Hidden

Mammography — Hidden Tier

(3 scenes)

Fully blind server-side evaluation — no data download.

No data downloadable. Algorithm runs server-side on hidden measurements.

Package algorithm as Docker container / Python script. Submit via link.

Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

🔒

True spec hidden — blind evaluation, only ranges available.

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	DiffusionCT + gradient	0.750	31.65	0.939	0.79	✓ Certified	Kazemi et al., ECCV 2024
2	CTFormer + gradient	0.745	30.5	0.924	0.85	✓ Certified	Li et al., ICCV 2024
3	DOLCE + gradient	0.742	30.68	0.927	0.82	✓ Certified	Liu et al., ICCV 2023
4	CT-ViT + gradient	0.741	31.2	0.933	0.78	✓ Certified	Guo et al., NeurIPS 2024
5	Score-CT + gradient	0.709	29.18	0.903	0.78	✓ Certified	Song et al., NeurIPS 2024
6	DuDoTrans + gradient	0.650	25.86	0.828	0.8	✓ Certified	Wang et al., MLMIR 2022
7	FBPConvNet + gradient	0.643	25.73	0.824	0.78	✓ Certified	Jin et al., IEEE TIP 2017
8	Learned Primal-Dual + gradient	0.640	25.69	0.823	0.77	✓ Certified	Adler & Oktem, IEEE TMI 2018
9	RED-CNN + gradient	0.635	24.66	0.791	0.86	✓ Certified	Chen et al., IEEE TMI 2017
10	PnP-ADMM + gradient	0.634	24.7	0.792	0.85	✓ Certified	Venkatakrishnan et al., IEEE GlobalSIP 2013
11	TV-ADMM + gradient	0.618	24.01	0.769	0.85	✓ Certified	Sidky et al., Phys. Med. Biol. 2008
12	PnP-DnCNN + gradient	0.600	23.29	0.742	0.85	✓ Certified	Zhang et al., IEEE TIP 2017
13	FBP + gradient	0.571	22.88	0.726	0.76	✓ Certified	Kak & Slaney, IEEE Press 1988

Scenes: 3

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

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