Electrical Impedance Tomography (EIT)

Electrical Impedance Tomography (EIT)

Join the Competition Contribute to Benchmark

Standard reconstruction benchmark — forward model perfectly known, no calibration needed. Score = 0.5 × clip((PSNR−15)/30, 0, 1) + 0.5 × SSIM

# Method Score PSNR (dB) SSIM Source
🥇 EIT-Former 0.690 30.0 0.880 ✓ Certified EIT reconstruction transformer, 2024
🥈 D-bar CNN 0.645 28.5 0.840 ✓ Certified Hamilton & Hauptmann, IEEE TMI 2018
🥉 TV-ADMM 0.508 24.5 0.700 ✓ Certified Borsic et al., Physiol. Meas. 2010
4 Gauss-Newton 0.375 21.0 0.550 ✓ Certified Cheney et al., SIAM Rev. 1999

Dataset: PWM Benchmark (4 algorithms)

Blind Reconstruction Challenge — forward model has unknown mismatch, must calibrate from data. Score = 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

# Method Overall Score Public
PSNR / SSIM
 Dev
PSNR / SSIM
 Hidden
PSNR / SSIM
 Trust Source
🥇 EIT-Former + gradient 0.593
0.705
27.9 dB / 0.879
0.561
21.73 dB / 0.678
0.512
20.27 dB / 0.612
✓ Certified EIT reconstruction transformer, 2024
🥈 D-bar CNN + gradient 0.530
0.706
27.47 dB / 0.869
0.480
19.32 dB / 0.566
0.404
16.8 dB / 0.440
✓ Certified Hamilton & Hauptmann, IEEE TMI 2018
🥉 TV-ADMM + gradient 0.485
0.611
23.02 dB / 0.732
0.458
18.41 dB / 0.520
0.386
16.18 dB / 0.410
✓ Certified Borsic et al., Physiol. Meas. 2010
4 Gauss-Newton + gradient 0.421
0.474
18.53 dB / 0.526
0.412
17.12 dB / 0.456
0.378
15.51 dB / 0.378
✓ Certified Cheney et al., SIAM Rev. 1999

Complete score requires all 3 tiers (Public + Dev + Hidden).

Join the competition →
Scoring: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖) PSNR 40% · SSIM 40% · Consistency 20%
Public 5 scenes

Full-access development tier with all data visible.

What you get & how to use

What you get: Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.

How to use: Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.

What to submit: Reconstructed signals (x_hat) and corrected spec as HDF5.

Public Leaderboard
# Method Score PSNR SSIM
1 D-bar CNN + gradient 0.706 27.47 0.869
2 EIT-Former + gradient 0.705 27.9 0.879
3 TV-ADMM + gradient 0.611 23.02 0.732
4 Gauss-Newton + gradient 0.474 18.53 0.526
Spec Ranges (4 parameters)
Parameter Min Max Unit
contact_impedance 80.0 140.0 ohm
electrode_position_error -0.4 0.8 mm
background_conductivity 0.16 0.28 S/m
current_amplitude_drift 0.99 1.02 -
View Details →
Dev 5 scenes

Blind evaluation tier — no ground truth available.

What you get & how to use

What you get: Measurements (y), ideal forward operator (H), and spec ranges only.

How to use: Apply your pipeline from the Public tier. Use consistency as self-check.

What to submit: Reconstructed signals and corrected spec. Scored server-side.

Dev Leaderboard
# Method Score PSNR SSIM
1 EIT-Former + gradient 0.561 21.73 0.678
2 D-bar CNN + gradient 0.480 19.32 0.566
3 TV-ADMM + gradient 0.458 18.41 0.52
4 Gauss-Newton + gradient 0.412 17.12 0.456
Spec Ranges (4 parameters)
Parameter Min Max Unit
contact_impedance 76.0 136.0 ohm
electrode_position_error -0.48 0.72 mm
background_conductivity 0.152 0.272 S/m
current_amplitude_drift 0.988 1.018 -
Submit Reconstruction View Details →
Hidden 5 scenes

Fully blind server-side evaluation — no data download.

What you get & how to use

What you get: No data downloadable. Algorithm runs server-side on hidden measurements.

How to use: Package algorithm as Docker container / Python script. Submit via link.

What to submit: Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

Hidden Leaderboard
# Method Score PSNR SSIM
1 EIT-Former + gradient 0.512 20.27 0.612
2 D-bar CNN + gradient 0.404 16.8 0.44
3 TV-ADMM + gradient 0.386 16.18 0.41
4 Gauss-Newton + gradient 0.378 15.51 0.378
Spec Ranges (4 parameters)
Parameter Min Max Unit
contact_impedance 86.0 146.0 ohm
electrode_position_error -0.28 0.92 mm
background_conductivity 0.172 0.292 S/m
current_amplitude_drift 0.993 1.023 -
Submit Algorithm View Details →

Blind Reconstruction Challenge

Challenge

Given measurements with unknown mismatch and spec ranges (not exact params), reconstruct the original signal. A method must be evaluated on all three tiers for a complete score. Scored on a composite metric: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖).

Input

Measurements y, ideal forward model H, spec ranges

Output

Reconstructed signal x̂

Spec DAG — Forward Model Pipeline

M → D

M Modulation
D Detector

Mismatch Parameters

Symbol Parameter Description Nominal Perturbed
c_i contact_impedance Contact impedance (ohm) 100.0 120.0
e_p electrode_position_error Electrode position error (mm) 0.0 0.4
b_c background_conductivity Background conductivity (S/m) 0.2 0.24
c_a current_amplitude_drift Current amplitude drift (-) 1.0 1.01

Credits System

40%
Platform Profit Pool
Revenue allocated to benchmark rewards
30%
Winner Share
Top algorithm receives from pool
$100
Min Withdrawal
Minimum payout threshold
Spec Primitives Reference (11 primitives)
P Propagation

Free-space or medium propagation kernel (Fresnel, Rayleigh-Sommerfeld).

M Mask / Modulation

Spatial or spatio-temporal amplitude modulation (coded aperture, SLM pattern).

Π Projection

Geometric projection operator (Radon transform, fan-beam, cone-beam).

F Fourier Sampling

Sampling in the Fourier / k-space domain (MRI, ptychography).

C Convolution

Shift-invariant convolution with a point-spread function (PSF).

Σ Summation / Integration

Summation along a physical dimension (spectral, temporal, angular).

D Detector

Sensor readout with gain g and noise model η (Gaussian, Poisson, mixed).

S Structured Illumination

Patterned illumination (block, Hadamard, random) applied to the scene.

W Wavelength Dispersion

Spectral dispersion element (prism, grating) with shift α and aperture a.

R Rotation / Motion

Sample or gantry rotation (CT, electron tomography).

Λ Wavelength Selection

Spectral filter or monochromator selecting a wavelength band.