Dev

Muon Tomo — 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
angular_resolution -2.4 – 3.6 mrad
momentum_estimate -12.0 – 18.0 %
detector_efficiency -3.6 – 5.4 %

Dev Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 TransEM + gradient 0.729 29.62 0.911 0.84 ✓ Certified Xie et al., 2023
2 PETFormer + gradient 0.716 28.42 0.889 0.88 ✓ Certified Li et al., ECCV 2024
3 PET-ViT + gradient 0.676 27.05 0.859 0.81 ✓ Certified Smith et al., ICCV 2024
4 DeepPET + gradient 0.643 25.46 0.816 0.81 ✓ Certified Haggstrom et al., MIA 2019
5 MAPEM-RDP + gradient 0.619 24.5 0.786 0.8 ✓ Certified Nuyts et al., IEEE TMI 2002
6 FBP-PET + gradient 0.616 24.18 0.775 0.82 ✓ Certified Analytical baseline
7 ML-EM + gradient 0.595 22.86 0.726 0.88 ✓ Certified Shepp & Vardi, IEEE TPAMI 1982
8 OSEM + gradient 0.577 22.62 0.716 0.82 ✓ Certified Hudson & Larkin, IEEE TMI 1994
9 U-Net-PET + gradient 0.567 21.74 0.679 0.89 ✓ Certified Ronneberger et al. variant, MICCAI 2020
10 OS-EM + gradient 0.540 21.11 0.651 0.84 ✓ 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 Muon Tomo