Dev

Spectral CT — Dev Tier

(3 scenes)

Blind evaluation tier — no ground truth available.

Measurements (y), ideal forward operator (H), and spec ranges only.

Apply your pipeline from the Public tier. Use consistency as self-check.

Reconstructed signals and corrected spec. Scored server-side.

🔒

True spec hidden — estimate parameters from spec ranges below.

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	CT-ViT + gradient	0.783	33.59	0.958	0.82	✓ Certified	Guo et al., NeurIPS 2024
2	DiffusionCT + gradient	0.774	32.77	0.95	0.83	✓ Certified	Kazemi et al., ECCV 2024
3	CTFormer + gradient	0.771	32.09	0.944	0.86	✓ Certified	Li et al., ICCV 2024
4	Score-CT + gradient	0.760	31.09	0.932	0.88	✓ Certified	Song et al., NeurIPS 2024
5	FBPConvNet + gradient	0.734	29.2	0.904	0.9	✓ Certified	Jin et al., IEEE TIP 2017
6	PnP-DnCNN + gradient	0.732	30.23	0.92	0.81	✓ Certified	Zhang et al., IEEE TIP 2017
7	DuDoTrans + gradient	0.726	29.2	0.904	0.86	✓ Certified	Wang et al., MLMIR 2022
8	Learned Primal-Dual + gradient	0.707	28.91	0.899	0.79	✓ Certified	Adler & Oktem, IEEE TMI 2018
9	TV-ADMM + gradient	0.707	28.49	0.891	0.83	✓ Certified	Sidky et al., Phys. Med. Biol. 2008
10	DOLCE + gradient	0.706	28.85	0.898	0.79	✓ Certified	Liu et al., ICCV 2023
11	PnP-ADMM + gradient	0.693	27.06	0.86	0.89	✓ Certified	Venkatakrishnan et al., IEEE GlobalSIP 2013
12	RED-CNN + gradient	0.673	26.22	0.838	0.88	✓ Certified	Chen et al., IEEE TMI 2017
13	FBP + gradient	0.630	24.19	0.775	0.89	✓ Certified	Kak & Slaney, IEEE Press 1988

y H_ideal spec_ranges

Format: HDF5

Scenes: 3

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

PSNR: 40% SSIM: 40% Consistency: 20%