This project presents a novel Electrode Selection Algorithm (ESA) for use in Electrical Impedance Tomography (EIT) on 2-dimensional samples. It also features an implementation of a solver of the EIT forward problem that considers the Complete Electrode Model. The model uses the GREIT implementation of the PyEIT package (https://github.com/liubenyuan/pyEIT) for EIT computations, as well as some other parts of the PyEIT software. Our team does not have any claims for these parts of the code.
Note that the mesh and eit folders contain copies of the corresponding files found in the PyEIT package (https://github.com/liubenyuan/pyEIT). The exceptions are the fem.py and the other python files starting with fem- in the eit folder. They are new implementations of Finite Electrode Method solvers written exclusively by our team - Ivo Mihov and Vasil Avramov.
| Packages | Notes |
|---|---|
| NumPy | tested with numpy 1.16.5 |
| PyEIT * | tested with pyeit 0.0.1 |
| CuPy | tested with cupy 6.0.0 |
| Scikit-Learn | tested with scikit-learn 0.21.3 |
| Tensorflow 2 | tested with tensorflow-gpu 2.2.0 |
| h5py | tested with h5py 2.10.0 |
| SciPy | tested with scipy 1.4.1 |
| Scikit-image | tested with scikit-image 0.15.0 |
| Matplotlib | tested with matplotlib 3.1.1 |
| Scikit-optimize | tested with scikit-optimize 0.7.4 |
* PyEIT package was modified in this project, so you may need to substitute the package files with the ones included in this project to use this software.