Dev

Confocal Laser Endomicroscopy (CLE) — 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.

Parameter	Spec Range	Unit
fiber_bundle_honeycomb_pattern	-0.15 – 0.15	-
motion_artifact	-2.4 – 3.6	px/frame
fluorescein_concentration_variation	0.52 – 1.72	relative

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	Restormer-CLE + gradient	0.793	34.34	0.963	0.82	✓ Certified	Zamir et al., CVPR 2022 (CLE)
2	DiffusionEndo + gradient	0.769	33.03	0.953	0.79	✓ Certified	Li et al., Med. Image Anal. 2024
3	SwinIR-CLE + gradient	0.714	29.23	0.904	0.8	✓ Certified	Liang et al., ICCV 2021 (CLE)
4	CARE-CLE + gradient	0.708	28.56	0.892	0.83	✓ Certified	Weigert et al., Nat. Methods 2018 (CLE)
5	BM3D-CLE + gradient	0.671	26.48	0.845	0.84	✓ Certified	Dabov et al., IEEE TIP 2007
6	PINN-CLE + gradient	0.656	26.16	0.836	0.8	✓ Certified	Kang et al., Med. Phys. 2022
7	DnCNN-CLE + gradient	0.649	25.02	0.803	0.89	✓ Certified	Zhang et al., IEEE TIP 2017
8	U-Net-CLE + gradient	0.614	24.45	0.784	0.78	✓ Certified	Andre et al., Med. Image Anal. 2011 (updated DL)
9	NLM-Speckle + gradient	0.610	23.76	0.76	0.84	✓ Certified	Buades et al., CVPR 2005

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%