Random features meet MIL: a deep GP approach to colorectal MSI prediction
Shixuan Shen, Zeyang Wang, Tianmu Liu, Kangle Ma, Zhen Tian, Fuqiang Zhang, Qingyue Zhang · 2025
Read Random features meet MIL: a deep GP approach to colorectal MSI prediction by Shixuan Shen, Zeyang Wang, Tianmu Liu, Kangle Ma, Zhen Tian, Fuqiang Zhang, Qingyue Zhang (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-12800130.txt.
Description
Shixuan Shen and co-authors predict a genetic trait of colorectal tumors from slide images in a 2025 paper. The trait, microsatellite instability, guides therapy, but testing every case costs time and money. Slides are labeled only as whole cases, a weak supervision problem with huge images.
Their model couples a deep Gaussian process with multiple instance learning, using random features to keep it tractable. An attention step highlights the slide regions driving each call, aiding trust. On a public cancer dataset the method beat prior models on standard scoring. Image-based triage could reserve lab tests for the cases that truly need them.
Catalog record
| CREATOR | Shixuan Shen, Zeyang Wang, Tianmu Liu, Kangle Ma, Zhen Tian, Fuqiang Zhang, Qingyue Zhang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 88 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |