Eidors-logo    

EIDORS: Electrical Impedance Tomography and Diffuse Optical Tomography Reconstruction Software

EIDORS (mirror)
Main
Documentation
Tutorials
− Image Reconst
− Data Structures
− Applications
− FEM Modelling
− GREIT
− Old tutorials
Workshop
Download
Contrib Data
GREIT
Browse Docs
Browse SVN

News
Mailing list
(archive)
FAQ
Developer
                       

 

Hosted by
SourceForge.net Logo

 

Samples for the EIDORS Workshop

To run tutorials, you need to download and install EIDORS and then run this command in a matlab (or octave) session.
>>run /path/to/eidors3d/startup.m

EIDORS Workshop @ EIT 2017 Conf

  1. Create a directory tutorial
  2. Download eidors-v3.9-ng.zip and unzip into the directory
  3. Start Matlab
  4. In matlab, do
    • cd H:/path/to/tutorial/eidors-v3.9
    • startup
    • compare_2d_algs(1) %(this tests if eidors is working)
    • show_fem(ng_mk_cyl_models(3,[0],[])) %(this tests if netgen is working)
Challenge problem:
  1. Create a 3D elliptical cylinder with 16 circular electrodes (see here)
    fmdl= ng_mk_ellip_models([1,1.2,0.8],[16,0.5],[0.1]); show_fem(fmdl);
  2. Put a ball into the elliptical cylinder
    extra={'ball','solid ball = sphere(0.5,0.5,0.5;0.1);'};
    fmdl= ng_mk_ellip_models([1,1.2,0.8],[16,0.5],[0.1],extra);
    show_fem(fmdl);
  3. Put two balls into the elliptical cylinder
    extra={'ball','solid ball = sphere(0.5,0.5,0.5;0.1) or sphere(0.5,-0.5,0.5;0.1);'};
    fmdl= ng_mk_ellip_models([1,1.2,0.8],[16,0.5],[0.1],extra);
    show_fem(fmdl);
  4. Set the model to use adjacent current patterns
    stim = mk_stim_patterns(16,1,[0,1],[0,1],{}); fmdl.stimulation = stim;
  5. Simulate homogeneous voltages (background conductivity = 0.5);
    img = mk_image(fmdl, 0.5); vh = fwd_solve(img);
    show_fem(img);
  6. Simulate inhomogeneous voltages (ball conductivity = 1.0). Add noise.
    img.elem_data(fmdl.mat_idx{2})= 1.0; vi = fwd_solve(img); vi = add_noise(0.5, vi, vh);
    show_fem(img);
  7. Reconstruct the image using a GN solver
    imdl= mk_common_model('b2c2',16); imdl.stimulation = stim; imdl= select_imdl(imdl,{'Basic GN dif'}); img= inv_solve(imdl,vh,vi);
    show_fem(img);
  8. Reconstruct with a different hyperparameter (lower resolution)
    imdl.hyperparameter.value = .3; img= inv_solve(imdl,vh,vi);
    show_fem(img);
  9. Reconstruct the image using GREIT (see here)
    opt.noise_figure = 0.5;
    fmdl.normalize_measurements = 0;
    imdl = mk_GREIT_model(mk_image(fmdl,0.5), 0.25, [], opt);
    img= inv_solve(imdl,vh,vi);
    show_fem(img);
  10. Compare images for skip 4 stimulation/measurement pattern (see here)

EIDORS Workshop

Last Modified: $Date: 2017-06-21 08:59:21 -0400 (Wed, 21 Jun 2017) $ by $Author: aadler $