Dev

PET — Dev Tier

(3 scenes)

Blind evaluation tier — no ground truth available.

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.

Parameter Specifications

🔒

True spec hidden — estimate parameters from spec ranges below.

Parameter Spec Range Unit
attenuation -6.0 – 9.0 %
scatter_frac 0.24 – 0.39
timing_res 140.0 – 290.0 ps
normalization -2.4 – 3.6 %

Dev Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 PET-ViT + gradient 0.778 33.2 0.954 0.82 ✓ Certified Smith et al., ICCV 2024
2 U-Net-PET + gradient 0.735 30.38 0.922 0.81 ✓ Certified Ronneberger et al. variant, MICCAI 2020
3 PETFormer + gradient 0.706 28.1 0.883 0.86 ✓ Certified Li et al., ECCV 2024
4 TransEM + gradient 0.679 27.19 0.863 0.81 ✓ Certified Xie et al., 2023
5 DeepPET + gradient 0.677 26.67 0.85 0.85 ✓ Certified Haggstrom et al., MIA 2019
6 OS-EM + gradient 0.666 26.03 0.833 0.86 ✓ Certified Hudson & Larkin, IEEE TMI 1994
7 ML-EM + gradient 0.651 25.56 0.819 0.84 ✓ Certified Shepp & Vardi, IEEE TPAMI 1982
8 FBP-PET + gradient 0.637 25.02 0.803 0.83 ✓ Certified Analytical baseline
9 MAPEM-RDP + gradient 0.557 21.52 0.669 0.87 ✓ Certified Nuyts et al., IEEE TMI 2002
10 OSEM + gradient 0.556 21.33 0.661 0.89 ✓ Certified Hudson & Larkin, IEEE TMI 1994

Visible Data Fields

y H_ideal spec_ranges

Dataset

Format: HDF5
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