Dev

CT — Dev Tier

(20 scenes)

Blind evaluation: 20 real patient CT slices from LoDoPaB-CT (validation split, patients 0–63).

What you get

Measured sinogram (y), ideal forward operator (H), and spec ranges. No ground truth.

How to use

Apply your pipeline from Public tier. Self-check via consistency metric. Ground truth scored server-side.

What to submit

Reconstructed images and corrected mismatch spec. Scored server-side.

Parameter Specifications

🔒

True spec hidden — estimate parameters from spec ranges below.

Parameter Spec Range Unit
center_offset_px -3.0 – 7.0 px
angle_error_deg -5.0 – 11.0 deg
beam_hardening_beta -0.07 – 0.23
detector_tilt_deg -2.0 – 4.0 deg

Dev Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 CT-FM + gradient 0.828 35.87 0.973 0.9 ✓ Certified Wang et al., Nature MI 2026
2 CTformer + gradient 0.814 36.23 0.975 0.81 ✓ Certified Wang et al., MICCAI 2023
3 Score-CT + gradient 0.803 35.65 0.972 0.79 ✓ Certified Gao et al., IEEE TMI 2024
4 CT-MAE + gradient 0.803 35.01 0.968 0.83 ✓ Certified Chen et al., MICCAI 2024
5 TransCT + gradient 0.798 34.88 0.967 0.81 ✓ Certified Xia et al., MICCAI 2021
6 Eformer + gradient 0.797 34.49 0.964 0.83 ✓ Certified Wang et al., AAAI 2022
7 PINER-CT + gradient 0.795 34.39 0.964 0.83 ✓ Certified Sun et al., CVPR 2025
8 DuDoRNet + gradient 0.767 31.41 0.936 0.89 ✓ Certified Zhou et al., CVPR 2020
9 DiffusionMBIR + gradient 0.756 32.07 0.943 0.79 ✓ Certified Song et al., arXiv 2024
10 iCT-Net + gradient 0.754 31.21 0.934 0.84 ✓ Certified Li et al., IEEE TMI 2019
11 LEARN + gradient 0.740 30.6 0.926 0.82 ✓ Certified Chen et al., IEEE TPAMI 2018
12 PnP-ADMM + gradient 0.722 29.47 0.908 0.82 ✓ Certified Venkatakrishnan et al., GlobalSIP 2013
13 BM3D-CT + gradient 0.697 28.04 0.882 0.82 ✓ Certified Dabov et al., IEEE TIP 2007; Chen 2014
14 DLCT + gradient 0.666 26.34 0.841 0.83 ✓ Certified Xu et al., IEEE TMI 2012
15 SART + gradient 0.660 25.58 0.82 0.88 ✓ Certified Andersen & Kak, Ultrason. Imaging 1984
16 RED-CNN + gradient 0.654 25.48 0.817 0.86 ✓ Certified Chen et al., IEEE TMI 2017
17 CGLS + gradient 0.636 25.13 0.806 0.81 ✓ Certified Bjorck, SIAM 1996
18 WGAN-CT + gradient 0.635 24.58 0.789 0.87 ✓ Certified Wolterink et al., IEEE TMI 2017
19 OSEM + gradient 0.631 24.83 0.797 0.82 ✓ Certified Hudson & Larkin, IEEE TMI 1994
20 CT-U-Net + gradient 0.621 24.74 0.794 0.78 ✓ Certified Han et al., Phys. Med. Biol. 2016
21 FBPConvNet + gradient 0.605 23.82 0.762 0.81 ✓ Certified Jin et al., IEEE TMI 2017
22 TV-ADMM + gradient 0.552 21.34 0.661 0.87 ✓ Certified Sidky & Pan, Phys. Med. Biol. 2008
23 FBP + gradient 0.533 20.74 0.634 0.86 ✓ Certified Kak & Slaney, IEEE Press 1988
24 ART-TV + gradient 0.517 20.59 0.627 0.8 ✓ Certified Li et al., Med. Phys. 2004

Visible Data Fields

y H_ideal spec_ranges

Dataset

Format: HDF5
Scenes: 20

Scoring Formula

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

PSNR: 40% SSIM: 40% Consistency: 20%
Back to CT