Uncertainty Quantification in Medical Image Segmentation: A Comprehensive Survey
Seyed Sina Ziaee, Katie Ovens · 2026
Read Uncertainty Quantification in Medical Image Segmentation: A Comprehensive Survey by Seyed Sina Ziaee, Katie Ovens (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/PMC13514988.txt.
Description
Seyed Sina Ziaee and Katie Ovens survey, in a 2026 review, methods for measuring uncertainty in artificial intelligence systems that outline structures in medical scans. Modern neural networks segment tumours and organs impressively, yet they hand back a single answer with no hint of doubt, which can breed overconfidence where an image is ambiguous. Their survey sorts the field into Bayesian methods, ensembles of models, deterministic approaches, test-time augmentation, and hybrids. Foundation-model techniques are treated as a cross-cutting theme.
Standard tools receive close treatment, including Monte Carlo dropout and Bayesian neural networks. Other methods separate uncertainty from noise in the data itself from uncertainty born of limited knowledge. Applications span brain tumour scans, heart imaging, and lung nodule detection, while evaluation turns on calibration and how well uncertainty tracks actual errors.
Catalog record
| CREATOR | Seyed Sina Ziaee, Katie Ovens |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |