Public
Spectral CT — Public Tier
(3 scenes)Full-access development tier with all data visible.
What you get
Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.
How to use
Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.
What to submit
Reconstructed signals (x_hat) and corrected spec as HDF5.
Parameter Specifications
✓
True spec visible — use these exact values for Scenario III oracle reconstruction.
| Parameter | Spec Range | True Value | Unit |
|---|---|---|---|
| energy_calibration_error | -4.0 – 8.0 | 2.0 | keV |
| scatter_fraction | -0.2 – 0.4 | 0.1 | |
| detector_crosstalk | -0.1 – 0.2 | 0.05 | |
| beam_hardening | -0.2 – 0.4 | 0.1 |
InverseNet Baseline Scores
Method: CPU_baseline — Mismatch parameter: nominal
Scenario I (Ideal)
14.88 dB
SSIM 0.2885
Scenario II (Mismatch)
11.68 dB
SSIM 0.0479
Scenario III (Oracle)
14.89 dB
SSIM 0.1290
Per-scene breakdown (4 scenes)
| Scene | PSNR I | SSIM I | PSNR II | SSIM II | PSNR III | SSIM III |
|---|---|---|---|---|---|---|
| scene_00 | 14.80 | 0.2898 | 11.69 | 0.0458 | 14.77 | 0.1268 |
| scene_01 | 15.05 | 0.2857 | 11.70 | 0.0477 | 14.91 | 0.1282 |
| scene_02 | 14.99 | 0.2846 | 11.68 | 0.0506 | 14.96 | 0.1312 |
| scene_03 | 14.69 | 0.2938 | 11.67 | 0.0475 | 14.91 | 0.1299 |
Public Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | CTFormer + gradient | 0.859 | 38.05 | 0.982 | 0.93 | ✓ Certified | Li et al., ICCV 2024 |
| 2 | DiffusionCT + gradient | 0.843 | 37.33 | 0.979 | 0.89 | ✓ Certified | Kazemi et al., ECCV 2024 |
| 3 | Score-CT + gradient | 0.843 | 36.97 | 0.978 | 0.91 | ✓ Certified | Song et al., NeurIPS 2024 |
| 4 | DuDoTrans + gradient | 0.838 | 36.03 | 0.974 | 0.94 | ✓ Certified | Wang et al., MLMIR 2022 |
| 5 | CT-ViT + gradient | 0.835 | 36.26 | 0.975 | 0.91 | ✓ Certified | Guo et al., NeurIPS 2024 |
| 6 | DOLCE + gradient | 0.828 | 36.4 | 0.975 | 0.87 | ✓ Certified | Liu et al., ICCV 2023 |
| 7 | Learned Primal-Dual + gradient | 0.822 | 34.78 | 0.966 | 0.94 | ✓ Certified | Adler & Oktem, IEEE TMI 2018 |
| 8 | FBPConvNet + gradient | 0.794 | 33.4 | 0.956 | 0.89 | ✓ Certified | Jin et al., IEEE TIP 2017 |
| 9 | PnP-DnCNN + gradient | 0.785 | 32.29 | 0.946 | 0.92 | ✓ Certified | Zhang et al., IEEE TIP 2017 |
| 10 | RED-CNN + gradient | 0.785 | 32.17 | 0.944 | 0.93 | ✓ Certified | Chen et al., IEEE TMI 2017 |
| 11 | PnP-ADMM + gradient | 0.750 | 30.43 | 0.923 | 0.88 | ✓ Certified | Venkatakrishnan et al., IEEE GlobalSIP 2013 |
| 12 | TV-ADMM + gradient | 0.706 | 27.78 | 0.876 | 0.89 | ✓ Certified | Sidky et al., Phys. Med. Biol. 2008 |
| 13 | FBP + gradient | 0.644 | 24.71 | 0.793 | 0.9 | ✓ Certified | Kak & Slaney, IEEE Press 1988 |
Visible Data Fields
y
H_ideal
spec_ranges
x_true
true_spec
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%