This script installs a large number of NeuroImaging related software on Linux (I prefer Linux Mint 21.1, but Ubuntu 22.04 also works). It is not maintained anymore to install on Windows-WSL2 (latest Win11 preferably) or macOS. Results may vary...
It is assumed you have an nvidia gpu.
The script installs the latest stable versions of:
- Anaconda3
- Video Studio Code
- nvidia cuda
- mrtrix3
- FSL
- Docker
- Freesurfer
- ANTs
- dcm2niix
- dcm2bids
- DCMTK
- GDCM
- SPM12
- cat12
- lead-DBS
- conn toolbox
- KUL_NeuroImaging_tools
- KUL_VBG
- KUL_FWT
- Mevislab
- Scilpy (in the conda environment "scilpy")
- FastSurfer (in the conda environment "fastsurfer_gpu")
- Karawun (in the conda environment "KarawunEnv")
The terminal in WSL2, macOS and Linux will show which version of software is currently installed.
- Matlab (R2017b or later) - only needed if you use spm12, conn, cat12, lead-dbs
- Should run out of the box.
- follow the instructions given by the installer
- Install win11, perform all updates
- Install nvidia-cuda: follow https://docs.microsoft.com/en-us/windows/ai/directml/gpu-cuda-in-wsl
- Open powershell as administrator and exe: wsl --install
- reboot when asked
- after reboot you will be asked to setup the user/passwd
- INSIDE WIN11: install
- visual studio code
- docker desktop & setup the ubuntu as wsl integration
- open Ubuntu in wsl2
- follow the instructions given by the installer
- Should work out of the box
- follow the instructions given by the installer
Open a terminal. Copy the following commands into terminal.
wget https://raw.githubusercontent.com/treanus/KUL_Linux_Installation/master/KUL_Install_NeuroImagingSoftware.sh
chmod +x KUL_Install_NeuroImagingSoftware.sh
Now run the script
./KUL_Install_NeuroImagingSoftware.sh
In a first step this will install Anaconda3. Follow the installation instruction, and accept the license and initialization. Reboot when asked.
Rerun the script.
./KUL_Install_NeuroImagingSoftware.sh
Depending on the cpu and the internet connection speed this will take a few hours.
On some occasions you will be asked to enter your password and/or to confirm the installation.
- The freesurfer license file needs to be installed. Put it in $KUL_apps_DIR/freesurfer_license
- Start matlab, add spm12 and conn to the path. The path for spm12 and conn is $HOME/KUL_apps/spm and conn.
- On macOS open Docker Desktop; in Resources/FileSharing add the folder with the freesurfer license.