Hello, I am a total novice when it comes to using these types of programs. I have been in Windows IT support for two decades, so compiling things is not something I've ever needed to do; plus, very little Terminal use. I've been tasked with determining if MRI Reface can work for our use-cases, so I’m looking for help/direction on how to get these applications to function. I’m not a programmer or developer, so I’m really at a loss.

I'm attempting to first do a POC with the native 3.5 Mac version. Then, later, potentially having this run on a Linux EC2 using the native Linux 3.5 release.

We are using a Mac Studio M1 Max on Tahoe 26.5. I’ve installed MATLAB 2022a. Easy. Next steps: the Niftyreg and ANTS components.

#1 - Are these the best links to use for these components for this Mac?

Nifty reg: https://github.com/KCL-BMEIS/niftyreg

ANTS: https://github.com/antsx/ants

#2 - Should different links be used if trying to set up the native Linux install? If so, which?

#3 – Am I right that I should be using CMake to compile both Nifty reg and ANTS?

#4 - I see the CMake text files in the Github data. Is this the best resource for which cmds to run? Are people using these successfully to compile and run the applications?

#5 - Is Ubuntu still the recommended Linux OS? Is there a release version that should be used or is known-good?

#6 – Once the Niftyreg and ANTS components are believed to be successfully ‘installed’, how do you run MRI Reface?

Thanks for any help you can provide or resources you can point me to.

Jesse,

Firstly, if you're not pretty comfortable in Linux to compile and set up all the dependencies, I would recommend skipping the Mac step and running on Linux using the Docker image. Docker relieves you of having to do any of that, but the Docker image is x86/x64 and won't run on Apple Silicon (ARM).  If your final production environment is Linux, your easiest route would be Docker on Linux.

For your more specific questions:
1) yes, those are the links to the ANTs and Niftyreg projects. 
2) On NITRC, our "default" version is Linux native (without Docker). The "Docker" version should run on any x86/x64 but Linux is the only thing we've tested. I'm not aware of anyone running it on Windows. The mac version (native) will run on Mac.
3) Yes, both of those projects use CMake to build
4) In an ideal case cmake builds look something like--
cd folder/with/source/code # folder will contain CMakeLists.txt
mkdir build
cd build
cmake ..
make
sudo make install

and repeat for ANTs.

5) The distribution shouldn't really make a difference. Ubuntu should be fine and if it wasn't working on Ubuntu I'm sure someone would have reported it by now, but we are running Rocky Linux 8 (descends from Centos, which descends from Redhat).

6) Use the included run_mri_reface.sh script

Good luck!
Chris Schwarz

Thanks. Really appreciate the information. We'll try to implement on the Mac first. If we don't have success, then we'll try with Docker.

Hi Chris,

I running across some errors and I was hoping you could help.

I cloned both repos (ANTS and Niftyreg) into our home directory, created the build folders, but ran into an error for niftyreg:

 corelabadmin@Corelab-Mac-Studio01-PRD build % make
[  1%] Building CXX object reg-lib/CMakeFiles/_reg_maths.dir/cpu/Maths.cpp.o
clang++: error: unsupported option '-msse3' for target 'arm64-apple-darwin25.5.0'
make[2]: *** [reg-lib/CMakeFiles/_reg_maths.dir/cpu/Maths.cpp.o] Error 1
make[1]: *** [reg-lib/CMakeFiles/_reg_maths.dir/all] Error 2
make: *** [all] Error 2
corelabadmin@Corelab-Mac-Studio01-PRD build % cd ..
corelabadmin@Corelab-Mac-Studio01-PRD niftyreg % cd ..

I suspect that it is anticipating x86 architecture instead of the more recent ARM architecture on Mac computers.

Any help that can be provided would be appreciated.

It's not something I remember encountering in the one time (so far) that I built this on Apple. When I Google your error, it suggests that disabling SSE would probably remove the Intel-specific flag. The niftyreg functionality mri_reface is using doesn't require any of the optional features, so for any niftyreg build problems I would disable whatever you can via cmake (here, SSE, but if errors with libjpeg, disable jpeg, etc.) and try again. The "ccmake" GUI interface to cmake would help you see what options are available. 

I gave up on the manual compiling setup as I kept facing issues regarding implementation (not being able to find entities, etc.). So, I tried setting up the docker environment on Windows by leveraging WSL. There was no problem with setting up the image, but when I attempt to run the .sh, I get the following errors.

Kazim,