Public

SPECT/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
ct_registration_shift -5.0 – 10.0 2.5 pixels
hu_to_mu_scale -12.0 – 24.0 6.0 %
scatter_fraction -0.35 – 0.7 0.175
collimator_blur -1.0 – 9.5 4.25 pixels

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

18.94 dB

SSIM 0.8860

Scenario II (Mismatch)

18.92 dB

SSIM 0.8920

Scenario III (Oracle)

24.85 dB

SSIM 0.9356

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 19.41 0.8948 20.28 0.9037 25.26 0.9319
scene_01 14.84 0.8471 14.95 0.8690 21.47 0.9436
scene_02 21.65 0.9006 20.03 0.8849 27.33 0.9330
scene_03 19.87 0.9014 20.43 0.9103 25.33 0.9340

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 DL-SPECT + gradient 0.807 34.38 0.964 0.89 ✓ Certified Ramon et al., IEEE TMI 2020
2 MAP-OSEM + gradient 0.784 32.07 0.943 0.93 ✓ Certified Nuyts et al., 2002
3 AC-OSEM + gradient 0.710 27.88 0.878 0.9 ✓ Certified CT-based attenuation correction
4 OSEM + gradient 0.576 21.91 0.686 0.91 ✓ Certified Hudson & Larkin, IEEE TMI 1994

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%
Back to SPECT/CT