ImmuVis : Hyperconvolutional Foundation Model for Imaging Mass Cytometry
Marcin Możejko, Dawid Uchal, Krzysztof Gogolewski, Piotr Kupidura, Szymon Łukasik, Jakub Giezgała, Tomasz Nocoń, Kacper Pietrzyk, Robert Pieniuta, Mateusz Sulimowicz, Michal Orzyłowski, Tomasz Siłkowski, Karol Zagródka, Eike Staub, Ewa Szczurek · 2026
Read ImmuVis : Hyperconvolutional Foundation Model for Imaging Mass Cytometry by Marcin Możejko, Dawid Uchal, Krzysztof Gogolewski, Piotr Kupidura, Szymon Łukasik, Jakub Giezgała, Tomasz Nocoń, Kacper Pietrzyk, Robert Pieniuta, Mateusz Sulimowicz, Michal Orzyłowski, Tomasz Siłkowski, Karol Zagródka, Eike Staub, Ewa Szczurek (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/arxiv-2602-04585.txt.
Description
Imaging mass cytometry lets researchers map many molecular markers across tissue at once, but each study measures a varied marker set. Dawid Uchal and 14 co-authors present ImmuVis in a 2026 paper, a base model built for that shifting channel space.
Special hyperconvolutions make their filters from learned marker descriptions, so one model handles any measured subset with no retraining. Pretraining used a very large set of tissue images and markers through masked reconstruction, a self-supervised task. ImmuVis beat leading baselines on virtual staining and classification while using far less compute than transformer rivals. It also gave calibrated uncertainty estimates, useful when judging how much to trust a prediction.
Catalog record
| CREATOR | Marcin Możejko, Dawid Uchal, Krzysztof Gogolewski, Piotr Kupidura, Szymon Łukasik, Jakub Giezgała, Tomasz Nocoń, Kacper Pietrzyk, Robert Pieniuta, Mateusz Sulimowicz, Michal Orzyłowski, Tomasz Siłkowski, Karol Zagródka, Eike Staub, Ewa Szczurek |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 49 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-04585