Fourier Transform Multiple Instance Learning for whole slide image classification
FFT-MIL establishes frequency-domain learning as an effective and efficient mechanism for capturing global dependencies in WSI classification, complementing spatial features and advancing the scalability and accuracy of MIL-based computational pathology. The source code is publicly available at https://github.com/irulenot/FFT-MIL.
Authors
- Dexter Hadley (ORCID: https://orcid.org/0000-0003-0990-4674)
- Laura J. Brattain
- Anthony Bilic
- Ming Li (ORCID: https://orcid.org/0000-0001-6162-6291)
- Chen Chen (ORCID: https://orcid.org/0000-0002-3525-9755)
- Wei Zhang (ORCID: https://orcid.org/0000-0003-3605-9373)
- Yu Tian (ORCID: https://orcid.org/0009-0000-6623-1491)
- Guangyu Sun (ORCID: https://orcid.org/0000-0001-6436-0820)
- Md. Sanzid Bin Hossain
Institutions
- University of Central Florida (US)
- Artificial Intelligence in Medicine (Canada) (CA)
- Florida College (US)
Publication Details
- Journal
- Journal of Medical Imaging
- Published
- 2025-10-22
- DOI
- https://doi.org/10.1117/1.jmi.12.6.061409
- Primary Topic
- Digital Imaging for Blood Diseases
- Type
- article
- Field-Weighted Citation Impact
- 0.00