Frequency-domain multi-scale hybrid attention for pathological image classification
Yang Zhang, Junjie Li, Qiushi Wang, Luyan Zhang, Lingjiang Deng, Qiyun Tian, Feixu Lu, Gang Li, Chuanyun Xu, Fanghong Zhang, Hualiang Xiao · 2026
Read Frequency-domain multi-scale hybrid attention for pathological image classification by Yang Zhang, Junjie Li, Qiushi Wang, Luyan Zhang, Lingjiang Deng, Qiyun Tian, Feixu Lu, Gang Li, Chuanyun Xu, Fanghong Zhang, Hualiang Xiao (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/PMC13574826.txt.
Description
Yang Zhang, with ten colleagues, presents a 2026 attention scheme for sorting pathology images, the microscope slides used to spot cancer early. Such slides mix tangled textures and wide variety, and most tools lean on local detail alone. Frequency views, which split an image into wave bands, rarely let those bands talk to each other.
The scheme, called MFHA, blends attention across scales in the frequency view. Coarse bands carry the broad layout of tissue, while fine bands hold cell-level clues. Letting the bands interact aims to fuse both into one read. The authors pitch the fusion as a fix for missed detail in hard cases. Finer sorting of slide types guides treatment choice, so gains here feed straight into care. The work targets labs where expert slide readers are scarce.
Catalog record
| CREATOR | Yang Zhang, Junjie Li, Qiushi Wang, Luyan Zhang, Lingjiang Deng, Qiyun Tian, Feixu Lu, Gang Li, Chuanyun Xu, Fanghong Zhang, Hualiang Xiao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62071 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |