Hidden

PET/MR — Hidden Tier

(3 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
mr_attenuation_error -17.5 – 57.5 %
motion_shift -4.2 – 13.8 pixels
b0_inhomogeneity -28.0 – 92.0 Hz
pet_scatter_fraction -0.175 – 0.575

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 PET-ViT + gradient 0.717 29.63 0.911 0.78 ✓ Certified Smith et al., ICCV 2024
2 PETFormer + gradient 0.682 27.57 0.871 0.79 ✓ Certified Li et al., ECCV 2024
3 FBP-PET + gradient 0.666 27.16 0.862 0.75 ✓ Certified Analytical baseline
4 TransEM + gradient 0.661 25.83 0.827 0.86 ✓ Certified Xie et al., 2023
5 U-Net-PET + gradient 0.653 25.93 0.83 0.81 ✓ Certified Ronneberger et al. variant, MICCAI 2020
6 ML-EM + gradient 0.628 25.24 0.81 0.76 ✓ Certified Shepp & Vardi, IEEE TPAMI 1982
7 OS-EM + gradient 0.596 23.85 0.763 0.76 ✓ Certified Hudson & Larkin, IEEE TMI 1994
8 OSEM + gradient 0.564 22.23 0.7 0.81 ✓ Certified Hudson & Larkin, IEEE TMI 1994
9 DeepPET + gradient 0.524 20.55 0.625 0.84 ✓ Certified Haggstrom et al., MIA 2019
10 MAPEM-RDP + gradient 0.469 19.31 0.565 0.75 ✓ Certified Nuyts et al., IEEE TMI 2002

Dataset

Scenes: 3

Scoring Formula

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

PSNR: 40% SSIM: 40% Consistency: 20%
Back to PET/MR