rordenlab / mricrogl Goto Github PK

It would be very useful to be able to see live x,y,z coordinates of a mouse click in render view. Thanks!

Hi, I'm a big fan of the scripting with python function and have been using it to auto-generate images. However, I noticed that python 3.7 embedded in the Linux release of MRIcroGL has a quite limited number of packages. json, for one, is not available but is extremely useful to read json files, which contain parameters I'd like MRIcroGL to use when running built-in functions.

What would be the correct way to add packages to the embedded python library? Or would it be possible to keep the libraries consistent across MRIcroGL on different OS (e.g., the windows version has json but Linux doesn't )?

The latest release of MRIcroGL for macOS (v1.2.20211006) transitions from using the system Python 2.7 to a statically linked Python 3.7. This is in response to deprecation of Python 2.7 and Apple's requests that applications should not use the system Python. Some users have reported that scripting no longer works, though using previous versions works. Best guess is that this reflects people upgrading where the old installation set the PyLib preference to Python 2.7.

1. Does installing the v1.2.20211007 dmg resolve this?
2. Does using the -r argument from the terminal to reset the preferences resolve the issue? e.g. /Applications/MRIcroGL.app/Contents/MacOS/MRIcroGL -r

Dear Chris,

Would it be possible to include better handling for complex data types in MRIcroGL?

Recently someone reported that MRIcro couldn't load NIfTI-MRS formatted data (created by spec2nii), see spec2nii issue #23. From testing I believe that this arises from MRS data being complex. MRIcroGL reports 'unable to load "my_complex_data.nii.gz"', but will load the same file when the data is non-complex.

Your handling of a 4th time dimension means that (somewhat) meaningful viewing of MRS data can almost be handled. Would you consider loading and simply taking the absolute or real part of the data?

I've attached two very simple files that I tested with. These have 1x1x1x100 random real/complex float data, and in addition a header extension such as NIfTI-MRS uses.
n2_float_4dim_hdr.nii.gz
n2_cfloat_4dim_hdr.nii.gz

Dear MRIcroGL experts,

Summary:
I would like to obtain the MNI coordinates of the center of a cluster of particularly intensely bright voxels. I see that I can get the X,Y,Z coordinates of a cluster of particularly intensely bright voxels but would like to know how to get the MNI coordinates of that cluster. How can I do that?

More information:
I am working on a dataset of ~100 participants with lesions. I segmented and normalized my T1 images to MNI space, registered my native FLAIR images to my standardized T1 images so now the FLAIR images are also standardized, ran these images through the Lesion Segmentation Tool (LST) Lesion Growth Algorithm (LGA) pipeline (Schmidt et al., 2012), and obtained two output images: a version of the FLAIR and a lesion probability map. The lesion probability map is a binary map (each voxel is either white with an intensity of 1 indicating that voxel is a lesion or dark with an intensity of 0 indicating that voxel is not a lesion). Both files are .nii.

I open both of these images (the FLAIR as an image and the lesion probability map as an added overlay) into MRIcroGL. I then select the lesion probability map, click "options", "generate cluster table with options", and then set the threshold intensity to be 1 and the minimum cluster size to be 0. I obtain a table that tells me the volume and X,Y,Z coordinates of each lesion.

How do I get the MNI coordinates of each lesion?
(also, what exactly are the X,Y,Z coordinates? what space is that in?)

Note:
I read some posts that told me I can enter MNI coordinates and then view that brain location. I don't want to enter MNI coordinates, I want to be given MNI coordinates.

Any assistance you might be able to provide would be greatly appreciated. Additionally, please let me know if my question and/or explanation did not make sense.

Thank you so very much,
Emily Harriott
PhD student, Vanderbilt University

As per the title. Using the latest pre-compiled Linux release, opening or trying to run any script (either through the GUI or the command-line options) makes this error pop out:

Error: could not find symbol "PyImport_Cleanup"
Python could not be properly initialized. We must quit.

And then MRIcroGL proceeds to fail abruptly with an access violation. This is probably because PyImport_Cleanup has been removed after Python 3.8. Just to double-check, I installed Python 3.8 (via pyenv), updated the PyLib in the MRIcroGL12.ini file, and scripting works as normal again.

PS: MRIcroGL is awesome, thanks a lot for a really nice piece of software!

Hi @neurolabusc,

Just downloaded the latest release in this repo. The metal version appears to not be rendering the UI widgets in retina quality on mac OS 10.14.3. I tried the both the intel UHD graphics and the dedicated Radeon 560X. I imagine this is still experimental, so just thought I'd show what it's doing on my system.

screen shots attached.

metal

not metal

On macOS 10.15, the option for "Background transparent in bitmaps" always resets back to false on reopening MRIcroGL. In MRIcroGL.ini, there is no ScreenCaptureTransparentBackground field, and adding that field manually to the ini file has no effect (ScreenCaptureTransparentBackground=1 in the [BOOL] section).

Hi,

I am using the released version of MRIcroGL for windows, great software!

MRIcroGL can load some built-in parcellations, such as AAL, and it will automatically use the digitalized "random rainbow" colorbar for parcellation visualization.

Here, I want to use other non-built-in parcellation and to visualize the appearance of this parcellation. However, the "random rainbow" colorbar is not activated and I can not find the options for that. I wonder if there is any solution.

Thanks and best wishes,

Xiaoyu Chen

@ghisvail, @mih, and @yarikoptic:

I am wondering if you can give me feedback on my first attempt at a Debian package. This uses your advice from this issue and attempts to copy your prior work with the neurologist-debian micron package.

• Experimental deb files are here, built using the script in the Debian folder of this repository.
• I want to provide both a GTK2 and QT5 release. What is the expected naming scheme? Here I use mricrogl_gtk2 and mricrogl. I realize that GTK2 is EOL, but GTK3 is limited to Core OpenGL support and is unable to support multi-sampling. I think that the QT5 widgetset should be the default, but I do not think this will work on older Debian releases (e.g. Ubuntu 18.04).
• Perhaps @dbannon can provide an update, but the QT5 version requires libqt5pas
  1.2.9, which I believe is newer than in Debian trunk. Previously, I have included release notes to use this version, but not sure how to capture that in a .deb file.
• As suggested by @ghisvail, the Python scripting will use the system libpython, rather than statically embedding a copy of Python. I specify Python 3.7 and greater. However, since a user might have several versions of Python installed, I am not sure what the right approach for identifying the preferred Python version to use. At the moment, I use the highest version number, but in theory a future version of Python 3 could remove a feature that breaks my software.
• My software will use dcm2niix if it is installed. If it is not installed, the Import/ConvertDICOMtoNIfTI menu item is not functional, but everything else works. If this is an option, should the Debian control file list dcm2niix in the Depends or not?

Any advice appreciated. This is my first attempt at supporting Debian and I would like to hear any wisdom you might have.

I appreciate any advice

Latest commit includes enhancement suggested by Brandon Gunasekera to change the colorbar color and translucency.

The new command colorbarcolor(R,G,B,A) allows you to set the color and transparency of the color bar. The text on the colorbar will be white if the specified colorbar luminance is dark, otherwise the text will be black.

Be aware that the command resetdefaults() always sets your colorbar to opaque black (as if the script included the command colorbarcolor(0,0,0,255)), so if you want to reset defaults and set a custom colorbar color, you need to consider the order of these two commands.

This example script sets a translucent red color bar:

import gl
gl.resetdefaults()
gl.loadimage('spm152')
gl.colorbarcolor(255,0,0,160)

The Color/Colobar menu allows the user to choose from four default options. The equivalent script commands are:

• opaque white: gl.colorbarcolor(255,255,255,255)
• translucent white: gl.colorbarcolor(255,255,255,168)
• opaque black: gl.colorbarcolor(0,0,0,255)
• translucent black: gl.colorbarcolor(0,0,0,168)

This function will be included in the next release. To test compile from code.

Dear rordenlab,
the the mosaic function is really cool and I like it even better, that I can script it in python.
However, when I save the image (either using File > save Bitmap or using gl.savebmp) the slice numbers are all over the place, depending on the size of my window.

I am running MRIcroGL on a newly set up Ubuntu (20.04. LTS), placed it in /opt and created a mricorgl.desktop file (don't really think it makes a difference).

Best Vincent

It looks like that MIP rendering results in a 3D image, and the image can be rotated or flipped from any angle. So if the MIP rendering results a 3D image, can we save that 3D image into Nifti format image? If not, why the MIP rendering looks like a 3D image?

Hello!

I‘m using MRIcroGL on macOS 10.14.6. Is it the intended behavior, that opening multiple images side by side (multiple windows) is a bit complicated? If MRIcroGL is open, double clicking another file in the Finder doesn‘t open it in a new window, but instead uses the existing one. I have to manually open a new window from the menu (as there‘s no shortcut for it) and then open the file I want. If I forget this step, I have to manually reopen the one from before, again after opening a new window. Am I missing something?

Thanks!

Thank you for continuous development of MRIcroGL.

I installed the latest version (v1.2.20220720) on both Ubuntu 18.04 and Ubuntu 20.04.

Though MRIcroGL runs without any problems on Ubuntu 20.04, it fails with Ubuntu 18.04 with the following error;

MRIcroGL: /lib/x86_64-linux-gnu/libm.so.6: version `GLIBC_2.29' not found (required by ./MRIcroGL)

(Both MRIcroGL and MRIcroGL_QT result in the same error.)

The shipped version of libc6 on Ubuntu 18.04 is 2.27 whereas the latest MRIcroGL requires 2.29.

I tried to find a solution, but it seems installing libc>=2.29 might cause several problems system-wide.

Any ideas on how to handle this?

I have encountered a display/load error (Borg_Error) for large 4D nii.gz files in Windows.
Although I still cannot figure out the exact situations that lead to this error, I have seen the latest MRIcroGL (1.2.20220720) sometimes fails to load large t (4th dimension) nii.gz files and shows Borg_Error.

Some (marginal) info:

• The older 1.2.20211006 version has no Borg_Error issues in the same environments.
• If I converted them to uncompressed NIfTI files (*.nii), MRIcroGL displayed them without any issues.
• Large t (4th dim) is not always critical. Not found Borg_Error for HCPpipeline preprocessed nii.gz files (4th dim > 1000)
• (At the moment) I have not seen any issues in small t files (4th dim < 50).
• No issues in 3D volumes (4th dim = 1).
• Modality (function or diffusion) does not matter.

To reproduce:
Dataset 1: Haxby 2001 subj2

# haxby2001/subj2/bold.nii.gz
from nilearn import datasets
nilearn.datasets.fetch_haxby()

Dataset 2: ABIDE
https://s3.amazonaws.com/fcp-indi/data/Projects/ABIDE_Initiative/Outputs/cpac/func_minimal/OHSU_0050147_func_minimal.nii.gz

My Environment 1:
MRIcroGL 1.2.20220720 x86-64 FPC
Author: Chris Rorden
License BSD 2-Clause
Windows: 10.0
Intel; OpenGL= 4.6.0 - Build 30.0.101.1122; Shader=4.60 - Build 30.0.101.1122
2.1 Max 3D texture: 2048
Current texture: 64x64x64 1mb

My Environment 2:
MRIcroGL 1.2.20220720 x86-64 FPC
Author: Chris Rorden
License BSD 2-Clause
Windows: 10.0
NVIDIA Corporation; OpenGL= 4.6.0 NVIDIA 512.15; Shader=4.60 - NVIDIA
2.1 Max 3D texture: 16384
Current texture: 64x64x64 1mb

Version 1.2.20190902+
Tested in both windows and linux release builds.

Attempting to open any nifti image I have lying around will fail (pixelated image and color scale locked to rainbow) if the directory path contains the string /atlases/

The /atlases/ string can occur anywhere in the file path, as far as I can tell, to trigger the bug. /home/user/atlases/foo.nii triggers it just the same as /home/user/foo/bar/atlases/foo.nii

Directories named atlases_ etc. are fine. It seems to be specific to /atlases/.

This issue did not occur on 1.1x builds.

Thank you! Please let me know if I can be of further help.

It seems affine transforms in NIfTI are not calculated into the MVP matrix in glvolume2.pas. How do I add that if needed for certain visualizations?

Consider this kind of layer to layer scan, with each layer's displacement not perpendicular to layer itself:

MRIcroGL version: 1.2.20200707e Cocoa x86-64

@neurolabusc , some regions with low integer values in atlases rendered in 3D mode appear more transparent than high integer label regions.

For example, an atlas image with integer labels ranging from 1-140 (background == 0) will render regions with low values (1,2,3...) as more transparent.

Would it be possible to add a toggle setting in preferences to disable alpha modulation by voxel intensity?

No rush on this. Just submitting the issue as a possible enhancement for the future.

FSLeyes has recently added this feature in the master branch (not released), and it can be useful to toggle that switch when viewing some images.

I would be happy to help implement this. It would take some time to get up to speed with the MRIcroGL build system and code base.

https://git.fmrib.ox.ac.uk/fsl/fsleyes/fsleyes/-/blob/master/assets/gl/gl21/glvolume_3d_frag.glsl#L404

https://github.com/rordenlab/MRIcroGL12/blob/70d20ec1f1fb913e7efe5861e8d9dbb6ac77f52b/Resources/shader/Default.glsl#L51

atlas shown in screenshots:

https://git.fmrib.ox.ac.uk/open-science/UK_biobank_pipeline_v_1/-/raw/master/templates/GMatlas/GMatlas.nii.gz?inline=false

example MRIcroGL screenshot:

FSLeyes with intensity modulation

FSLeyes WITHOUT intensity modulation

I am trying to compile this package, and the README indicates this should be done via the same command line script as MRIcroGL-1.0.

I use:

lazbuild -B --lazarusdir="/usr/share/lazarus/" --pcp="system-lazarus-config" simplelaz.lpi

and get:

silenthost ~ # emerge -v mricrogl

 * IMPORTANT: 5 news items need reading for repository 'gentoo'.
 * Use eselect news read to view new items.


These are the packages that would be merged, in order:

Calculating dependencies... done!
[ebuild  N     ] sci-visualization/mricrogl-1.2.20190902::chymeric  0 KiB

Total: 1 package (1 new), Size of downloads: 0 KiB


>>> Verifying ebuild manifests

>>> Emerging (1 of 1) sci-visualization/mricrogl-1.2.20190902::chymeric
 * mricrogl-1.2.20190902.tar.gz SHA256 SHA512 WHIRLPOOL size ;-) ...                                                                 [ ok ]
>>> Unpacking source...
>>> Unpacking mricrogl-1.2.20190902.tar.gz to /var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work
>>> Source unpacked in /var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work
>>> Preparing source in /var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work/MRIcroGL12-1.2.20190902 ...
>>> Source prepared.
>>> Configuring source in /var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work/MRIcroGL12-1.2.20190902 ...
>>> Source configured.
>>> Compiling source in /var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work/MRIcroGL12-1.2.20190902 ...
cp: cannot stat '/etc/lazarus': No such file or directory
SetPrimaryConfigPath NewValue="system-lazarus-config" -> "/var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work/MRIcroGL12-1.2.20190902/system-lazarus-config"
Error: (lazbuild) package not found: simplelaz.lpi
 * ERROR: sci-visualization/mricrogl-1.2.20190902::chymeric failed (compile phase):
 *   (no error message)
 *
 * Call stack:
 *     ebuild.sh, line 125:  Called src_compile
 *   environment, line 785:  Called die
 * The specific snippet of code:
 *       lazbuild -B --lazarusdir="/usr/share/lazarus/" --pcp="system-lazarus-config" simplelaz.lpi || die
 *
 * If you need support, post the output of `emerge --info '=sci-visualization/mricrogl-1.2.20190902::chymeric'`,
 * the complete build log and the output of `emerge -pqv '=sci-visualization/mricrogl-1.2.20190902::chymeric'`.
 * The complete build log is located at '/var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/temp/build.log'.
 * The ebuild environment file is located at '/var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/temp/environment'.
 * Working directory: '/var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work/MRIcroGL12-1.2.20190902'
 * S: '/var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work/MRIcroGL12-1.2.20190902'

The README also mentions a “Run” command, but I cannot find one inside the source code:

silenthost /var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work/MRIcroGL12-1.2.20190902 # find -name "*run*"
silenthost /var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work/MRIcroGL12-1.2.20190902 # find -name "*Run*"
silenthost /var/tmp/portage/sci-visualization/mricrogl-1.2.20190902/work/MRIcroGL12-1.2.20190902 # find -name "*un*"
./mainunit.pas
./mainunit.lfm
./MRIcroGL.app/Contents/Resources/matcap/unused
./MRIcroGL.app/Contents/Resources/matcap/Burgundy.jpg
./MRIcroGL.app/Contents/Resources/lut/CT_Lungs.clut
./Resources/shader/unused
./Resources/shader/ancient/unused
./Resources/matcap/unused
./Resources/matcap/Burgundy.jpg
./Resources/lut/CT_Lungs.clut

This is a feature request to be able to both display cross hairs on the render view and allow mouse clicks on the render view to change the cross-hair position to the location of a mouse click (see issue 22).

Hi, I hope this is the right place to ask for help.

I'm using the latest precompiled version and whenever I try to open 4D nii.gz files (from bold fmri runs), a little window pops up, with the message "Warning: Loaded X out of 527 volumes".

Some more details:

• When I convert my DICOMS to a nii.gz using dcm2niix, MRIcroGL manages to load the first 30 out of the 527 volumes.
• When I try to view the preprocessed nii.gz file (fmriprep), it only loads the first 3 out of 527 volumes
• The size difference from original to preprocessed is roughly a factor of 10 (70MB vs 640MB). Could this be a memory issue?

Could provide the nifti files privately, if needed.
Thanks in advance!

Having compiled and installed the following files:

silenthost ~ # equery f mricrogl
 * Searching for mricrogl ...
 * Contents of sci-visualization/mricrogl-1.2.20200324:
/usr
/usr/bin
/usr/bin/MRIcroGL
/usr/share
/usr/share/MRIcroGL
/usr/share/MRIcroGL/lut
/usr/share/MRIcroGL/lut/1red.clut
/usr/share/MRIcroGL/lut/2green.clut
/usr/share/MRIcroGL/lut/3blue.clut
/usr/share/MRIcroGL/lut/4hot.clut
/usr/share/MRIcroGL/lut/5winter.clut
/usr/share/MRIcroGL/lut/6bluegrn.clut
/usr/share/MRIcroGL/lut/6warm.clut
/usr/share/MRIcroGL/lut/7cool.clut
/usr/share/MRIcroGL/lut/8redyell.clut
/usr/share/MRIcroGL/lut/CT_Bones.clut
/usr/share/MRIcroGL/lut/CT_Kidneys.clut
/usr/share/MRIcroGL/lut/CT_Liver.clut
/usr/share/MRIcroGL/lut/CT_Muscles.clut
/usr/share/MRIcroGL/lut/CT_Skull.clut
/usr/share/MRIcroGL/lut/CT_Soft_Tissue.clut
/usr/share/MRIcroGL/lut/CT_Surface.clut
/usr/share/MRIcroGL/lut/CT_Vessels.clut
/usr/share/MRIcroGL/lut/CT_w_Contrast.clut
/usr/share/MRIcroGL/lut/GE_color.clut
/usr/share/MRIcroGL/lut/HOTIRON.clut
/usr/share/MRIcroGL/lut/Inferno.clut
/usr/share/MRIcroGL/lut/NIH.clut
/usr/share/MRIcroGL/lut/Plasma.clut
/usr/share/MRIcroGL/lut/Viridis.clut
/usr/share/MRIcroGL/lut/actc.clut
/usr/share/MRIcroGL/lut/blue2red.clut
/usr/share/MRIcroGL/lut/bone.clut
/usr/share/MRIcroGL/lut/bronze.clut
/usr/share/MRIcroGL/lut/copper.clut
/usr/share/MRIcroGL/lut/copper2.clut
/usr/share/MRIcroGL/lut/cubehelix.clut
/usr/share/MRIcroGL/lut/electric_blue.clut
/usr/share/MRIcroGL/lut/gold.clut
/usr/share/MRIcroGL/lut/jet.clut
/usr/share/MRIcroGL/lut/linspecer.clut
/usr/share/MRIcroGL/lut/surface.clut
/usr/share/MRIcroGL/lut/x_rain.clut
/usr/share/MRIcroGL/script
/usr/share/MRIcroGL/script/basic.py
/usr/share/MRIcroGL/script/clip.py
/usr/share/MRIcroGL/script/cluster.py
/usr/share/MRIcroGL/script/ct_abdomen.py
/usr/share/MRIcroGL/script/ct_head.py
/usr/share/MRIcroGL/script/cutout.py
/usr/share/MRIcroGL/script/explode.py
/usr/share/MRIcroGL/script/explode2.py
/usr/share/MRIcroGL/script/glass.py
/usr/share/MRIcroGL/script/help.py
/usr/share/MRIcroGL/script/hidezeros.py
/usr/share/MRIcroGL/script/invert.py
/usr/share/MRIcroGL/script/jagged.py
/usr/share/MRIcroGL/script/mip.py
/usr/share/MRIcroGL/script/mosaic.py
/usr/share/MRIcroGL/script/mosaic2.py
/usr/share/MRIcroGL/shader
/usr/share/MRIcroGL/shader/Default.glsl
/usr/share/MRIcroGL/shader/Edges.glsl
/usr/share/MRIcroGL/shader/Glass.glsl
/usr/share/MRIcroGL/shader/MIP.glsl
/usr/share/MRIcroGL/shader/Matte.glsl
/usr/share/MRIcroGL/shader/Minimal.glsl
/usr/share/MRIcroGL/shader/Occlusion.glsl
/usr/share/MRIcroGL/shader/OverlaySurface.glsl
/usr/share/MRIcroGL/shader/Shell.glsl
/usr/share/MRIcroGL/shader/ShellEdges.glsl
/usr/share/MRIcroGL/shader/Slow 2.glsl
/usr/share/MRIcroGL/shader/Slow.glsl
/usr/share/MRIcroGL/shader/SpecialEffects.glsl
/usr/share/MRIcroGL/shader/Standard.glsl
/usr/share/MRIcroGL/shader/Tomography.glsl

I get lots of issues from MRIcroGL, and while the interface starts, it's not able to do much (I can try to do a video recording if the exact behaviour is of interest to you).

I am assuming perhaps MRIcriGL does not find the files under /usr/share/MRIcroGL (which would be the correct place to install them according to the FHS). Is this the case? And if so, how can I tell it to look in the right place?

Somehow, whenever I use the save as bitmap function, images come out with blue tint, especially for overlay images. Attached are the screenshot of the console, and the image generated with save as bitmap.

I'm using the precompiled MRIcroGL 1.2.20190902++ GTK2 on Linux. Load up any image and overlay, and overlay color will come out with blue tint when saved as bitmap.

Screenshot:

Saved file:

I'm trying to reproduce a visualization that looks like this:

But with the glass brain option in MRIcroGL, the thickness of the edges gets exagerated, giving it almost a marker-drawn, chunky aesthetic:

Is there a way to tweak the glass brain renderer so that it looks more similar to the images I'm trying to reproduce?

see for examples: https://github.com/niivue/niivue/tree/main/src/cmaps

related to #39 as an option to adjust display of filenames an example of which I use here

if I have a following filenames structure

d1/mn152.nii.gz
d2/mn152.nii.gz

opening both of them at the same time would just show two mni152 entries in the pane, making it impossible to distinguish between the two. It would be cool if mricrogl helped to make difference obvious. It could be

• manual option on how many leading parent paths to include in the display
• automagic option: if conflict (the same name) is detected, choose sufficient number of leading paths to distinguish

thank you very much in advance

Is it possible to include the following option:

gl.overlaydepth(depthPct)  -> depth percentage of overlay

The upcoming release of MRIcroGL for Linux will target the GTK2 and QT5 widget sets. The following (fixable) issues remain:

• There are autosizing issues and some controls are frozen until the screen is resized.

There are inherent limitations with GTK3 that make it poorly suited for this project. Since GTK 3.22 was supposed to be the last version, these are unlikely to be resolved.

• OpenGL [multi-sampling is not supported](https://github.com/aklomp
  /gtk3-opengl/issues/2), therefore the quality of MRIcroGL images in GTK3 is poorer than QT5 and GTK2.
• Complicated GTK3 widgets like GtkSpinButton never have real GtkWidget children. This creates limitations inusing these widgets with MRIcroGL.

Therefore, while one can compile MRIcroGL for GTK3, the functionality is limited.

Hello,

Using the latest release of MRIcroGL, I have the following message on startup:

GLSL error 0:57(28): warning: `textureSz' used uninitialized
0:67(20): warning: `textureSz' used uninitialized

Then the display looks incomplete:

I'm using Ubuntu 18.04

Thanks for your help.

For MacOS and Linux, MRIcroGL will report progress for loading volumes in the terminal. However, this is not the case with Windows. 3D images are often slow to load relative to 2D images due to file size and cache proximity between slices. Therefore, interactive information during loads would be useful.

I downloaded MRIcroGL 2-Novemember from https://www.nitrc.org/projects/mricrogl to convert PAR/REC file into Nifti format. Now, I got stuck with a problem. When I tried to convert a PAR/REC file into Nifti format, the MRIcroGL doesn't work. The error message is as below.

However, if I modify the diffusion (ap, fh, rl) parameter in image information session from (0.000, 0.000, -1.000) to (0.000001, 0.000, -1.000), MRIcroGL could convert the data into Nifti format. Can you help me fix the problem?

Many DICOM viewers include 10cm rulers. Add "Ruler" checkbox to "2D Slice Selection" tools to show/hide ruler when in 2D slice view.

Hi,

On MRIcron, when you do a render the brain is opaque, so you only see the surface activation.

However on MRIcroGL, when you do "surface overlay" the brain is transparent, so you see activation deep inside as well.

The issue is, when you modify the "overlay depth" it does not specify how deep you are looking into the cortex. Can someone please tell me how much in mm/cm does the overlay depth correspond to? Or how can I replicate the exact search depth that is used in MRIcron?

How can I make the brain opaque in MRIcroGL?

How can I find out the voxel-wise search depth in MRIcroGL? Which is an option in MRIcron like this:

user reported that when opening a git-annex'ed file, mricrogl shows the filename of the resolved file which is a long git-annex key:

I wonder if there could be an option to avoid resolving symlinks either globally or for specific files? Thanks in advance!

MRIcroGL 1.0 had a menu item "Add Overlay (remove small clusters)" - it would be nice to include this in MRIcroGL 1.2. In the meantime, users can use the Matlab script nii_threshreslicecluster.m.

Suggested by Jake.

I am trying to install the linux version of MRIcroGL on WSL. I can doenload the zip file but the download help page says to extract from the ZIP and then run the executable. But what is the executable? I can see a file called MRIcroGL_QT. When I try to execute that I get the error:

Unable to find AFNI_ATLAS_PATH
MRIcroGL/MRIcroGL_QT: symbol lookup error: MRIcroGL/MRIcroGL_QT: undefined symbol: QLCLOpenGLWidget_Create

As if it is expecing that I have setup some environmental variables.

If possible could the 'Show Header' information get updated when if it is open when a new image is loaded in?
Roger

I think it is possible in pascal/lazarus, but I don't know if that implementation would work cross platform

Here's a reference implementation in NiiVue. I'm just creating a css linear gradient style string by looping over the RGBA nodes in the JSON color map.

Similar to the NiiVue and FSLeyes color drop down widgets

The name of this release is incorrect. See attached screen capture:

Text in addition to colors for Orientation Cube (e.g. S\I\A\P\L\R). Suggested by las nikolay.

Is it possible to save a T1 (added through open file) and it's spmT overlay (added as an overlay) together as a single nifti file? So that the T1 image is modified to have itself along with the activation maps of the spmT.nii file?

So when I start MRIcroGL I get a blank window, after which I can either wait a while or just close it, and the viewer starts up normally.

I also see this in the command line:

[deco]~ ❱ MRIcroGL
If there is a long delay at launch, ensure full GTK2 install: "sudo apt-get install appmenu-gtk2-module"
Unable to find AFNI_ATLAS_PATH
Loading preferences /home/chymera/.MRIcroGL12.ini
Unable to find folder "/usr/share/MRIcroGL/standard"
Unable to find folder "/usr/share/MRIcroGL/atlas"
Load time 26
float range 0..0.94197642803192139
float window 0...0.83225166797637939
Init time 5
Set Min/Max Window 0..0.83225166797637939 time 1
Update RGBA time 1

Is this the issue I'm getting? If so, what file does it need exactly? that's as far as I can tell a specific package name from Debian or Ubuntu.

Also, for reference I tried both =mricrogl-1.2.20211007 and =mricrogl-1.2.20220720, both of them with either

lazbuild -B --lazarusdir="/usr/share/lazarus/" --pcp="system-lazarus-config" MRIcroGL_Debian.lpi || die

or

lazbuild -B --lazarusdir="/usr/share/lazarus/" --pcp="system-lazarus-config" MRIcroGL_NoPython.lpi || die

all 4 with the same results.

As reported by Marcus Heldmann adding statistical overlays that have thresholded by SPM12 can look slightly eroded when saved as 32-bit float relative to saving as 16-bit integer. This only occurs when overlay loading is smoothed (Load Smooth Overlays in Options button menu is checked), and only for maps from SPM, not other tools. The reason for this is that SPM sets voxels that do not reach the threshold to not-a-number (NaN), whereas for the integer data the voxels that do not reach threshold are set to zero. Interpolating a positive value with a zero yields a (smaller) positive number, but interpolating a positive number with a NaN yields a NaN. The upcoming release will convert NaNs to zeros.

As ever, it is always recommended to apply a threshold AFTER reslicing the statistical map to the resolution of the background image. Since statistical maps are intentionally smoothed, sub-threshold voxels near the edge of a significant cluster are virtually always near threshold. Resampling before thresholding preserves this information. However, the upcoming release also changes the interpolation of thresholded images a little bit to ty to fight some of the artifacts of re-sampling a thresholded image. The issue is illustrated in the 1D figure below. The red rectangles represent original voxel values, and the red horizontal dotted line shows an arbitrary threshold applied to these voxels (here at 5.5). The voxels shown in solid red survive the threshold, those that do not survive are shown with a dashed outline. The blue line shows the interpolation one would get by linear interpolation of all the voxels regardless of whether they are above or below the threshold. The intersection of the blue line and the dotted red line shows the ideal thresholding. Simple linear interpolation applied to the thresholded data yields the green line. The artificially zeroed voxels have eroded this object: fewer voxels will survive the threshold. However, the interpolation also shows an artificial border of sub-threshold signal. The proposed solution is shown in orange. In this case, when an image is interpolated, one detects the smallest positive signal in the entire 3D volume (which is likely to be extremely close to the threshold applied to the input image). For our example, lets call say this is 5.52 (the picture only shows surviving voxels with an intensity of 6, but presumably other surviving voxels in the image are a bit closer to the threshold). We filter the interpolated image so any voxel less than 2.51 will get the value zero, and any voxel with an intensity between 2.51 and 5.52 will get the value 5.52. This preserves the volume of the ideal outcome, and avoids the artificial dark ring surrounding thresholded objects.

Fix Windows OS issue saving bitmaps.

How can specify an ROI color with the input of RGB values, instead of using the "colorname"?

The Options button has an item Generate cluster table with options which allows the user to specify thresholds for filtering data. @jpurcel8 has suggested an enhancement: providing a minimum separation between peaks to ensure they are not too close together. A reference implementation exists in bspm.

I've got some points and ROIs I've mapped in MNI152, but I've just realized they may have been origialy plotted on something closer to the SPM152 template. Is there a transformation from MNI152 to SPM152 available? Is is just a rescaling?

Hi, I'm trying to build the currently latest version (1.2.20210317) and I get this build error: build.log.

I think it could be a missing file?

Fatal: (2013) Cannot open include file "../Metal-Demos/common/glopts.inc"

I am using =fpc-3.2.2 and =lazarus-2.0.12.

Hi! I got a fresh installation of MRIcroGL on an Ubuntu 22.04 laptop. I would like to make MRIcroGL the default application for opening .nii and .nii.gz files. However, when I right click on a .nii file and select MRIcroGL as application for opening, I get the following error:

Pressing "OK" does nothing.