U-Net accelerated MRI yields higher PSNR and SSIM but lower AUC on a Rayleigh discrimination task than fully sampled rSOS at the same acceleration factors.
Estimating task- based performance bounds for accelerated MRI image reconstruction methods by use of learned-ideal observers,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.med-ph 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Evaluating the resolution of AI-based accelerated MR reconstruction using a deep learning-based model observer
U-Net accelerated MRI yields higher PSNR and SSIM but lower AUC on a Rayleigh discrimination task than fully sampled rSOS at the same acceleration factors.