Hidden

PET — 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
attenuation -3.5 – 11.5 %
scatter_frac 0.265 – 0.415
timing_res 165.0 – 315.0 ps
normalization -1.4 – 4.6 %

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 PET-ViT + gradient 0.744 30.97 0.93 0.81 ✓ Certified Smith et al., ICCV 2024
2 U-Net-PET + gradient 0.715 28.47 0.89 0.87 ✓ Certified Ronneberger et al. variant, MICCAI 2020
3 OS-EM + gradient 0.676 27.05 0.859 0.81 ✓ Certified Hudson & Larkin, IEEE TMI 1994
4 PETFormer + gradient 0.667 25.92 0.83 0.88 ✓ Certified Li et al., ECCV 2024
5 DeepPET + gradient 0.602 23.47 0.749 0.84 ✓ Certified Haggstrom et al., MIA 2019
6 TransEM + gradient 0.593 22.95 0.729 0.86 ✓ Certified Xie et al., 2023
7 ML-EM + gradient 0.584 23.39 0.746 0.76 ✓ Certified Shepp & Vardi, IEEE TPAMI 1982
8 FBP-PET + gradient 0.578 22.61 0.715 0.83 ✓ Certified Analytical baseline
9 OSEM + gradient 0.517 20.25 0.611 0.85 ✓ Certified Hudson & Larkin, IEEE TMI 1994
10 MAPEM-RDP + gradient 0.504 19.8 0.589 0.85 ✓ 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