A python-based, open-source package for Electrical Impedance Tomography (EIT)

1. Introduction

1.1 Dependencies

Packages	Optional	Note
numpy		tested with numpy-1.10.4, numpy-1.11.rc1
scipy		tested with scipy-0.17.0
matplotlib		tested with matplotlib-1.5.1
vispy		tested with vispy-git
pandas	Optional	tested with pandas-0.17.1
xarray	Optional	for long term data analysis
distmesh	Optional	it currently has a build-in distmesh module (supports 2D and 3D!)
tetgen	Optional	generate 3D meshes

Note 1, Why vispy ? pyEIT uses vispy for visualizing 3D meshes (tetrahedron). vispy has minimal system dependencies, all you need is a decent graphical card with OpenGL support. It supports fast rendering, which I think is more superior to vtk or mayavi. Please go to the website vispy.org or github repository vispy.github for more details.

Note 2, How to contribute ? The interested user can contribute (create a PR! any type of improvement is welcome) forward simulation, inverse solving algorithms as well as their models at current stage. We will setup a wiki page dedicated to this topic.

Note 3, Fast setup. Anaconda from continuum is highly recommended for this package.

1.2 Features

• 2D forward and inverse computing of EIT
• Reconstruction algorithms : Gauss-Newton solver (JAC), Back-projection (BP), 2D GREIT
• 2D/3D visualization!

1.3 Todo

• Generate complex shape using distmesh
• Add support for 3D forward and inverse computing
• 3D mesh generation and visualization
• More algorithms and data pre-processing modules
• Complete electrode model

2. Installation

pyEIT is purely python based (in current version), so it can be installed and run without any difficulty.

2.1 Install global

$ python setup.py build
$ python setup.py install

2.2 set PYTHONPATH (recommended)

export PYTHONPATH=/path/to/pyEIT

In windows, you may set PYTHONPATH as a system wide environment. If you are using spyder-IDE, or pyCharm, you may also set PYTHONPATH in the IDE, which is more convenient. Please refer to a specific tool for detailed information.

3. Run the demo

Enter the demo folder, pick one demo, and run !

3.1 (2D) forward and inverse computing

Using demo/demo_dynamic_bp.py

Using demo/demo_dynamic_greit.py

Using demo/demo_dynamic_jac.py

Using demo/demo_static_jac.py

3.2 (3D) forward and inverse computing