Hidden

Magnetic Particle Imaging (MPI) — Hidden Tier

(5 scenes)

Fully blind server-side evaluation — no data download.

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.

Parameter Specifications

🔒

True spec hidden — blind evaluation, only ranges available.

Parameter Spec Range Unit
drive_field_amplitude 24.58 – 26.38 mT
selection_field_gradient 2.43 – 2.73 T/m
particle_relaxation_time 1.86 – 2.46 us
receive_coil_sensitivity 0.979 – 1.069 -

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 ScoreExperimental + gradient 0.633 24.56 0.788 0.86 ✓ Certified Wei et al., 2025
2 Domain-Adapted-CNN + gradient 0.624 24.52 0.787 0.82 ✓ Certified Domain adaptation CNN
3 PnP-RED + gradient 0.614 24.1 0.772 0.82 ✓ Certified Romano et al., IEEE TIP 2017
4 DiffusionExperimental + gradient 0.586 23.24 0.74 0.79 ✓ Certified Zhang et al., 2024
5 SwinIR + gradient 0.576 22.51 0.711 0.83 ✓ Certified Liang et al., ICCVW 2021
6 ResUNet + gradient 0.552 21.93 0.687 0.79 ✓ Certified Residual U-Net baseline
7 Wiener Filter + gradient 0.552 21.43 0.665 0.86 ✓ Certified Wiener filtering baseline
8 ExpFormer + gradient 0.541 21.67 0.676 0.77 ✓ Certified Experimental science transformer, 2024
9 PnP-ADMM + gradient 0.541 21.81 0.682 0.75 ✓ Certified ADMM + denoiser prior
10 Tikhonov + gradient 0.536 21.2 0.655 0.81 ✓ Certified Tikhonov, Doklady 1963
11 Matched Filter + gradient 0.501 20.41 0.618 0.75 ✓ Certified Optimal linear filter

Dataset

Scenes: 5

Scoring Formula

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%
Back to Magnetic Particle Imaging (MPI)