NITRC VisuAlign - Nonlinear adjustments after QuickNII Forum: general

RE: QuickNII-Kim-UnifiedMouse-v1 atlas is not read by VisuAlign?

Gergely Csucs — Thu, 10 Jul 2025 10:33:37 GMT

Hi,
While it's not released "officially" yet, there's a version of VisuAlign with the Kim atlas at https://www.nesys.uio.no/VisuAlign/
I hope this helps (and that you are on Windows)

QuickNII-Kim-UnifiedMouse-v1 atlas is not read by VisuAlign?

nina1996 — Wed, 09 Jul 2025 18:16:39 GMT

Hi, 
 
I am using QuickNII-Kim-UnifiedMouse-v1 but I noticed that the file that I downloaded for VisuAlign only includes ABA atlases: ABA_Mouse_CCFv3_2015_25um.cutlas and ABA_Mouse_CCFv3_2017_25um.cutlas and rat atlases but no KIM atlas, because of this I can not further align the brain regions to my picture. Could someone please help me how to circumvent this issue?

RE: Issue with downloading VisuAlign

Gergely Csucs — Tue, 28 Jan 2025 17:20:06 GMT

Hi,
Windows has built-in support for zip archives, just instead of opening it directly, you should right-click on VisuAlign-v0_9.zip, and select the option "Extract all...". Then it will let you specify a location for the files (the automatic choice will be a new VisuAlign-v0_9 folder - so same name, just no .zip at the end), and it will also offer to show the files after finishing, picking that option is handy.
Then try starting VisuAlign.bat from this new location. While it may or may not work, it definitely should not want you to buy anything.
I hope this helps, best regards,
Gergely

RE: Questions regarding using QuickMaskNL for batch processing rat brain samples

Janine Reinert — Tue, 28 Jan 2025 10:01:53 GMT

Hi Gergely, thanks a lot for the detailed reply and apologies for my late response (the email notifying about the post was deemed spam by our university filter...).
Thanks for explaining how the coordinates come about. I tried to retroactively confirm the trial-and-error ones I came up with using QuickNII and I think I get it. For anyone that stumbles upon this: it appears in the example coordinates given "left" is posterior location, while "right" is anterior (to me this was counter intuitive but inverting them also works). It's also explained a bit here <a href="https://www.protocols.io/view/quicknii-brain-atlas-registration-e6nvwdme2lmk/v1?version_warning=no&step=21" target="_blank" rel="noopener">Protocols Step 21</a>
 
Originally posted by Gergely Csucs:
<blockquote>
Setup:
<ul>
<li>find a convenient location, if you're not familiar with navigating the command line, try creating a new folder in the root of a drive where you have write access (in Windows Explorer, navigate to "This PC" - will be different if Windows is set to German -, and check locations that have a drive letter, like C:, D:, etc.). Let's assume it's M: and the folder is called MaskTest</li>
<li>download QNLpack.zip from https://www.nesys.uio.no/QNL/ and extract it. There is a single QNL folder inside, that's the one you will need to copy into the previously created MaskTest folder on the M: drive</li>
<li>you will need Java, download it from https://openjdk.org/ - it's not friendly to the eye, but there's a link "jdk.java.net/23" in the middle of the second paragraph, that's where you can find the current downloads). Get the one for Windows, extract it, and right now it contains a jdk-23.0.2 folder, copy that into MaskTest too</li>
<li>copy all of the JSON files you want to process into the same MaskTest folder.</li>
</ul>
So now MaskTest contains QNL and jdk-23.0.2 folders, and lots of JSON files.
Now start a command prompt, like open the Start Menu and type cmd. Probably it will arrive in your user folder, so type the drive letter where you have MaskTest, and a colon, then press Enter. In my case I type m:. Now type cd \MaskTest and press Enter. These two steps set the working location in the command prompt.
Here we can test if the installation works, first try jdk-23.0.2\bin\java --version to see if Java works properly, and then try jdk-23.0.2\bin\java -cp QNL qnlmask.QuickMask to check if QuickMaskNL works too. Then just close it.
If everything worked so far, two steps remain:
<ol>
<li>creating output folders for every JSON: for %f in (*.json) do mkdir "%~nf"</li>
<li>and running the tool: for %f in (*.json) do jdk-23.0.2\bin\java -cp QNL qnlmask.QuickMask "%f" 242 32.2 292.8 242 367.9 291.8 242 34.5 48.7 "%~nf"</li>
</ol>
 
</blockquote>
Thank you very much for sharing this in a very beginner friendly way!! I'll give it a go today but at first glance it seems doable even for someone with my non-existent java knowledge. Cheers
Janine

Issue with downloading VisuAlign

cicechen — Mon, 27 Jan 2025 20:45:42 GMT

Hi, I'm trying to download VisuAlign on Windows for QUINT workflow. I downloaded v0.9 zip file and tried to open VisuAlign.bat from there. However, the file couldn't open on my computer. I was wondering if I'm doing any steps wrong or do I need to make purchase (it was telling me to purchase rarlab)? Thank you for your help!

RE: Questions regarding using QuickMaskNL for batch processing rat brain samples

Gergely Csucs — Wed, 22 Jan 2025 15:41:18 GMT

Hi,
The masking tool is a quick'n'dirty solution, the coordinates are the corners of a triangle from the cutting plane. Usually they're coming from QuickNII:
<ul>
<li>open QuickNII, don't load any series</li>
<li>set up the main view in the way how you want to cut the atlas into half</li>
<li>read coordinates for 3 suitably distant points, e.g. move the mouse cursor somewhere in the top-left, read the coordinates (they're displayed in the purple title bar), then repeat this with top-right and bottom-center.</li>
</ul>
So in the given case the atlas is halved along the x=242 plane (all 3 points have this coordinate).
For mass-producing these masks you will have to use the command line, and that can be painful.
Setup:
<ul>
<li>find a convenient location, if you're not familiar with navigating the command line, try creating a new folder in the root of a drive where you have write access (in Windows Explorer, navigate to "This PC" - will be different if Windows is set to German -, and check locations that have a drive letter, like C:, D:, etc.). Let's assume it's M: and the folder is called MaskTest</li>
<li>download QNLpack.zip from https://www.nesys.uio.no/QNL/ and extract it. There is a single QNL folder inside, that's the one you will need to copy into the previously created MaskTest folder on the M: drive</li>
<li>you will need Java, download it from https://openjdk.org/ - it's not friendly to the eye, but there's a link "jdk.java.net/23" in the middle of the second paragraph, that's where you can find the current downloads). Get the one for Windows, extract it, and right now it contains a jdk-23.0.2 folder, copy that into MaskTest too</li>
<li>copy all of the JSON files you want to process into the same MaskTest folder.</li>
</ul>
So now MaskTest contains QNL and jdk-23.0.2 folders, and lots of JSON files.
Now start a command prompt, like open the Start Menu and type cmd. Probably it will arrive in your user folder, so type the drive letter where you have MaskTest, and a colon, then press Enter. In my case I type m:. Now type cd \MaskTest and press Enter. These two steps set the working location in the command prompt.
Here we can test if the installation works, first try jdk-23.0.2\bin\java --version to see if Java works properly, and then try jdk-23.0.2\bin\java -cp QNL qnlmask.QuickMask to check if QuickMaskNL works too. Then just close it.
If everything worked so far, two steps remain:
<ol>
<li>creating output folders for every JSON: for %f in (*.json) do mkdir "%~nf"</li>
<li>and running the tool: for %f in (*.json) do jdk-23.0.2\bin\java -cp QNL qnlmask.QuickMask "%f" 242 32.2 292.8 242 367.9 291.8 242 34.5 48.7 "%~nf"</li>
</ol>
In the download location there's a screenshot of the ideal case too (I processed only 2 files), though it won't do much if anything goes wrong.
I hope this helps, best regards:
Gergely

Questions regarding using QuickMaskNL for batch processing rat brain samples

Janine Reinert — Tue, 21 Jan 2025 16:58:03 GMT

Hello everyone, I've been using VisuAlign as part of the QUINT workflow to adjust slices registered to the Waxholm Space Atlas (using QuickNII 2.2. with v4 of the Waxholm Atlas).
This has been working but I've come across two issues now 1. for generation of hemibrain maps using QuickMaskNL, the reference coordinates supplied on the Nutil documentation (https://nutil.readthedocs.io/en/latest/QuantifierMasks.html) appear to be for a mouse brain or at least not for v4 of the WHS atlas. I've managed to generate roughly fitting maps through trial and error (coordinate string: 242 32.2 292.8 242 367.9 291.8 242 34.5 48.7) but I'd be grateful if anyone could explain what the coordinates actually signify and how they were determined in the first place so I might be able to replicate this for the WHS atlas 2. I have a large number of brains (> 100) that I've registered and now need to generate the hemibrain maps for. Is there any way to automate QuickMaskNL for this? Manually entering all the info for the location of the json files, etc. is cumbersom work I'd like to avoid if at all possible

RE: Ability to use custom atlas labels

Aaron Sathyanesan — Tue, 20 Aug 2024 17:59:50 GMT

Hi Gergely
Attached is the traceback. Also, I looked into your other solutions regarding upscaling the images, but for some reason, the downscaled images and the hi-res segmentations have different sizes and lengthXwidth ratios? Not sure what is going on here. I attached an example of this.
Appreciate your help on this matter very much.
Best Aaron

RE: Ability to use custom atlas labels

Gergely Csucs — Tue, 06 Aug 2024 11:52:04 GMT

Hi Aaron,
<blockquote>
<div>I tried using a modified python code to read the flat file, however, the output is a low-res PNG. What would I need to do in order to get a high-res PNG similar to what is the final output images using NUTIL?</div>
</blockquote>
<div>Nutil gets that low-resolution image too, then scales it up to the desired resolution (of the image coming from Ilastik - typically). ImageJ/Fiji can do this too: Image menu, "Scale...", then type the needed resolution and set Interpolation to None.</div>
<blockquote>
<div>I am specifically interested in extracting the distinct cerebellar nuclei. If each of these nuclei have different colors, I can use fiji to extract ROI based on color information. However, currently the labeling in the final output of NUTIL yields the same color for all cerebellar nuclei.</div>
</blockquote>
<div>Yes, this issue comes up from time to time. A workaround is provided at https://www.nesys.uio.no/VisuAlign/relabel.html - it's a small tool that creates a label file with distinct colors (not necessarily for the human eye, but for programs like Fiji), that can be used with VisuAlign.</div>
I hope this helps.
Best regards, Gergely

RE: Ability to use custom atlas labels

Aaron Sathyanesan — Mon, 05 Aug 2024 19:54:15 GMT

Hello Gergely
<div>I tried using a modified python code to read the flat file, however, the output is a low-res PNG. What would I need to do in order to get a high-res PNG similar to what is the final output images using NUTIL? I am specifically interested in extracting the distinct cerebellar nuclei. If each of these nuclei have different colors, I can use fiji to extract ROI based on color information. However, currently the labeling in the final output of NUTIL yields the same color for all cerebellar nuclei.</div>
<div> </div>
<div>However, since I do not need object counts, I just obtained the high-res color images using empty segmentation images in the NUTIL workflow. I got the high res maps but the distinct nuclei do not show up. I tried using one of your answers to relabel the label.txt file, but I am unsure how to proceed from here. Because the json file which was exported from Visualign does not open up in visualign itself. So, how would I go about getting only the high-resolution PNGs with each of the cerebellar nuclei (fastigial nucleus, interposed nucleus, and dentate nucleus) distinctly labeled?</div>
<div> </div>
<div>Best</div>
<div>Aaron</div>