Artificial intelligence for white matter hyperintensity segmentation: performance, determinants, and clinical implications
Jinhua Hu, Zhihao Zhang, Peng Lei, Jupeng Zhang, Xiqi Zhu, Baosheng Li · 2026
Read Artificial intelligence for white matter hyperintensity segmentation: performance, determinants, and clinical implications by Jinhua Hu, Zhihao Zhang, Peng Lei, Jupeng Zhang, Xiqi Zhu, Baosheng Li (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/pmc-13624168.txt.
Description
Jinhua Hu and five co-authors pool studies from 2026 on computer tools that outline bright spots in brain scans. Those spots, seen in white matter, mark small-vessel disease and link to stroke and memory decline. Drawing them by hand is slow and varies between raters, so teams train models to do the task.
Across twenty-six studies with more than four thousand people, the models scored well on overlap with hand-drawn maps. A net design called U-Net did best. Scan settings seemed to sway scores, and study methods varied a lot. The authors call for shared scan rules and tests across many hospitals. Disease-specific scores for Alzheimer's disease and small-vessel disease were close to the general score.
Catalog record
| CREATOR | Jinhua Hu, Zhihao Zhang, Peng Lei, Jupeng Zhang, Xiqi Zhu, Baosheng Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |