A residual neural network using patient-derived anatomical priors outperforms a near-optimal classical per-bundle algorithm in multi-source CT inversion at high attenuation, reaching error ratios of 0.096 on patient data while classical methods cannot cross the single-source performance floor.
First performance evaluation of a dual-source CT (DSCT) system
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Neural-Network Inversion for the Temporal CT Multi-Source Bundle Problem: Per-Bundle Statistical Limits and Near-Optimal Performance
A residual neural network using patient-derived anatomical priors outperforms a near-optimal classical per-bundle algorithm in multi-source CT inversion at high attenuation, reaching error ratios of 0.096 on patient data while classical methods cannot cross the single-source performance floor.