Mind the Detail: Uncovering Clinically Relevant Image Details in Accelerated MRI with Semantically Diverse Reconstructions
Jan Nikolas Morshuis, Christian Schlarmann, Thomas Küstner, Christian F. Baumgartner *, Matthias Hein * · 2025
Read Mind the Detail: Uncovering Clinically Relevant Image Details in Accelerated MRI with Semantically Diverse Reconstructions by Jan Nikolas Morshuis, Christian Schlarmann, Thomas Küstner, Christian F. Baumgartner *, Matthias Hein * (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-00670.txt.
Description
Jan Nikolas Morshuis and co-authors warned in a 2025 paper that fast MRI can erase the very detail doctors need. Deep learning rebuilds sharp scans from sparse data, scoring well on image quality. Yet a rebuilt image can look fine while a small, rare lesion vanishes, steering diagnosis astray.
The team show that even diffusion-based rebuilds, with repeat sampling, miss such details. Their fix draws several semantically diverse rebuilds of the same scan. Where the versions disagree, doubt is flagged instead of hidden. A truth score then rates each version. The aim is scans that confess their weak spots.
Catalog record
| CREATOR | Jan Nikolas Morshuis, Christian Schlarmann, Thomas Küstner, Christian F. Baumgartner *, Matthias Hein * |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 20 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-00670