Knowledge-guided adaptation of pathology foundation models effectively improves cross-domain generalization and demographic fairness
Yanyan Huang, Weiqin Zhao, Zhengyu Zhang, Yihang Chen, Yu Fu, Feng Wu, Yuming Jiang, Li Liang, Shujun Wang, Lequan Yu · 2025
Read Knowledge-guided adaptation of pathology foundation models effectively improves cross-domain generalization and demographic fairness by Yanyan Huang, Weiqin Zhao, Zhengyu Zhang, Yihang Chen, Yu Fu, Feng Wu, Yuming Jiang, Li Liang, Shujun Wang, Lequan Yu (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/pmc-12749728.txt.
Description
Yanyan Huang and co-authors adapt pathology AI for fairer use across hospitals in 2025. Big models trained on tissue slides often falter at new sites. They also score unevenly across patient groups, which raises fairness worries. Their framework, called FLEX, filters each task signal through a knowledge guided bottleneck.
The filter keeps sturdy disease traits and drops site and group artifacts. Tests spanned three large cohorts and many clinical tasks with thousands of slides. The adapted model held up better at unseen hospitals than baselines. Fairness checks showed smaller gaps between demographic groups as well.
Catalog record
| CREATOR | Yanyan Huang, Weiqin Zhao, Zhengyu Zhang, Yihang Chen, Yu Fu, Feng Wu, Yuming Jiang, Li Liang, Shujun Wang, Lequan Yu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 81 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |