to this list here.
| 2005 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(28) |
Dec
(35) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2006 |
Jan
(59) |
Feb
(35) |
Mar
(45) |
Apr
(20) |
May
(34) |
Jun
(33) |
Jul
(38) |
Aug
(7) |
Sep
(47) |
Oct
(31) |
Nov
(23) |
Dec
(9) |
| 2007 |
Jan
(5) |
Feb
(19) |
Mar
(18) |
Apr
(20) |
May
(24) |
Jun
(23) |
Jul
(23) |
Aug
(59) |
Sep
(35) |
Oct
(36) |
Nov
(18) |
Dec
(13) |
| 2008 |
Jan
(26) |
Feb
(76) |
Mar
(53) |
Apr
(36) |
May
(24) |
Jun
(26) |
Jul
(42) |
Aug
(50) |
Sep
(37) |
Oct
(36) |
Nov
(47) |
Dec
(34) |
| 2009 |
Jan
(36) |
Feb
(25) |
Mar
(35) |
Apr
(59) |
May
(23) |
Jun
(48) |
Jul
(32) |
Aug
(11) |
Sep
(37) |
Oct
(26) |
Nov
(48) |
Dec
(17) |
| 2010 |
Jan
(13) |
Feb
(45) |
Mar
(52) |
Apr
(63) |
May
(49) |
Jun
(75) |
Jul
(113) |
Aug
(64) |
Sep
(52) |
Oct
(50) |
Nov
(57) |
Dec
(36) |
| 2011 |
Jan
(43) |
Feb
(27) |
Mar
(37) |
Apr
(56) |
May
(87) |
Jun
(80) |
Jul
(144) |
Aug
(149) |
Sep
(125) |
Oct
(112) |
Nov
(129) |
Dec
(96) |
| 2012 |
Jan
(104) |
Feb
(82) |
Mar
(76) |
Apr
(69) |
May
(136) |
Jun
(58) |
Jul
(61) |
Aug
(54) |
Sep
(58) |
Oct
(139) |
Nov
(56) |
Dec
(28) |
| 2013 |
Jan
(95) |
Feb
(91) |
Mar
(128) |
Apr
(72) |
May
(44) |
Jun
(71) |
Jul
(99) |
Aug
(79) |
Sep
(77) |
Oct
(118) |
Nov
(70) |
Dec
(112) |
| 2014 |
Jan
(81) |
Feb
(101) |
Mar
(121) |
Apr
(102) |
May
(57) |
Jun
(37) |
Jul
(92) |
Aug
(134) |
Sep
(71) |
Oct
(76) |
Nov
(56) |
Dec
(71) |
| 2015 |
Jan
(70) |
Feb
(77) |
Mar
(87) |
Apr
(79) |
May
(89) |
Jun
(123) |
Jul
(98) |
Aug
(189) |
Sep
(116) |
Oct
(81) |
Nov
(85) |
Dec
(19) |
| 2016 |
Jan
(46) |
Feb
(62) |
Mar
(67) |
Apr
(68) |
May
(40) |
Jun
(34) |
Jul
(48) |
Aug
(81) |
Sep
(44) |
Oct
(48) |
Nov
(52) |
Dec
(25) |
| 2017 |
Jan
(39) |
Feb
(49) |
Mar
(51) |
Apr
(48) |
May
(45) |
Jun
(40) |
Jul
(29) |
Aug
(25) |
Sep
(35) |
Oct
(39) |
Nov
(19) |
Dec
(26) |
| 2018 |
Jan
(23) |
Feb
(9) |
Mar
(48) |
Apr
(30) |
May
(33) |
Jun
(45) |
Jul
(44) |
Aug
(9) |
Sep
(28) |
Oct
(27) |
Nov
(20) |
Dec
(10) |
| 2019 |
Jan
(18) |
Feb
(6) |
Mar
(7) |
Apr
(19) |
May
(34) |
Jun
(71) |
Jul
(42) |
Aug
(13) |
Sep
(27) |
Oct
(50) |
Nov
(21) |
Dec
(17) |
| 2020 |
Jan
(6) |
Feb
(27) |
Mar
(23) |
Apr
(23) |
May
(19) |
Jun
(20) |
Jul
(12) |
Aug
(5) |
Sep
(26) |
Oct
(22) |
Nov
(7) |
Dec
(13) |
| 2021 |
Jan
(41) |
Feb
(8) |
Mar
(15) |
Apr
(22) |
May
(22) |
Jun
(6) |
Jul
(4) |
Aug
(1) |
Sep
(6) |
Oct
(12) |
Nov
(4) |
Dec
(3) |
| 2022 |
Jan
(7) |
Feb
(3) |
Mar
(10) |
Apr
(2) |
May
(2) |
Jun
(6) |
Jul
(10) |
Aug
(4) |
Sep
(3) |
Oct
(5) |
Nov
(5) |
Dec
(1) |
| 2023 |
Jan
|
Feb
(9) |
Mar
(19) |
Apr
(1) |
May
|
Jun
(4) |
Jul
|
Aug
(4) |
Sep
(2) |
Oct
(1) |
Nov
(11) |
Dec
(6) |
| 2024 |
Jan
(9) |
Feb
(8) |
Mar
(3) |
Apr
|
May
(2) |
Jun
|
Jul
(2) |
Aug
(3) |
Sep
(6) |
Oct
(2) |
Nov
|
Dec
|
| 2025 |
Jan
|
Feb
(19) |
Mar
(7) |
Apr
(16) |
May
(1) |
Jun
|
Jul
|
Aug
(3) |
Sep
(7) |
Oct
|
Nov
|
Dec
|
|
From: Tobias S. <t.s...@tu...> - 2025-09-08 11:37:36
|
Hello Stephen, thanks a lot for all the hints. In the end I was able to use the volume visualisation which I had encountered before but I couldn't see any 3D rendering. The problem was apparently that I had to turn on and off the "Volume rendering" several times and that it took some time to show up. It is really amazing what this project has achieved! I am very grateful for all the descriptions and support and I hope other future users can benefit from it. I don't know if people usually know how to use the program. Maybe the essence could go into a tutorial or a video tutorials that shows how to start into MITK or something. Best regards, Tobias Von: Schaumann, Stephen <s.s...@dk...> Gesendet: Dienstag, 2. September 2025 17:18 An: Tobias Schmetzer <t.s...@tu...>; mit...@li... Betreff: AW: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Tobias, the demo videos were made in napari, but you can get comparable visualisations in MITK. The lower-right render window of the 2x2 Standard Display in MITK is the 3D window. You can maximize it by hovering your mouse over it and then clicking the middle square of the three options that pop up in the upper right of the render window (between the "+" and tools-icon). Click the "+"-icon and disable "Show crosshair" so that the 2D planes are not in the way. To show your segmentation labels in 3D, you probably want to do a right-click on the corresponding node in the Data Manager and "Create smoothed polygon model". This will create a new node for each label that is shown in 3D. For an exact representation of the label with distinct voxels, you can instead choose "Create polygon model" or select the segmentation node in the "Properties" view and enable the multilabel -> 3D -> visualize option. If you want to overlay the segmentation with a 3D rendering of the base image, you can use the "Volume Visualization" view. There, you can select your image and turn on "Volume rendering". There are several presets available, but you can also fully customize the visualization to your liking (take a look at the F1 help page for more information). I hope this helps you create the visualization you seek. Hint: if you want to save / share the rendering, MITK lets you take snapshots with the "Screenshot Maker" view. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...<mailto:t.s...@tu...>> Gesendet: Dienstag, 2. September 2025 16:49:27 An: mit...@li...<mailto:mit...@li...> Betreff: Re: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Stephen, apparently something was wrong when I tried it the last time. I think all were greyed out. Today it worked. Maybe the images were not the same ones or something was not aligned and I wasn't aware. The OpenGL message was exactly that one. I will ignore it then. One last thing that I couldn't find out so far: How can I activate the 3D view on Windows as in the demo video? Do I manually need to install napari somehow (as described in https://github.com/MIC-DKFZ/napari-nninteractive) or how do I start it? Best regards, Tobias Von: Schaumann, Stephen <s.s...@dk...<mailto:s.s...@dk...>> Gesendet: Dienstag, 2. September 2025 11:12 An: Tobias Schmetzer <t.s...@tu...<mailto:t.s...@tu...>>; mit...@li...<mailto:mit...@li...> Betreff: AW: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Tobias, do the other segmentation tools work and only nnInteractive is unavailable, or are all tools greyed out? If it's the latter, can you check that the correct image is selected at the top of the Segmentation View and that this image is currently visible + you have performed a reinit on it? If it is not visible / not centered, there should also be a red text below the selection about that. Does the OpenGL warning say "WARNING: An OpenGL Core Profile was requested, but it is not supported on the current platform. Falling back to a non-Core profile. Note that this might cause rendering issues." ? In that case, the warning is known/expected behavior and you can ignore it. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...<mailto:t.s...@tu...>> Gesendet: Montag, 1. September 2025 18:37:49 An: mit...@li...<mailto:mit...@li...> Betreff: Re: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Stephen, Sorry for not describing this earlier. I meant in the segmentation menu in the 3d tools I cannot press the button nnInteractive. By the way I get a warning about a unsupported OpenGL profile no matter if the computer has a Nvidia graphics card or not (I tried 3 different computers) Viele Grüße, Tobias Schmetzer ________________________________ Von: Schaumann, Stephen <s.s...@dk...<mailto:s.s...@dk...>> Gesendet: Montag, 1. September 2025 15:42:00 An: Tobias Schmetzer; mit...@li...<mailto:mit...@li...> Betreff: AW: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello Tobias, could you elaborate what you mean with it not working? Is the nnInteractive Segmentation of bad quality or does it fail to produce any segmentation at all? Are there any warning / error messages in the MITK Workbench or in the console log? In principle, nnInteractive should work for any image loaded into MITK. Results may vary, of course, based on the image (modality) and how it compares to the training data used for nnInteractive. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...<mailto:t.s...@tu...>> Gesendet: Montag, 1. September 2025 12:05:15 An: mit...@li...<mailto:mit...@li...> Betreff: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello, I have MRT acquisitions from two different radiologies. Both are in DICOM format. nnInteractive seems to work on only one of them. Is the nnInteractive feature restricted to certain DICOM formats or is there a different problem? Do you want me to post the DICOM specifics here or somewhere else? Viele Grüße, Tobias Schmetzer |
|
From: Schaumann, S. <s.s...@dk...> - 2025-09-02 15:17:50
|
Hello Tobias, the demo videos were made in napari, but you can get comparable visualisations in MITK. The lower-right render window of the 2x2 Standard Display in MITK is the 3D window. You can maximize it by hovering your mouse over it and then clicking the middle square of the three options that pop up in the upper right of the render window (between the "+" and tools-icon). Click the "+"-icon and disable "Show crosshair" so that the 2D planes are not in the way. To show your segmentation labels in 3D, you probably want to do a right-click on the corresponding node in the Data Manager and "Create smoothed polygon model". This will create a new node for each label that is shown in 3D. For an exact representation of the label with distinct voxels, you can instead choose "Create polygon model" or select the segmentation node in the "Properties" view and enable the multilabel -> 3D -> visualize option. If you want to overlay the segmentation with a 3D rendering of the base image, you can use the "Volume Visualization" view. There, you can select your image and turn on "Volume rendering". There are several presets available, but you can also fully customize the visualization to your liking (take a look at the F1 help page for more information). I hope this helps you create the visualization you seek. Hint: if you want to save / share the rendering, MITK lets you take snapshots with the "Screenshot Maker" view. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...> Gesendet: Dienstag, 2. September 2025 16:49:27 An: mit...@li... Betreff: Re: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Stephen, apparently something was wrong when I tried it the last time. I think all were greyed out. Today it worked. Maybe the images were not the same ones or something was not aligned and I wasn’t aware. The OpenGL message was exactly that one. I will ignore it then. One last thing that I couldn’t find out so far: How can I activate the 3D view on Windows as in the demo video? Do I manually need to install napari somehow (as described in https://github.com/MIC-DKFZ/napari-nninteractive) or how do I start it? Best regards, Tobias Von: Schaumann, Stephen <s.s...@dk...> Gesendet: Dienstag, 2. September 2025 11:12 An: Tobias Schmetzer <t.s...@tu...>; mit...@li... Betreff: AW: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Tobias, do the other segmentation tools work and only nnInteractive is unavailable, or are all tools greyed out? If it's the latter, can you check that the correct image is selected at the top of the Segmentation View and that this image is currently visible + you have performed a reinit on it? If it is not visible / not centered, there should also be a red text below the selection about that. Does the OpenGL warning say "WARNING: An OpenGL Core Profile was requested, but it is not supported on the current platform. Falling back to a non-Core profile. Note that this might cause rendering issues." ? In that case, the warning is known/expected behavior and you can ignore it. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...<mailto:t.s...@tu...>> Gesendet: Montag, 1. September 2025 18:37:49 An: mit...@li...<mailto:mit...@li...> Betreff: Re: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Stephen, Sorry for not describing this earlier. I meant in the segmentation menu in the 3d tools I cannot press the button nnInteractive. By the way I get a warning about a unsupported OpenGL profile no matter if the computer has a Nvidia graphics card or not (I tried 3 different computers) Viele Grüße, Tobias Schmetzer ________________________________ Von: Schaumann, Stephen <s.s...@dk...<mailto:s.s...@dk...>> Gesendet: Montag, 1. September 2025 15:42:00 An: Tobias Schmetzer; mit...@li...<mailto:mit...@li...> Betreff: AW: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello Tobias, could you elaborate what you mean with it not working? Is the nnInteractive Segmentation of bad quality or does it fail to produce any segmentation at all? Are there any warning / error messages in the MITK Workbench or in the console log? In principle, nnInteractive should work for any image loaded into MITK. Results may vary, of course, based on the image (modality) and how it compares to the training data used for nnInteractive. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...<mailto:t.s...@tu...>> Gesendet: Montag, 1. September 2025 12:05:15 An: mit...@li...<mailto:mit...@li...> Betreff: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello, I have MRT acquisitions from two different radiologies. Both are in DICOM format. nnInteractive seems to work on only one of them. Is the nnInteractive feature restricted to certain DICOM formats or is there a different problem? Do you want me to post the DICOM specifics here or somewhere else? Viele Grüße, Tobias Schmetzer |
|
From: Tobias S. <t.s...@tu...> - 2025-09-02 14:49:45
|
Hello Stephen, apparently something was wrong when I tried it the last time. I think all were greyed out. Today it worked. Maybe the images were not the same ones or something was not aligned and I wasn't aware. The OpenGL message was exactly that one. I will ignore it then. One last thing that I couldn't find out so far: How can I activate the 3D view on Windows as in the demo video? Do I manually need to install napari somehow (as described in https://github.com/MIC-DKFZ/napari-nninteractive) or how do I start it? Best regards, Tobias Von: Schaumann, Stephen <s.s...@dk...> Gesendet: Dienstag, 2. September 2025 11:12 An: Tobias Schmetzer <t.s...@tu...>; mit...@li... Betreff: AW: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Tobias, do the other segmentation tools work and only nnInteractive is unavailable, or are all tools greyed out? If it's the latter, can you check that the correct image is selected at the top of the Segmentation View and that this image is currently visible + you have performed a reinit on it? If it is not visible / not centered, there should also be a red text below the selection about that. Does the OpenGL warning say "WARNING: An OpenGL Core Profile was requested, but it is not supported on the current platform. Falling back to a non-Core profile. Note that this might cause rendering issues." ? In that case, the warning is known/expected behavior and you can ignore it. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...<mailto:t.s...@tu...>> Gesendet: Montag, 1. September 2025 18:37:49 An: mit...@li...<mailto:mit...@li...> Betreff: Re: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Stephen, Sorry for not describing this earlier. I meant in the segmentation menu in the 3d tools I cannot press the button nnInteractive. By the way I get a warning about a unsupported OpenGL profile no matter if the computer has a Nvidia graphics card or not (I tried 3 different computers) Viele Grüße, Tobias Schmetzer ________________________________ Von: Schaumann, Stephen <s.s...@dk...<mailto:s.s...@dk...>> Gesendet: Montag, 1. September 2025 15:42:00 An: Tobias Schmetzer; mit...@li...<mailto:mit...@li...> Betreff: AW: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello Tobias, could you elaborate what you mean with it not working? Is the nnInteractive Segmentation of bad quality or does it fail to produce any segmentation at all? Are there any warning / error messages in the MITK Workbench or in the console log? In principle, nnInteractive should work for any image loaded into MITK. Results may vary, of course, based on the image (modality) and how it compares to the training data used for nnInteractive. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...<mailto:t.s...@tu...>> Gesendet: Montag, 1. September 2025 12:05:15 An: mit...@li...<mailto:mit...@li...> Betreff: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello, I have MRT acquisitions from two different radiologies. Both are in DICOM format. nnInteractive seems to work on only one of them. Is the nnInteractive feature restricted to certain DICOM formats or is there a different problem? Do you want me to post the DICOM specifics here or somewhere else? Viele Grüße, Tobias Schmetzer |
|
From: Schaumann, S. <s.s...@dk...> - 2025-09-02 14:31:46
|
Hello Tobias, do the other segmentation tools work and only nnInteractive is unavailable, or are all tools greyed out? If it's the latter, can you check that the correct image is selected at the top of the Segmentation View and that this image is currently visible + you have performed a reinit on it? If it is not visible / not centered, there should also be a red text below the selection about that. Does the OpenGL warning say "WARNING: An OpenGL Core Profile was requested, but it is not supported on the current platform. Falling back to a non-Core profile. Note that this might cause rendering issues." ? In that case, the warning is known/expected behavior and you can ignore it. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...> Gesendet: Montag, 1. September 2025 18:37:49 An: mit...@li... Betreff: Re: [mitk-users] [Extern] - nnInteractive only for specific DICOM formats? Hello Stephen, Sorry for not describing this earlier. I meant in the segmentation menu in the 3d tools I cannot press the button nnInteractive. By the way I get a warning about a unsupported OpenGL profile no matter if the computer has a Nvidia graphics card or not (I tried 3 different computers) Viele Grüße, Tobias Schmetzer ________________________________ Von: Schaumann, Stephen <s.s...@dk...> Gesendet: Montag, 1. September 2025 15:42:00 An: Tobias Schmetzer; mit...@li... Betreff: AW: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello Tobias, could you elaborate what you mean with it not working? Is the nnInteractive Segmentation of bad quality or does it fail to produce any segmentation at all? Are there any warning / error messages in the MITK Workbench or in the console log? In principle, nnInteractive should work for any image loaded into MITK. Results may vary, of course, based on the image (modality) and how it compares to the training data used for nnInteractive. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...> Gesendet: Montag, 1. September 2025 12:05:15 An: mit...@li... Betreff: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello, I have MRT acquisitions from two different radiologies. Both are in DICOM format. nnInteractive seems to work on only one of them. Is the nnInteractive feature restricted to certain DICOM formats or is there a different problem? Do you want me to post the DICOM specifics here or somewhere else? Viele Grüße, Tobias Schmetzer |
|
From: Tobias S. <t.s...@tu...> - 2025-09-01 16:38:08
|
Hello Stephen, Sorry for not describing this earlier. I meant in the segmentation menu in the 3d tools I cannot press the button nnInteractive. By the way I get a warning about a unsupported OpenGL profile no matter if the computer has a Nvidia graphics card or not (I tried 3 different computers) Viele Grüße, Tobias Schmetzer ________________________________ Von: Schaumann, Stephen <s.s...@dk...> Gesendet: Montag, 1. September 2025 15:42:00 An: Tobias Schmetzer; mit...@li... Betreff: AW: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello Tobias, could you elaborate what you mean with it not working? Is the nnInteractive Segmentation of bad quality or does it fail to produce any segmentation at all? Are there any warning / error messages in the MITK Workbench or in the console log? In principle, nnInteractive should work for any image loaded into MITK. Results may vary, of course, based on the image (modality) and how it compares to the training data used for nnInteractive. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...> Gesendet: Montag, 1. September 2025 12:05:15 An: mit...@li... Betreff: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello, I have MRT acquisitions from two different radiologies. Both are in DICOM format. nnInteractive seems to work on only one of them. Is the nnInteractive feature restricted to certain DICOM formats or is there a different problem? Do you want me to post the DICOM specifics here or somewhere else? Viele Grüße, Tobias Schmetzer |
|
From: Schaumann, S. <s.s...@dk...> - 2025-09-01 15:29:58
|
Hello Tobias, could you elaborate what you mean with it not working? Is the nnInteractive Segmentation of bad quality or does it fail to produce any segmentation at all? Are there any warning / error messages in the MITK Workbench or in the console log? In principle, nnInteractive should work for any image loaded into MITK. Results may vary, of course, based on the image (modality) and how it compares to the training data used for nnInteractive. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...> Gesendet: Montag, 1. September 2025 12:05:15 An: mit...@li... Betreff: [Extern] - [mitk-users] nnInteractive only for specific DICOM formats? Hello, I have MRT acquisitions from two different radiologies. Both are in DICOM format. nnInteractive seems to work on only one of them. Is the nnInteractive feature restricted to certain DICOM formats or is there a different problem? Do you want me to post the DICOM specifics here or somewhere else? Viele Grüße, Tobias Schmetzer |
|
From: Tobias S. <t.s...@tu...> - 2025-09-01 10:05:25
|
Hello, I have MRT acquisitions from two different radiologies. Both are in DICOM format. nnInteractive seems to work on only one of them. Is the nnInteractive feature restricted to certain DICOM formats or is there a different problem? Do you want me to post the DICOM specifics here or somewhere else? Viele Grüße, Tobias Schmetzer |
|
From: Dinkelacker, S. <s.d...@dk...> - 2025-08-27 13:54:11
|
Hi Tobias, I assume you did get your image data in DICOM format. While the data is typically saved in multiple files (for example one file per image slice of a 3-d or 4-d image), it is actually organized a bit differently based on patient, study, and series. I can only further assume that your DICOM data may contain multiple images of your hand, possibly mixing all those files of individual slices in a single folder. For example, it is quite common to also have low-res localizer images or images for different MR sequences. Hence, I highly recommend to use the DICOM Browser in MITK to open your image(s), instead of File->Open. Click on the big DICOM button in the toolbar at the top. The DICOM Browser should open. On the Import tab click on Scan directory and select the root directory of your data. You should now see at least an entry in the patient widget. Click on it to see at least an entry below in the study and series widgets. The entries in the series widget are the actual individual 3-d images of your hand. You can select them and click on the View button at the top of the DICOM Browser or just double-click on them to open them in the Data Manager and finally see them in the Standard Display of the MITK Workbench. Hope that helps, Stefan PS: Try out the latest MITK v2025.08, which was released today! :) PPS: While the DICOM browser is active, you can press F1 to open a context-specific help page where you can read more about the DICOM browser. The same applies to other plugins of the MITK Workbench. ________________________________________ Von: Schaumann, Stephen <s.s...@dk...> Gesendet: Dienstag, 26. August 2025 16:29 An: Tobias Schmetzer; mit...@li... Betreff: Re: [mitk-users] [Extern] - How to view the whole data Hello Tobias, the statement regarding use in diagnosis or treatment is of a legal nature. Functionally, there is no restriction as to which data can be loaded because of this. If your data is in a common file format that MITK can handle, you should be able to load it. When you've loaded your data, do you see multiple entries listed in the "Data Manager" view? If so, by default the highest one will be rendered on top of the entries below it. In that case, try changing the visibility or remove the highest entry to see the image listed below. If there is only a single entry in the "Data Manager", what you see is the only image currently loaded. Either there are more images that you haven't loaded into MITK yet, or that's all there is. In case you are unable to load all your data, maybe give us an overview of the folder structure and what exactly you are currently doing to load the data. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...> Gesendet: Dienstag, 26. August 2025 15:58:15 An: mit...@li... Betreff: [Extern] - [mitk-users] How to view the whole data Dear Co-users, I’ve searched the Wiki and the mailing list archive but couln’t find any hint regarding a solution for my question so I now consult this mailing list. I am a computer engineer and I had an accident on my hand. I came across this great tool to analyse my personal MRT data of my hand in new ways such as advanced segmentation because the radiologists seem to have a hard time encountering the location causing the trouble with conventional tools. After I loaded my data I can only see a low resolution of the data and in the main window I find the statement „Not for use in diagnosis ir treatment“. The radiology that recorded the data claims that all data is available in the data package but the computer doesn’t have the required setup/license/calibration to display all data. What setup is required to display it or is there any option to avoid this check at my own risk? I mean, I can still discuss this with a specialialist on a calibrated computer later. Thanks in advance for any hints! Best regards, Tobias Schmetzer |
|
From: Schaumann, S. <s.s...@dk...> - 2025-08-26 14:30:00
|
Hello Tobias, the statement regarding use in diagnosis or treatment is of a legal nature. Functionally, there is no restriction as to which data can be loaded because of this. If your data is in a common file format that MITK can handle, you should be able to load it. When you've loaded your data, do you see multiple entries listed in the "Data Manager" view? If so, by default the highest one will be rendered on top of the entries below it. In that case, try changing the visibility or remove the highest entry to see the image listed below. If there is only a single entry in the "Data Manager", what you see is the only image currently loaded. Either there are more images that you haven't loaded into MITK yet, or that's all there is. In case you are unable to load all your data, maybe give us an overview of the folder structure and what exactly you are currently doing to load the data. Best regards, Stephen ________________________________ Von: Tobias Schmetzer <t.s...@tu...> Gesendet: Dienstag, 26. August 2025 15:58:15 An: mit...@li... Betreff: [Extern] - [mitk-users] How to view the whole data Dear Co-users, I’ve searched the Wiki and the mailing list archive but couln’t find any hint regarding a solution for my question so I now consult this mailing list. I am a computer engineer and I had an accident on my hand. I came across this great tool to analyse my personal MRT data of my hand in new ways such as advanced segmentation because the radiologists seem to have a hard time encountering the location causing the trouble with conventional tools. After I loaded my data I can only see a low resolution of the data and in the main window I find the statement „Not for use in diagnosis ir treatment“. The radiology that recorded the data claims that all data is available in the data package but the computer doesn’t have the required setup/license/calibration to display all data. What setup is required to display it or is there any option to avoid this check at my own risk? I mean, I can still discuss this with a specialialist on a calibrated computer later. Thanks in advance for any hints! Best regards, Tobias Schmetzer |
|
From: Tobias S. <t.s...@tu...> - 2025-08-26 14:12:59
|
Dear Co-users, I've searched the Wiki and the mailing list archive but couln't find any hint regarding a solution for my question so I now consult this mailing list. I am a computer engineer and I had an accident on my hand. I came across this great tool to analyse my personal MRT data of my hand in new ways such as advanced segmentation because the radiologists seem to have a hard time encountering the location causing the trouble with conventional tools. After I loaded my data I can only see a low resolution of the data and in the main window I find the statement "Not for use in diagnosis ir treatment". The radiology that recorded the data claims that all data is available in the data package but the computer doesn't have the required setup/license/calibration to display all data. What setup is required to display it or is there any option to avoid this check at my own risk? I mean, I can still discuss this with a specialialist on a calibrated computer later. Thanks in advance for any hints! Best regards, Tobias Schmetzer |
|
From: Mujassim J. <muj...@gm...> - 2025-05-01 03:21:59
|
Thanks for the clarification and for creating the ticket—sounds good for now! On Tue, 29 Apr 2025 at 19:29, Dinkelacker, Stefan <s.d...@dk...> wrote: > > This should work for our GUI applications like MitkWorkbench. However, I double-checked and noticed that the setting influences our console applications as well, which can be considered a bug as-is. For now I advice to live with the console window until we fix it. I'll create an issue ticket for the topic. Thanks for reporting! > > Best, > Stefan > ________________________________________ > Von: Mujassim Jamal <muj...@gm...> > Gesendet: Dienstag, 29. April 2025 08:11 > An: Dinkelacker, Stefan > Cc: mit...@li... > Betreff: Re: [Extern] - [mitk-users] How to hide or disable the console window at startup > > Thanks for pointing that out. I tried changing it, and now it does > show "Windows (/SUBSYSTEM:WINDOWS)" in the SubSystem property. > However, I'm now getting another error: > > unresolved external symbol WinMain referenced in function "int __cdecl > __scrt_common_main_seh(void)" (?__scrt_common_main_seh@@YAHXZ). > > On Tue, 29 Apr 2025 at 10:21, Dinkelacker, Stefan > <s.d...@dk...> wrote: > > > > It is an MITK-build setting, not a superbuild setting. Make sure to set it while the MITK-build subfolder is selected as build folder in CMake GUI. Hence, you do not need to rebuild anything on superbuild level. > > > > ________________________________ > > From: Mujassim Jamal <muj...@gm...> > > Sent: Monday, April 28, 2025 11:12:20 AM > > To: Dinkelacker, Stefan <s.d...@dk...> > > Cc: mit...@li... <mit...@li...> > > Subject: Re: [Extern] - [mitk-users] How to hide or disable the console window at startup > > > > Hi Stefan, > > > > I have built MITK on Windows 10. Here are the steps I followed, but > > the console window still appears: > > 1. In the Visual Studio IDE, under the MITK-build project's > > CMakeLists.txt, I turned OFF MITK_SHOW_CONSOLE_WINDOW. > > 2. Opened CMake-GUI and regenerated the CMake files by clicking the > > Generate button. > > 3. Opened the MITK-superbuild.sln from CMake-GUI itself by clicking > > the Open Project button. > > 4. Rebuilt the solution, then opened MITK.sln and ran the Workbench, > > but the console window still appears. > > > > The SubSystem property still shows Console (/SUBSYSTEM:CONSOLE). When > > I explicitly printed the value of MITK_SHOW_CONSOLE_WINDOW, it shows > > ON even though it is set to OFF. Maybe it is related to a CMake cache > > issue? > > > > Best, > > Mujassim > > > > On Mon, 28 Apr 2025 at 00:47, Dinkelacker, Stefan > > <s.d...@dk...> wrote: > > > > > > Hi, > > > > > > I just switched off MITK_SHOW_CONSOLE_WINDOW to OFF in my develop branch MITK-build and the console window disappeared. Which OS and version are you using? Did you regenerate the CMake files after switching it off? > > > > > > It is supposed to influence the "Linker > System > SubSystem" project property of the MitkWorkbench project in the MITK solution. Instead of "Console (/SUBSYSTEM:CONSOLE)" it should display "Windows (/SUBSYSTEM:WINDOWS)". > > > > > > Best, > > > Stefan > > > ________________________________________ > > > Von: Mujassim Jamal <muj...@gm...> > > > Gesendet: Sonntag, 27. April 2025 06:19 > > > An: mit...@li... > > > Betreff: [Extern] - [mitk-users] How to hide or disable the console window at startup > > > > > > Hi again, > > > > > > I want to hide or disable the console window that appears at the > > > startup of MitkWorkbench. I tried building MITK by setting > > > MITK_SHOW_CONSOLE_WINDOW to OFF in CMake-GUI. However, the console > > > window still appears. > > > > > > I would appreciate any help. > > > > > > Best, > > > Mujassim > > > > > > > > > _______________________________________________ > > > mitk-users mailing list > > > mit...@li... > > > https://lists.sourceforge.net/lists/listinfo/mitk-users |
|
From: Dinkelacker, S. <s.d...@dk...> - 2025-04-29 14:00:23
|
This should work for our GUI applications like MitkWorkbench. However, I double-checked and noticed that the setting influences our console applications as well, which can be considered a bug as-is. For now I advice to live with the console window until we fix it. I'll create an issue ticket for the topic. Thanks for reporting! Best, Stefan ________________________________________ Von: Mujassim Jamal <muj...@gm...> Gesendet: Dienstag, 29. April 2025 08:11 An: Dinkelacker, Stefan Cc: mit...@li... Betreff: Re: [Extern] - [mitk-users] How to hide or disable the console window at startup Thanks for pointing that out. I tried changing it, and now it does show "Windows (/SUBSYSTEM:WINDOWS)" in the SubSystem property. However, I'm now getting another error: unresolved external symbol WinMain referenced in function "int __cdecl __scrt_common_main_seh(void)" (?__scrt_common_main_seh@@YAHXZ). On Tue, 29 Apr 2025 at 10:21, Dinkelacker, Stefan <s.d...@dk...> wrote: > > It is an MITK-build setting, not a superbuild setting. Make sure to set it while the MITK-build subfolder is selected as build folder in CMake GUI. Hence, you do not need to rebuild anything on superbuild level. > > ________________________________ > From: Mujassim Jamal <muj...@gm...> > Sent: Monday, April 28, 2025 11:12:20 AM > To: Dinkelacker, Stefan <s.d...@dk...> > Cc: mit...@li... <mit...@li...> > Subject: Re: [Extern] - [mitk-users] How to hide or disable the console window at startup > > Hi Stefan, > > I have built MITK on Windows 10. Here are the steps I followed, but > the console window still appears: > 1. In the Visual Studio IDE, under the MITK-build project's > CMakeLists.txt, I turned OFF MITK_SHOW_CONSOLE_WINDOW. > 2. Opened CMake-GUI and regenerated the CMake files by clicking the > Generate button. > 3. Opened the MITK-superbuild.sln from CMake-GUI itself by clicking > the Open Project button. > 4. Rebuilt the solution, then opened MITK.sln and ran the Workbench, > but the console window still appears. > > The SubSystem property still shows Console (/SUBSYSTEM:CONSOLE). When > I explicitly printed the value of MITK_SHOW_CONSOLE_WINDOW, it shows > ON even though it is set to OFF. Maybe it is related to a CMake cache > issue? > > Best, > Mujassim > > On Mon, 28 Apr 2025 at 00:47, Dinkelacker, Stefan > <s.d...@dk...> wrote: > > > > Hi, > > > > I just switched off MITK_SHOW_CONSOLE_WINDOW to OFF in my develop branch MITK-build and the console window disappeared. Which OS and version are you using? Did you regenerate the CMake files after switching it off? > > > > It is supposed to influence the "Linker > System > SubSystem" project property of the MitkWorkbench project in the MITK solution. Instead of "Console (/SUBSYSTEM:CONSOLE)" it should display "Windows (/SUBSYSTEM:WINDOWS)". > > > > Best, > > Stefan > > ________________________________________ > > Von: Mujassim Jamal <muj...@gm...> > > Gesendet: Sonntag, 27. April 2025 06:19 > > An: mit...@li... > > Betreff: [Extern] - [mitk-users] How to hide or disable the console window at startup > > > > Hi again, > > > > I want to hide or disable the console window that appears at the > > startup of MitkWorkbench. I tried building MITK by setting > > MITK_SHOW_CONSOLE_WINDOW to OFF in CMake-GUI. However, the console > > window still appears. > > > > I would appreciate any help. > > > > Best, > > Mujassim > > > > > > _______________________________________________ > > mitk-users mailing list > > mit...@li... > > https://lists.sourceforge.net/lists/listinfo/mitk-users |
|
From: Mujassim J. <muj...@gm...> - 2025-04-29 06:11:59
|
Thanks for pointing that out. I tried changing it, and now it does show "Windows (/SUBSYSTEM:WINDOWS)" in the SubSystem property. However, I'm now getting another error: unresolved external symbol WinMain referenced in function "int __cdecl __scrt_common_main_seh(void)" (?__scrt_common_main_seh@@YAHXZ). On Tue, 29 Apr 2025 at 10:21, Dinkelacker, Stefan <s.d...@dk...> wrote: > > It is an MITK-build setting, not a superbuild setting. Make sure to set it while the MITK-build subfolder is selected as build folder in CMake GUI. Hence, you do not need to rebuild anything on superbuild level. > > ________________________________ > From: Mujassim Jamal <muj...@gm...> > Sent: Monday, April 28, 2025 11:12:20 AM > To: Dinkelacker, Stefan <s.d...@dk...> > Cc: mit...@li... <mit...@li...> > Subject: Re: [Extern] - [mitk-users] How to hide or disable the console window at startup > > Hi Stefan, > > I have built MITK on Windows 10. Here are the steps I followed, but > the console window still appears: > 1. In the Visual Studio IDE, under the MITK-build project's > CMakeLists.txt, I turned OFF MITK_SHOW_CONSOLE_WINDOW. > 2. Opened CMake-GUI and regenerated the CMake files by clicking the > Generate button. > 3. Opened the MITK-superbuild.sln from CMake-GUI itself by clicking > the Open Project button. > 4. Rebuilt the solution, then opened MITK.sln and ran the Workbench, > but the console window still appears. > > The SubSystem property still shows Console (/SUBSYSTEM:CONSOLE). When > I explicitly printed the value of MITK_SHOW_CONSOLE_WINDOW, it shows > ON even though it is set to OFF. Maybe it is related to a CMake cache > issue? > > Best, > Mujassim > > On Mon, 28 Apr 2025 at 00:47, Dinkelacker, Stefan > <s.d...@dk...> wrote: > > > > Hi, > > > > I just switched off MITK_SHOW_CONSOLE_WINDOW to OFF in my develop branch MITK-build and the console window disappeared. Which OS and version are you using? Did you regenerate the CMake files after switching it off? > > > > It is supposed to influence the "Linker > System > SubSystem" project property of the MitkWorkbench project in the MITK solution. Instead of "Console (/SUBSYSTEM:CONSOLE)" it should display "Windows (/SUBSYSTEM:WINDOWS)". > > > > Best, > > Stefan > > ________________________________________ > > Von: Mujassim Jamal <muj...@gm...> > > Gesendet: Sonntag, 27. April 2025 06:19 > > An: mit...@li... > > Betreff: [Extern] - [mitk-users] How to hide or disable the console window at startup > > > > Hi again, > > > > I want to hide or disable the console window that appears at the > > startup of MitkWorkbench. I tried building MITK by setting > > MITK_SHOW_CONSOLE_WINDOW to OFF in CMake-GUI. However, the console > > window still appears. > > > > I would appreciate any help. > > > > Best, > > Mujassim > > > > > > _______________________________________________ > > mitk-users mailing list > > mit...@li... > > https://lists.sourceforge.net/lists/listinfo/mitk-users |
|
From: Dinkelacker, S. <s.d...@dk...> - 2025-04-29 04:51:48
|
It is an MITK-build setting, not a superbuild setting. Make sure to set it while the MITK-build subfolder is selected as build folder in CMake GUI. Hence, you do not need to rebuild anything on superbuild level. ________________________________ From: Mujassim Jamal <muj...@gm...> Sent: Monday, April 28, 2025 11:12:20 AM To: Dinkelacker, Stefan <s.d...@dk...> Cc: mit...@li... <mit...@li...> Subject: Re: [Extern] - [mitk-users] How to hide or disable the console window at startup Hi Stefan, I have built MITK on Windows 10. Here are the steps I followed, but the console window still appears: 1. In the Visual Studio IDE, under the MITK-build project's CMakeLists.txt, I turned OFF MITK_SHOW_CONSOLE_WINDOW. 2. Opened CMake-GUI and regenerated the CMake files by clicking the Generate button. 3. Opened the MITK-superbuild.sln from CMake-GUI itself by clicking the Open Project button. 4. Rebuilt the solution, then opened MITK.sln and ran the Workbench, but the console window still appears. The SubSystem property still shows Console (/SUBSYSTEM:CONSOLE). When I explicitly printed the value of MITK_SHOW_CONSOLE_WINDOW, it shows ON even though it is set to OFF. Maybe it is related to a CMake cache issue? Best, Mujassim On Mon, 28 Apr 2025 at 00:47, Dinkelacker, Stefan <s.d...@dk...> wrote: > > Hi, > > I just switched off MITK_SHOW_CONSOLE_WINDOW to OFF in my develop branch MITK-build and the console window disappeared. Which OS and version are you using? Did you regenerate the CMake files after switching it off? > > It is supposed to influence the "Linker > System > SubSystem" project property of the MitkWorkbench project in the MITK solution. Instead of "Console (/SUBSYSTEM:CONSOLE)" it should display "Windows (/SUBSYSTEM:WINDOWS)". > > Best, > Stefan > ________________________________________ > Von: Mujassim Jamal <muj...@gm...> > Gesendet: Sonntag, 27. April 2025 06:19 > An: mit...@li... > Betreff: [Extern] - [mitk-users] How to hide or disable the console window at startup > > Hi again, > > I want to hide or disable the console window that appears at the > startup of MitkWorkbench. I tried building MITK by setting > MITK_SHOW_CONSOLE_WINDOW to OFF in CMake-GUI. However, the console > window still appears. > > I would appreciate any help. > > Best, > Mujassim > > > _______________________________________________ > mitk-users mailing list > mit...@li... > https://lists.sourceforge.net/lists/listinfo/mitk-users |
|
From: Mujassim J. <muj...@gm...> - 2025-04-28 09:13:00
|
Hi Stefan, I have built MITK on Windows 10. Here are the steps I followed, but the console window still appears: 1. In the Visual Studio IDE, under the MITK-build project's CMakeLists.txt, I turned OFF MITK_SHOW_CONSOLE_WINDOW. 2. Opened CMake-GUI and regenerated the CMake files by clicking the Generate button. 3. Opened the MITK-superbuild.sln from CMake-GUI itself by clicking the Open Project button. 4. Rebuilt the solution, then opened MITK.sln and ran the Workbench, but the console window still appears. The SubSystem property still shows Console (/SUBSYSTEM:CONSOLE). When I explicitly printed the value of MITK_SHOW_CONSOLE_WINDOW, it shows ON even though it is set to OFF. Maybe it is related to a CMake cache issue? Best, Mujassim On Mon, 28 Apr 2025 at 00:47, Dinkelacker, Stefan <s.d...@dk...> wrote: > > Hi, > > I just switched off MITK_SHOW_CONSOLE_WINDOW to OFF in my develop branch MITK-build and the console window disappeared. Which OS and version are you using? Did you regenerate the CMake files after switching it off? > > It is supposed to influence the "Linker > System > SubSystem" project property of the MitkWorkbench project in the MITK solution. Instead of "Console (/SUBSYSTEM:CONSOLE)" it should display "Windows (/SUBSYSTEM:WINDOWS)". > > Best, > Stefan > ________________________________________ > Von: Mujassim Jamal <muj...@gm...> > Gesendet: Sonntag, 27. April 2025 06:19 > An: mit...@li... > Betreff: [Extern] - [mitk-users] How to hide or disable the console window at startup > > Hi again, > > I want to hide or disable the console window that appears at the > startup of MitkWorkbench. I tried building MITK by setting > MITK_SHOW_CONSOLE_WINDOW to OFF in CMake-GUI. However, the console > window still appears. > > I would appreciate any help. > > Best, > Mujassim > > > _______________________________________________ > mitk-users mailing list > mit...@li... > https://lists.sourceforge.net/lists/listinfo/mitk-users |
|
From: Dinkelacker, S. <s.d...@dk...> - 2025-04-27 19:17:56
|
Hi, I just switched off MITK_SHOW_CONSOLE_WINDOW to OFF in my develop branch MITK-build and the console window disappeared. Which OS and version are you using? Did you regenerate the CMake files after switching it off? It is supposed to influence the "Linker > System > SubSystem" project property of the MitkWorkbench project in the MITK solution. Instead of "Console (/SUBSYSTEM:CONSOLE)" it should display "Windows (/SUBSYSTEM:WINDOWS)". Best, Stefan ________________________________________ Von: Mujassim Jamal <muj...@gm...> Gesendet: Sonntag, 27. April 2025 06:19 An: mit...@li... Betreff: [Extern] - [mitk-users] How to hide or disable the console window at startup Hi again, I want to hide or disable the console window that appears at the startup of MitkWorkbench. I tried building MITK by setting MITK_SHOW_CONSOLE_WINDOW to OFF in CMake-GUI. However, the console window still appears. I would appreciate any help. Best, Mujassim _______________________________________________ mitk-users mailing list mit...@li... https://lists.sourceforge.net/lists/listinfo/mitk-users |
|
From: Mujassim J. <muj...@gm...> - 2025-04-27 04:19:55
|
Hi again, I want to hide or disable the console window that appears at the startup of MitkWorkbench. I tried building MITK by setting MITK_SHOW_CONSOLE_WINDOW to OFF in CMake-GUI. However, the console window still appears. I would appreciate any help. Best, Mujassim |
|
From: Mujassim J. <muj...@gm...> - 2025-04-20 05:03:32
|
Thanks Stefan, that really helped.
Best,
Mujassim
On Thu, 17 Apr 2025 at 13:48, Dinkelacker, Stefan
<s.d...@dk...> wrote:
>
> Hi,
>
> as far as I know we do not support direct loading from Qt resources via IOUtil. If you want to load a resource you can use our own resource system, though:
>
> - Use one of the overloads of Load() that takes a us::ModuleResource instead of a path, e. g.:
>
> auto myStl = IOUtil::Load<Surface>(resource); // myStl is a mitk::Surface::Pointer
>
> - Where resource (us::ModuleResource) is typically loaded like:
>
> #include <usGetModuleContext.h>
> #include <usModuleResource.h>
>
> auto resource = us::GetModuleContext()->GetModule()->GetResource("my_stl.stl");
>
> If you call GetModule() from another module that does not contain the resource, you need to pass the name of the module to GetModule() like GetModule("MitkMyModule").
>
> To have the resource in your module in the first place, add the following to your module's files.cmake:
>
> set(RESOURCE_FILES my_stl.stl) # Or, if there are already resources, add my_stl.stl to the list.
>
> The file is expected to be located in "Resources/my_stl.stl".
>
> You find plenty of examples in the MitkSegmentation module.
>
> Best,
>
> Stefan
> ________________________________________
> Von: Mujassim Jamal <muj...@gm...>
> Gesendet: Dienstag, 15. April 2025 06:45
> An: mit...@li...
> Betreff: [Extern] - [mitk-users] Loading STL from Plugin Resource
>
> Hello again,
>
> I’ve created a plugin and would like to load an STL file located in
> the 'resources/stl' folder within the plugin. I'm attempting to load
> the STL by passing the resource path (e.g.,
> :/org.mitk.customplugin/stl/my_stl.stl) to
> 'mitk::IOUtil::Load<mitk::Surface>'.
>
> However, I'm getting the following error:
> "Description File: :/org.mitk.customplugin/stl/my_stl.stl does not exist"
>
> The strange part is that when I check the resource using
> 'QFile::exists', it confirms that the file exists. Could this be an
> issue with 'IOUtil::Load' not supporting Qt resource file paths?
>
> Thanks for any help!
>
> Best,
> Mujassim
>
>
> _______________________________________________
> mitk-users mailing list
> mit...@li...
> https://lists.sourceforge.net/lists/listinfo/mitk-users
|
|
From: Dinkelacker, S. <s.d...@dk...> - 2025-04-17 08:18:49
|
Hi,
as far as I know we do not support direct loading from Qt resources via IOUtil. If you want to load a resource you can use our own resource system, though:
- Use one of the overloads of Load() that takes a us::ModuleResource instead of a path, e. g.:
auto myStl = IOUtil::Load<Surface>(resource); // myStl is a mitk::Surface::Pointer
- Where resource (us::ModuleResource) is typically loaded like:
#include <usGetModuleContext.h>
#include <usModuleResource.h>
auto resource = us::GetModuleContext()->GetModule()->GetResource("my_stl.stl");
If you call GetModule() from another module that does not contain the resource, you need to pass the name of the module to GetModule() like GetModule("MitkMyModule").
To have the resource in your module in the first place, add the following to your module's files.cmake:
set(RESOURCE_FILES my_stl.stl) # Or, if there are already resources, add my_stl.stl to the list.
The file is expected to be located in "Resources/my_stl.stl".
You find plenty of examples in the MitkSegmentation module.
Best,
Stefan
________________________________________
Von: Mujassim Jamal <muj...@gm...>
Gesendet: Dienstag, 15. April 2025 06:45
An: mit...@li...
Betreff: [Extern] - [mitk-users] Loading STL from Plugin Resource
Hello again,
I’ve created a plugin and would like to load an STL file located in
the 'resources/stl' folder within the plugin. I'm attempting to load
the STL by passing the resource path (e.g.,
:/org.mitk.customplugin/stl/my_stl.stl) to
'mitk::IOUtil::Load<mitk::Surface>'.
However, I'm getting the following error:
"Description File: :/org.mitk.customplugin/stl/my_stl.stl does not exist"
The strange part is that when I check the resource using
'QFile::exists', it confirms that the file exists. Could this be an
issue with 'IOUtil::Load' not supporting Qt resource file paths?
Thanks for any help!
Best,
Mujassim
_______________________________________________
mitk-users mailing list
mit...@li...
https://lists.sourceforge.net/lists/listinfo/mitk-users
|
|
From: Mujassim J. <muj...@gm...> - 2025-04-15 04:46:37
|
Hello again, I’ve created a plugin and would like to load an STL file located in the 'resources/stl' folder within the plugin. I'm attempting to load the STL by passing the resource path (e.g., :/org.mitk.customplugin/stl/my_stl.stl) to 'mitk::IOUtil::Load<mitk::Surface>'. However, I'm getting the following error: "Description File: :/org.mitk.customplugin/stl/my_stl.stl does not exist" The strange part is that when I check the resource using 'QFile::exists', it confirms that the file exists. Could this be an issue with 'IOUtil::Load' not supporting Qt resource file paths? Thanks for any help! Best, Mujassim |
|
From: Mujassim J. <muj...@gm...> - 2025-04-13 11:42:02
|
Hi Stefan, Ah, my mistake — thanks for clarifying, and I really appreciate the guidance! Best, Mujassim On Tue, 8 Apr 2025 at 11:37, Dinkelacker, Stefan <s.d...@dk...> wrote: > > Hi Mujassim, > > I think there's been a mix-up — we're not actually working on this ourselves, just offering some guidance on how you might proceed. > > Best, > Stefan > ________________________________________ > Von: Mujassim Jamal <muj...@gm...> > Gesendet: Donnerstag, 3. April 2025 05:43 > An: Dinkelacker, Stefan > Cc: mit...@li... > Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK > > Hi Stefan, > > I just wanted to check if there are any updates on its implementation. No rush—just curious to know how it's progressing. > > Best, > Mujassim > > > On Mon, 24 Mar 2025 at 13:52, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...>> wrote: > > Hi Mujassim, > > yes, an approach in that direction is reasonable. I think I would write a function that traverses from the hover point in both directions along the contour for a certain distance D (should be easy since two points are always connected by a straight line). Each contour vertex that is within D I would pair with a weight like a vertex at the hover point would have weight 1, a vertex in distance D/2 would have weight 0.5 and a vertex in distance D would have weight 0. The returned list of pairs (probably made up of vertex index and weight) I would use to decide on which points the move operation has influence on and how much. The weight can also be directly used to calculate a color for rendering like in my example picture. > > There's an edge case for tiny contours, though, when basically more or less all points are within distance D. So probably I would also take the contour's complete circumferrence into account to decide if D should be decreased to never cover more than a certain percentage of a contour's circumferrence, or simething similar. A strategy for this is greatly dependent on the contour's you are expecting in the first place, like will they have dense or sparse vertices, what is the typical ratio of distance between vertices and the whole circumferrence etc... > > Best, > Stefan > > ________________________________________ > Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...>> > Gesendet: Freitag, 21. März 2025 12:21 > An: Dinkelacker, Stefan > Cc: mit...@li...<mailto:mit...@li...> > Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK > > Hi Stefan, > > While implementing the custom mapper for neighbor highlighting, I realized that I first need to identify the neighboring points before I can highlight them. This means that creating and identifying neighbor points within the contour itself is an essential prerequisite. > > It might be more effective to focus on neighbor detection first before moving on to the rendering and appearance aspects. For reference, I have attached a picture showing the contour I am working on, which consists of six vertices. Here’s the basic approach I have in mind: > > 1. A position is hovered over on the contour (marked in red at vertex 1 or between vertices 5 and 4). > 2. If a vertex is found at that position, proceed to the next step; otherwise, first add a point at that location. > 3. The algorithm will then attempt to find neighboring points on both sides of the hovered position within a specified radius (marked in green). > 4. If no neighboring points are found, 'X' number of points will be added in both directions of the hovered position within the radius. > 5. Finally, the identified points to be highlighted will be passed to the rendering process. > > Does this approach make sense to you? I’d appreciate your thoughts on it. > > Best, > Mujassim > > > On Mon, 17 Mar 2025 at 19:36, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>> wrote: > Hi Mujassim, > > interactors typically just change state/properties of data. The transformation into render primitives happens in mapper classes, which (re-)generate the VTK (or OpenGL in legacy cases) primitives on any data/property updates. Long story short: if you want to change anything regarding the appearance of data, you would need to write (potentially derive) your own mapper. Unfortunately mappers in the regime of PlanarFigures or ContourModelSets are, let's say, part of a technical dept that were never resolved, and while they are working, they shouldn't probably be used as examples for new implementations. You still can do so, of course, but in general new mappers should be derived from mitk::VtkMapper. > > Best regards, > Stefan > ________________________________________ > Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>> > Gesendet: Montag, 17. März 2025 13:10 > An: Dinkelacker, Stefan > Cc: mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>> > Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK > > Hi Stefan, > > Yes, the Lasso tool better aligns with my needs. I tested ContourModelInteractor, set up the MITK-ProjectTemplate, created a custom interactor class inheriting from ContourModelInteractor, and implemented a function to enable interaction on the saved contour node, following your instructions. I also prefer this approach from scratch rather than the vtkCutter approach you mentioned in your other post. > > Before implementing the OnMovePoint function, I plan to first develop a neighbor highlighting feature for the contour. This feature will highlight the neighboring areas of a point when hovered over with the mouse. Users will also be able to adjust the highlighted area, allowing modifications to only that portion of the contour rather than the entire area between two points (which I believe is the default behavior). Currently, ContourModelInteractor highlights the entire contour when hovered over. > > Could you share your thoughts on this approach? > > Best regards, > Mujassim > > On Fri, 21 Feb 2025 at 12:14, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>>> wrote: > Hi Mujassim, > > let's assume you would start with the lasso tool [1] instead of the livewire tool, which uses the same base class but seems to fit your needs even better. The base class EditableContourTool [2] sets up the state machine "EditableContourTool.xml" for the creation of the contour and connects most of the functions. The derived LassoTool class then only connects the MovePoint action in addition. But all of this interaction is for the creation of the contour, not for editing. In LassoTool::FinishTool() you will see, however, that another DataInteractor/StateMachine is associated with the created contour data node: ContourModelInteractor [3] with ContourModelModificationInteractor.xml as state machine. In this class you find the OnMovePoint() method which is called when moving a point of the contour. That's where you want to hook in to add your functionality. You can do this for example by subclassing ContourModelInteractor and overriding the OnMovePoint() method. > > To understand/test the ContourModelInteractor in the MITK Workbench first, activate the Segmentation plugin, create a segmentation, select the Lasso tool from the 2D tools and remove the checkmark from "Auto-confirm contour". This is crucial. Start a contour with a double click, add points with single clicks and add the last point again with a double click. Now the editable contour should still be on the screen and you can test the ContourModelnteractor by moving an existing point of the contour for example. > > To start with such a contour right from the beginning I recommend to save the contour so you can simply open it later without having to use the segmentation plugin. The contour node from the lasso tool is invisible by default, though. To make it visible, open the preferences (Ctrl+P), click on Data Manager on the very top and set the checkmark for "Show helper objects". While the Lasso tool is active with an editable contour, you will see a "working contour node" node in the Data Manager now. You can right-click on it and "Save..." it to a .cnt file, that you can later open again. > > When you implement your subclass you can activate the interaction by calling either DataInteractor::SetDataNode() or DataNode::SetDataInteractor() on a contour node. A good starting point for that is probably your own plugin/view and to wire up a button to do so. Start best with the MITK-ProjectTemplate [4] where you could modify the Example View for your first prototype [5]. > > [1] https://docs.mitk.org/2024.12/classmitk_1_1LassoTool.html > [2] https://docs.mitk.org/2024.12/classmitk_1_1EditableContourTool.html > [3] https://docs.mitk.org/2024.12/classmitk_1_1ContourModelInteractor.html > [4] https://github.com/MITK/MITK-ProjectTemplate/blob/master/README.md > [5] https://github.com/MITK/MITK-ProjectTemplate/blob/master/Plugins/org.mitk.gui.qt.exampleplugin/src/internal/QmitkExampleView.cpp > > Best, > Stefan > ________________________________________ > Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>>> > Gesendet: Dienstag, 18. Februar 2025 13:41 > An: Dinkelacker, Stefan > Cc: mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>> > Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK > > Hi Stefan, > > I followed your instructions and explored the LiveWire tool, which helped me understand how it leverages ContourModelLiveWireInteractor for various interactions. I also reviewed its corresponding state machine and configuration XML files, which were quite easy to follow. > > Next, I traced the inheritance hierarchy of ContourModelLiveWireInteractor up to DataInteractor, where each class has a ConnectActionsAndFunctions function responsible for linking state machine actions to interactor functions (as you mentioned in an earlier response). The description of the DataInteractor class also states that new interactor classes can be derived from it. > > From this point, what would you recommend as my next step? Should I proceed with creating a new interactor and its associated XML files, or is there anything else I should refer to before diving into the implementation? > > Looking forward to your guidance. > > Best, > Mujassim > > On Tue, 11 Feb 2025 at 15:33, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>>>> wrote: > Hi Mujassim, > > I think at this stage an example-based exploration is probably the most efficient way of going forward. For the Live Wire tool, see [1] for example and move up the inheritance until you find the code using an Interactor. An Interactor is connected to a state machine and a separate config for that statemachine. Both are written as .xml files. The Interactor code connects actions and conditions of a statemachine to its methods. > > Best, > Stefan > > [1] https://docs.mitk.org/2024.12/classmitk_1_1LiveWireTool2D.html > ________________________________________ > Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>>>> > Gesendet: Dienstag, 11. Februar 2025 04:46 > An: Dinkelacker, Stefan > Cc: mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>>> > Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK > > Hi, > > Yes, I previously built MITK v2023.12, but I encountered several compiler errors related to ITK (conversion type errors) and POCO (OpenSSL version mismatch). > > Thank you for suggesting these four tools. I have explored all of them and found that the Draw Polygon and LiveWire tools closely match my use case, though I will still need to make significant customizations. I plan to refer to the code of these two tools to get an initial understanding of point adding, dragging, picking, etc. > > I would appreciate it if you could share any resources on developing a custom interactor in MITK. > > Best, > Mujassim > > On Wed, 5 Feb 2025 at 17:28, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>>>>> wrote: > Hi and welcome to MITK! > > The last version of MITK that supported Qt 5 was MITK v2023.12, as far as I know. Starting with later versions, Qt 6.6+ is a strict requirement. > > To get started, I recommend running the MitkWorkbench application and exploring the contour interaction functionalities available in two different plugins: > > - Measurement [1]: The Draw Polygon and Draw Subdivision Polygon tools demonstrate how contours can be interacted with. > > - Segmentation [2]: The Lasso and LiveWire tools are both based on contour interaction. > > MITK provides at least two approaches for working with contours and interacting with them. However, you will need to write some code—for example, handling a Modified event to map contour changes back to a 3D model (referred to as a Surface in MITK). Additionally, MITK's current interactors only operate on individual contour vertices. > > If you aim to implement an approach similar to what you showed in your video—where multiple neighboring vertices of a contour are modified simultaneously—you will need to develop a custom interactor. While this is certainly possible, it is not the simplest task for a first MITK project. > > Best, > Stefan > > [1] https://docs.mitk.org/2024.12/org_mitk_views_measurement.html > [2] https://docs.mitk.org/2024.12/org_mitk_views_segmentation.html#org_mitk_views_segmentationlivewiretool > > ________________________________________ > Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>>>>> > Gesendet: Mittwoch, 5. Februar 2025 11:27 > An: mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>>>> > Betreff: [Extern] - [mitk-users] Contour Editing in MITK > > Hi, > > First of all, I am new to using MITK. I am working with Qt 5.15.2, but I couldn't find an MITK version compatible with this Qt version. Therefore, I built the latest version of MITK (v2024.12). > > My question is: Can I use the MITK toolkit to create a contour editing feature such that, after projecting a 3D model as a contour onto the 2D segmentation, I can pick and drag the points on the contour in 2D and see the effect directly on the 3D model? I found that MITK provides various contour-related classes, but I am unsure whether they can be used for my specific use case. > > I am attaching a Google Drive link that contains a video showcasing what I want to achieve using MITK. > Video: https://drive.google.com/file/d/1Wu3UJmQQeewlR0Thvx7TPP0iYLEmfu1J/view?usp=sharing > > Thank you. |
|
From: Mujassim J. <muj...@gm...> - 2025-04-13 11:39:14
|
Hi Stefan, Thanks a lot for the clarification, that really helps. I’ll wait for the new Python backend update, sounds exciting! Best, Mujassim On Tue, 8 Apr 2025 at 11:44, Dinkelacker, Stefan <s.d...@dk...> wrote: > > Hi again, > > since Python3 also affects the configuration/builds of a few third-party dependencies of MITK, you should configure and generate the MITK-superbuild with MITK_USE_Python3 instead of just the MITK-build. This will build the missing CTK libraries for example but also even more important changes and additions in third-party libraries like SWIG. That being said, the state of Python support in our develop and master branch is highly experimental and old. In fact, we are rewriting our whole Python backend right now and basically everything will change completely. I would advise to wait a few weeks in that regard if you can do so. > > Best, > Stefan > ________________________________________ > Von: Mujassim Jamal <muj...@gm...> > Gesendet: Samstag, 5. April 2025 07:09 > An: mit...@li... > Betreff: [Extern] - [mitk-users] Issue Enabling Python3 Support in MITK 2024.12 > > Hello, > > I have successfully built MITK 2024.12 without Python3. Now, I have decided to reconfigure it to include Python3 support. I followed the official documentation, which instructs setting the variable 'MITK_USE_Python3' in CMake-GUI. I enabled this option and reconfigured the project. It appears that CMake successfully detected Python along with the required modules. Here is the relevant log message from CMake-GUI during reconfiguration: > ''' > Found Python3: C:/Users/mj/AppData/Local/Programs/Python/Python313/python.exe (found suitable version "3.13.1", minimum required is "3.8") found components: Interpreter Development NumPy Development.Module Development.Embed > ''' > > After that, I generated the project and opened it in Visual Studio, but I encountered the following error: > ''' > 29>CMake Error at CMake/mitkFunctionCreatePlugin.cmake:80 (message): > 29> org_mitk_gui_qt_python is missing requirements and won't be built. > 29> Missing: MitkQtPython > 29>Call Stack (most recent call first): > 29> Plugins/org.mitk.gui.qt.python/CMakeLists.txt:5 (mitk_create_plugin) > 29> > ''' > > In the 'FEATURE SUMMARY FOR MITK' section of the build log, it shows that 'MitkPython' is enabled, but 'MitkQtPython' is missing. > > After debugging, I discovered that even though 'MITK_USE_Python3' is enabled, the variables 'CTKScriptingPythonCore_INCLUDE_DIRS' and 'CTKScriptingPythonWidgets_INCLUDE_DIRS' in Modules/QtPython/CMakeLists.txt appear to be missing. To verify this, I added a print statement inside the if condition that checks for these variables, but it was never logged during the build process. > > I believe this is likely the reason why 'MitkQtPython' is not being built and is missing. > > I would greatly appreciate any help on this matter. > > Best, > Mujassim > > |
|
From: Dinkelacker, S. <s.d...@dk...> - 2025-04-08 06:14:37
|
Hi again, since Python3 also affects the configuration/builds of a few third-party dependencies of MITK, you should configure and generate the MITK-superbuild with MITK_USE_Python3 instead of just the MITK-build. This will build the missing CTK libraries for example but also even more important changes and additions in third-party libraries like SWIG. That being said, the state of Python support in our develop and master branch is highly experimental and old. In fact, we are rewriting our whole Python backend right now and basically everything will change completely. I would advise to wait a few weeks in that regard if you can do so. Best, Stefan ________________________________________ Von: Mujassim Jamal <muj...@gm...> Gesendet: Samstag, 5. April 2025 07:09 An: mit...@li... Betreff: [Extern] - [mitk-users] Issue Enabling Python3 Support in MITK 2024.12 Hello, I have successfully built MITK 2024.12 without Python3. Now, I have decided to reconfigure it to include Python3 support. I followed the official documentation, which instructs setting the variable 'MITK_USE_Python3' in CMake-GUI. I enabled this option and reconfigured the project. It appears that CMake successfully detected Python along with the required modules. Here is the relevant log message from CMake-GUI during reconfiguration: ''' Found Python3: C:/Users/mj/AppData/Local/Programs/Python/Python313/python.exe (found suitable version "3.13.1", minimum required is "3.8") found components: Interpreter Development NumPy Development.Module Development.Embed ''' After that, I generated the project and opened it in Visual Studio, but I encountered the following error: ''' 29>CMake Error at CMake/mitkFunctionCreatePlugin.cmake:80 (message): 29> org_mitk_gui_qt_python is missing requirements and won't be built. 29> Missing: MitkQtPython 29>Call Stack (most recent call first): 29> Plugins/org.mitk.gui.qt.python/CMakeLists.txt:5 (mitk_create_plugin) 29> ''' In the 'FEATURE SUMMARY FOR MITK' section of the build log, it shows that 'MitkPython' is enabled, but 'MitkQtPython' is missing. After debugging, I discovered that even though 'MITK_USE_Python3' is enabled, the variables 'CTKScriptingPythonCore_INCLUDE_DIRS' and 'CTKScriptingPythonWidgets_INCLUDE_DIRS' in Modules/QtPython/CMakeLists.txt appear to be missing. To verify this, I added a print statement inside the if condition that checks for these variables, but it was never logged during the build process. I believe this is likely the reason why 'MitkQtPython' is not being built and is missing. I would greatly appreciate any help on this matter. Best, Mujassim |
|
From: Dinkelacker, S. <s.d...@dk...> - 2025-04-08 06:07:59
|
Hi Mujassim, I think there's been a mix-up — we're not actually working on this ourselves, just offering some guidance on how you might proceed. Best, Stefan ________________________________________ Von: Mujassim Jamal <muj...@gm...> Gesendet: Donnerstag, 3. April 2025 05:43 An: Dinkelacker, Stefan Cc: mit...@li... Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK Hi Stefan, I just wanted to check if there are any updates on its implementation. No rush—just curious to know how it's progressing. Best, Mujassim On Mon, 24 Mar 2025 at 13:52, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...>> wrote: Hi Mujassim, yes, an approach in that direction is reasonable. I think I would write a function that traverses from the hover point in both directions along the contour for a certain distance D (should be easy since two points are always connected by a straight line). Each contour vertex that is within D I would pair with a weight like a vertex at the hover point would have weight 1, a vertex in distance D/2 would have weight 0.5 and a vertex in distance D would have weight 0. The returned list of pairs (probably made up of vertex index and weight) I would use to decide on which points the move operation has influence on and how much. The weight can also be directly used to calculate a color for rendering like in my example picture. There's an edge case for tiny contours, though, when basically more or less all points are within distance D. So probably I would also take the contour's complete circumferrence into account to decide if D should be decreased to never cover more than a certain percentage of a contour's circumferrence, or simething similar. A strategy for this is greatly dependent on the contour's you are expecting in the first place, like will they have dense or sparse vertices, what is the typical ratio of distance between vertices and the whole circumferrence etc... Best, Stefan ________________________________________ Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...>> Gesendet: Freitag, 21. März 2025 12:21 An: Dinkelacker, Stefan Cc: mit...@li...<mailto:mit...@li...> Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK Hi Stefan, While implementing the custom mapper for neighbor highlighting, I realized that I first need to identify the neighboring points before I can highlight them. This means that creating and identifying neighbor points within the contour itself is an essential prerequisite. It might be more effective to focus on neighbor detection first before moving on to the rendering and appearance aspects. For reference, I have attached a picture showing the contour I am working on, which consists of six vertices. Here’s the basic approach I have in mind: 1. A position is hovered over on the contour (marked in red at vertex 1 or between vertices 5 and 4). 2. If a vertex is found at that position, proceed to the next step; otherwise, first add a point at that location. 3. The algorithm will then attempt to find neighboring points on both sides of the hovered position within a specified radius (marked in green). 4. If no neighboring points are found, 'X' number of points will be added in both directions of the hovered position within the radius. 5. Finally, the identified points to be highlighted will be passed to the rendering process. Does this approach make sense to you? I’d appreciate your thoughts on it. Best, Mujassim On Mon, 17 Mar 2025 at 19:36, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>> wrote: Hi Mujassim, interactors typically just change state/properties of data. The transformation into render primitives happens in mapper classes, which (re-)generate the VTK (or OpenGL in legacy cases) primitives on any data/property updates. Long story short: if you want to change anything regarding the appearance of data, you would need to write (potentially derive) your own mapper. Unfortunately mappers in the regime of PlanarFigures or ContourModelSets are, let's say, part of a technical dept that were never resolved, and while they are working, they shouldn't probably be used as examples for new implementations. You still can do so, of course, but in general new mappers should be derived from mitk::VtkMapper. Best regards, Stefan ________________________________________ Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>> Gesendet: Montag, 17. März 2025 13:10 An: Dinkelacker, Stefan Cc: mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>> Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK Hi Stefan, Yes, the Lasso tool better aligns with my needs. I tested ContourModelInteractor, set up the MITK-ProjectTemplate, created a custom interactor class inheriting from ContourModelInteractor, and implemented a function to enable interaction on the saved contour node, following your instructions. I also prefer this approach from scratch rather than the vtkCutter approach you mentioned in your other post. Before implementing the OnMovePoint function, I plan to first develop a neighbor highlighting feature for the contour. This feature will highlight the neighboring areas of a point when hovered over with the mouse. Users will also be able to adjust the highlighted area, allowing modifications to only that portion of the contour rather than the entire area between two points (which I believe is the default behavior). Currently, ContourModelInteractor highlights the entire contour when hovered over. Could you share your thoughts on this approach? Best regards, Mujassim On Fri, 21 Feb 2025 at 12:14, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>>> wrote: Hi Mujassim, let's assume you would start with the lasso tool [1] instead of the livewire tool, which uses the same base class but seems to fit your needs even better. The base class EditableContourTool [2] sets up the state machine "EditableContourTool.xml" for the creation of the contour and connects most of the functions. The derived LassoTool class then only connects the MovePoint action in addition. But all of this interaction is for the creation of the contour, not for editing. In LassoTool::FinishTool() you will see, however, that another DataInteractor/StateMachine is associated with the created contour data node: ContourModelInteractor [3] with ContourModelModificationInteractor.xml as state machine. In this class you find the OnMovePoint() method which is called when moving a point of the contour. That's where you want to hook in to add your functionality. You can do this for example by subclassing ContourModelInteractor and overriding the OnMovePoint() method. To understand/test the ContourModelInteractor in the MITK Workbench first, activate the Segmentation plugin, create a segmentation, select the Lasso tool from the 2D tools and remove the checkmark from "Auto-confirm contour". This is crucial. Start a contour with a double click, add points with single clicks and add the last point again with a double click. Now the editable contour should still be on the screen and you can test the ContourModelnteractor by moving an existing point of the contour for example. To start with such a contour right from the beginning I recommend to save the contour so you can simply open it later without having to use the segmentation plugin. The contour node from the lasso tool is invisible by default, though. To make it visible, open the preferences (Ctrl+P), click on Data Manager on the very top and set the checkmark for "Show helper objects". While the Lasso tool is active with an editable contour, you will see a "working contour node" node in the Data Manager now. You can right-click on it and "Save..." it to a .cnt file, that you can later open again. When you implement your subclass you can activate the interaction by calling either DataInteractor::SetDataNode() or DataNode::SetDataInteractor() on a contour node. A good starting point for that is probably your own plugin/view and to wire up a button to do so. Start best with the MITK-ProjectTemplate [4] where you could modify the Example View for your first prototype [5]. [1] https://docs.mitk.org/2024.12/classmitk_1_1LassoTool.html [2] https://docs.mitk.org/2024.12/classmitk_1_1EditableContourTool.html [3] https://docs.mitk.org/2024.12/classmitk_1_1ContourModelInteractor.html [4] https://github.com/MITK/MITK-ProjectTemplate/blob/master/README.md [5] https://github.com/MITK/MITK-ProjectTemplate/blob/master/Plugins/org.mitk.gui.qt.exampleplugin/src/internal/QmitkExampleView.cpp Best, Stefan ________________________________________ Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>>> Gesendet: Dienstag, 18. Februar 2025 13:41 An: Dinkelacker, Stefan Cc: mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>> Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK Hi Stefan, I followed your instructions and explored the LiveWire tool, which helped me understand how it leverages ContourModelLiveWireInteractor for various interactions. I also reviewed its corresponding state machine and configuration XML files, which were quite easy to follow. Next, I traced the inheritance hierarchy of ContourModelLiveWireInteractor up to DataInteractor, where each class has a ConnectActionsAndFunctions function responsible for linking state machine actions to interactor functions (as you mentioned in an earlier response). The description of the DataInteractor class also states that new interactor classes can be derived from it. >From this point, what would you recommend as my next step? Should I proceed with creating a new interactor and its associated XML files, or is there anything else I should refer to before diving into the implementation? Looking forward to your guidance. Best, Mujassim On Tue, 11 Feb 2025 at 15:33, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>>>> wrote: Hi Mujassim, I think at this stage an example-based exploration is probably the most efficient way of going forward. For the Live Wire tool, see [1] for example and move up the inheritance until you find the code using an Interactor. An Interactor is connected to a state machine and a separate config for that statemachine. Both are written as .xml files. The Interactor code connects actions and conditions of a statemachine to its methods. Best, Stefan [1] https://docs.mitk.org/2024.12/classmitk_1_1LiveWireTool2D.html ________________________________________ Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>>>> Gesendet: Dienstag, 11. Februar 2025 04:46 An: Dinkelacker, Stefan Cc: mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>>> Betreff: Re: [Extern] - [mitk-users] Contour Editing in MITK Hi, Yes, I previously built MITK v2023.12, but I encountered several compiler errors related to ITK (conversion type errors) and POCO (OpenSSL version mismatch). Thank you for suggesting these four tools. I have explored all of them and found that the Draw Polygon and LiveWire tools closely match my use case, though I will still need to make significant customizations. I plan to refer to the code of these two tools to get an initial understanding of point adding, dragging, picking, etc. I would appreciate it if you could share any resources on developing a custom interactor in MITK. Best, Mujassim On Wed, 5 Feb 2025 at 17:28, Dinkelacker, Stefan <s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>><mailto:s.d...@dk...<mailto:s.d...@dk...><mailto:s.d...@dk...<mailto:s.d...@dk...>>>>>> wrote: Hi and welcome to MITK! The last version of MITK that supported Qt 5 was MITK v2023.12, as far as I know. Starting with later versions, Qt 6.6+ is a strict requirement. To get started, I recommend running the MitkWorkbench application and exploring the contour interaction functionalities available in two different plugins: - Measurement [1]: The Draw Polygon and Draw Subdivision Polygon tools demonstrate how contours can be interacted with. - Segmentation [2]: The Lasso and LiveWire tools are both based on contour interaction. MITK provides at least two approaches for working with contours and interacting with them. However, you will need to write some code—for example, handling a Modified event to map contour changes back to a 3D model (referred to as a Surface in MITK). Additionally, MITK's current interactors only operate on individual contour vertices. If you aim to implement an approach similar to what you showed in your video—where multiple neighboring vertices of a contour are modified simultaneously—you will need to develop a custom interactor. While this is certainly possible, it is not the simplest task for a first MITK project. Best, Stefan [1] https://docs.mitk.org/2024.12/org_mitk_views_measurement.html [2] https://docs.mitk.org/2024.12/org_mitk_views_segmentation.html#org_mitk_views_segmentationlivewiretool ________________________________________ Von: Mujassim Jamal <muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>><mailto:muj...@gm...<mailto:muj...@gm...><mailto:muj...@gm...<mailto:muj...@gm...>>>>>> Gesendet: Mittwoch, 5. Februar 2025 11:27 An: mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>><mailto:mit...@li...<mailto:mit...@li...><mailto:mit...@li...<mailto:mit...@li...>>>>> Betreff: [Extern] - [mitk-users] Contour Editing in MITK Hi, First of all, I am new to using MITK. I am working with Qt 5.15.2, but I couldn't find an MITK version compatible with this Qt version. Therefore, I built the latest version of MITK (v2024.12). My question is: Can I use the MITK toolkit to create a contour editing feature such that, after projecting a 3D model as a contour onto the 2D segmentation, I can pick and drag the points on the contour in 2D and see the effect directly on the 3D model? I found that MITK provides various contour-related classes, but I am unsure whether they can be used for my specific use case. I am attaching a Google Drive link that contains a video showcasing what I want to achieve using MITK. Video: https://drive.google.com/file/d/1Wu3UJmQQeewlR0Thvx7TPP0iYLEmfu1J/view?usp=sharing Thank you. |
|
From: Mujassim J. <muj...@gm...> - 2025-04-05 05:10:30
|
Hello, I have successfully built MITK 2024.12 without Python3. Now, I have decided to reconfigure it to include Python3 support. I followed the official documentation, which instructs setting the variable 'MITK_USE_Python3' in CMake-GUI. I enabled this option and reconfigured the project. It appears that CMake successfully detected Python along with the required modules. Here is the relevant log message from CMake-GUI during reconfiguration: ''' Found Python3: C:/Users/mj/AppData/Local/Programs/Python/Python313/python.exe (found suitable version "3.13.1", minimum required is "3.8") found components: Interpreter Development NumPy Development.Module Development.Embed ''' After that, I generated the project and opened it in Visual Studio, but I encountered the following error: ''' 29>CMake Error at CMake/mitkFunctionCreatePlugin.cmake:80 (message): 29> org_mitk_gui_qt_python is missing requirements and won't be built. 29> Missing: MitkQtPython 29>Call Stack (most recent call first): 29> Plugins/org.mitk.gui.qt.python/CMakeLists.txt:5 (mitk_create_plugin) 29> ''' In the *'*FEATURE SUMMARY FOR MITK*'* section of the build log, it shows that 'MitkPython' is enabled, but 'MitkQtPython' is missing. After debugging, I discovered that even though 'MITK_USE_Python3' is enabled, the variables 'CTKScriptingPythonCore_INCLUDE_DIRS' and 'CTKScriptingPythonWidgets_INCLUDE_DIRS' in Modules/QtPython/CMakeLists.txt appear to be missing. To verify this, I added a print statement inside the if condition that checks for these variables, but it was never logged during the build process. I believe this is likely the reason why 'MitkQtPython' is not being built and is missing. I would greatly appreciate any help on this matter. Best, Mujassim |
255 messages has been excluded from this view by a project administrator.
