Hidden
Portal Imaging (EPID) — 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 |
|---|---|---|
| isocenter_shift | -0.28 – 0.92 | mm |
| beam_energy_variation | 5.972 – 6.092 | MV |
| detector_sag | -0.14 – 0.46 | mm |
| scatter_kernel_width | 4.72 – 5.92 | mm |
Hidden Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | CTFormer + gradient | 0.742 | 30.44 | 0.923 | 0.84 | ✓ Certified | Li et al., ICCV 2024 |
| 2 | DuDoTrans + gradient | 0.731 | 30.39 | 0.923 | 0.79 | ✓ Certified | Wang et al., MLMIR 2022 |
| 3 | CT-ViT + gradient | 0.721 | 29.41 | 0.907 | 0.82 | ✓ Certified | Guo et al., NeurIPS 2024 |
| 4 | Score-CT + gradient | 0.720 | 29.19 | 0.904 | 0.83 | ✓ Certified | Song et al., NeurIPS 2024 |
| 5 | Learned Primal-Dual + gradient | 0.693 | 27.9 | 0.879 | 0.81 | ✓ Certified | Adler & Oktem, IEEE TMI 2018 |
| 6 | DiffusionCT + gradient | 0.690 | 27.89 | 0.878 | 0.8 | ✓ Certified | Kazemi et al., ECCV 2024 |
| 7 | PnP-DnCNN + gradient | 0.682 | 28.11 | 0.883 | 0.74 | ✓ Certified | Zhang et al., IEEE TIP 2017 |
| 8 | FBPConvNet + gradient | 0.674 | 26.56 | 0.847 | 0.85 | ✓ Certified | Jin et al., IEEE TIP 2017 |
| 9 | PnP-ADMM + gradient | 0.673 | 27.3 | 0.865 | 0.77 | ✓ Certified | Venkatakrishnan et al., IEEE GlobalSIP 2013 |
| 10 | TV-ADMM + gradient | 0.648 | 25.05 | 0.804 | 0.88 | ✓ Certified | Sidky et al., Phys. Med. Biol. 2008 |
| 11 | DOLCE + gradient | 0.643 | 25.89 | 0.829 | 0.76 | ✓ Certified | Liu et al., ICCV 2023 |
| 12 | RED-CNN + gradient | 0.642 | 24.88 | 0.798 | 0.87 | ✓ Certified | Chen et al., IEEE TMI 2017 |
| 13 | FBP + gradient | 0.593 | 22.88 | 0.726 | 0.87 | ✓ Certified | Kak & Slaney, IEEE Press 1988 |
Dataset
Scenes: 3
Scoring Formula
0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)
PSNR: 40%
SSIM: 40%
Consistency: 20%