Segmenting with confidence through uncertainty quantification for brain tumor imaging
Yassine Guennoun, Pierre Nedelec, Mark McArthur, Evan Bloch, Jinchi Wei, Leo Sugrue, Evan Calabrese, Andreas M. Rauschecker · 2026
Read Segmenting with confidence through uncertainty quantification for brain tumor imaging by Yassine Guennoun, Pierre Nedelec, Mark McArthur, Evan Bloch, Jinchi Wei, Leo Sugrue, Evan Calabrese, Andreas M. Rauschecker (2026) 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/PMC13639024.txt.
Description
Yassine Guennoun and seven colleagues describe a 2026 deep learning system that outlines brain tumors on MRI scans while also reporting how unsure it is. Clinicians hesitate to trust automated tumor outlines that give no sense of their own uncertainty.
Trained on 1,655 scans from 788 patients, the system segmented meningiomas with a median Dice score of 0.93 on held-out scans, and its doubt maps matched areas that radiologists found ambiguous. A separate group of 353 patients confirmed the accuracy. The authors argue that honest uncertainty estimates can make medical imaging software safer to use in daily practice.
Catalog record
| CREATOR | Yassine Guennoun, Pierre Nedelec, Mark McArthur, Evan Bloch, Jinchi Wei, Leo Sugrue, Evan Calabrese, Andreas M. Rauschecker |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64566 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |