AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Imaging with Confidence: Uncertainty Quantification for High-dimensional Undersampled MR Images

Frederik Hoppe, Claudio Mayrink Verdun, Hannah Sophie Laus, Sebastian Endt, Marion Irene Menzel, Felix Krahmer, Holger Rauhut

ECCV 2024 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Priors and functional variational uncertainty · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
debias the TV-regularized reconstruction before building pixel-wise confidence intervalsTV minimization for undersampled MRI · In vivo brain MRI, spiral and radial samplingempirical coverage of 95% intervals: 93.82% (spiral, all pixels); 94.20% (radial, all pixels)Improved
debias the TV estimator so the error is dominated by a Gaussian termTV minimization for undersampled MRI · In vivo brain MRIstructure of the reconstruction errorImproved
debias the TV estimator (effect on l2 error)TV minimization for undersampled MRI · In vivo brain MRIl2 reconstruction errorHurt
increase the number of k-space measurements (ablation)debiased TV confidence intervals · In vivo brain MRIconfidence interval qualityImproved
use data with more than 20% noise (ablation)debiased TV confidence intervals · very noisy MRI regimeusefulness of intervalsHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy