Occiput - Tomographic Vision - version 1.1

Tomographic reconstruction software for PET, PET-MRI and SPECT in 2D, 3D (volumetric) and 4D (spatio-temporal) in Python.

The software provides high-speed reconstruction using Graphics Processing Units (GPU). Note: an NVidia CUDA-compatible GPU is required.

Occiput can be utilized with arbitrary scanner geometries. It can be utilized for abstract tomographic reconstruction experiments to develop new algorithms and explore new system geometries, or to connect to real-world scanners, providing production quality image reconstruction with standard algorithms (such as MLEM and OSEM).

Occiput implements advanced algorithms for motion correction, kinetic imaging, multi-modal reconstruction, respiratory and cardiac gated imaging. The source code contains Jupyter notebooks with examples.

A Python extension package Occiput_Interface_Biograph_mMR, implementing the interface to the Siemens Biograph mMR PET-MRI scanner is available upon request and following authorization from Siemens. Notebooks containing Biograph_mMR in the title can only be executed after installing the extension package. Please email us at occiput.reconstruction@gmail.com

Installation

Linux, Windows (not tested recently), MacOS

Pre-requisites: Occiput requires NVidia GPU Drivers, NVidia CUDA and the NiftyRec GPU accelerated tomographic ray-tracing library.

Install NVidia GPU Drivers and CUDA
Install NiftyRec libraries - build the latest version using CMake
Make sure that CUDA libraries and NiftyRec libraries are in the system path:

Linux:

export LD_LIBRARY_PATH:$LD_LIBRARY_PATH:/path_to_cuda_libraries:/path_to_niftyrec_libraries

MacOS:

export DYLD_LIBRARY_PATH:$DYLD_LIBRARY_PATH:/path_to_cuda_libraries:/path_to_niftyrec_libraries

Windows:

setx path "%path%;c:/path_to_cuda_libraries:/path_to_niftyrec_libraries;"

Install Occiput:

git clone https://github.com/mscipio/occiput.git

python setup.py build install (*not sure if this works right now, please let me know if you try!*)

Getting started

Examples and demos of the features of Occiput are in the /occiput/notebooks folder. To get started, install Python Jupyter and open the scripts in /occiput/notebooks. The notebook /occiput/notebooks/DOCUMENTATION.ipynb contains an index and short description of the notebooks.

Website

For more information see occiput.io.

Changelog