Saved in:
Bibliographic Details
Main Authors: Uchal, Dawid, Możejko, Marcin, Gogolewski, Krzysztof, Kupidura, Piotr, Łukasik, Szymon, Giezgała, Jakub, Nocoń, Tomasz, Pietrzyk, Kacper, Pieniuta, Robert, Sulimowicz, Mateusz, Orzyłowski, Michal, Siłkowski, Tomasz, Zagródka, Karol, Staub, Eike, Szczurek, Ewa
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.04585
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916010995482624
author Uchal, Dawid
Możejko, Marcin
Gogolewski, Krzysztof
Kupidura, Piotr
Łukasik, Szymon
Giezgała, Jakub
Nocoń, Tomasz
Pietrzyk, Kacper
Pieniuta, Robert
Sulimowicz, Mateusz
Orzyłowski, Michal
Siłkowski, Tomasz
Zagródka, Karol
Staub, Eike
Szczurek, Ewa
author_facet Uchal, Dawid
Możejko, Marcin
Gogolewski, Krzysztof
Kupidura, Piotr
Łukasik, Szymon
Giezgała, Jakub
Nocoń, Tomasz
Pietrzyk, Kacper
Pieniuta, Robert
Sulimowicz, Mateusz
Orzyłowski, Michal
Siłkowski, Tomasz
Zagródka, Karol
Staub, Eike
Szczurek, Ewa
contents We present ImmuVis, a family of efficient foundation models for imaging mass cytometry (IMC), a high-throughput multiplex imaging technology that handles molecular marker measurements as image channels and enables large-scale spatial tissue profiling. Unlike natural images, multiplex imaging lacks a fixed channel space, as real-world marker sets vary across studies, violating a core assumption of standard vision backbones. To address this, ImmuVis introduces marker-adaptive hyperconvolutions that generate convolutional kernels from learned marker embeddings, enabling a single model to operate on arbitrary measured marker subsets without retraining. We pretrain ImmuVis on the largest dataset to date, IMC17M (28 cohorts, 24,405 images, 265 markers, over 17M patches), using self-supervised masked reconstruction. ImmuVis outperforms state-of-the-art baselines and ablations in virtual staining and downstream classification tasks at substantially lower compute cost than transformer-based alternatives, and is the sole model that provides calibrated uncertainty via a heteroscedastic likelihood objective. These results position ImmuVis as a practical framework for real-world IMC modeling.
format Preprint
id arxiv_https___arxiv_org_abs_2602_04585
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ImmuVis: Hyperconvolutional Foundation Model for Imaging Mass Cytometry
Uchal, Dawid
Możejko, Marcin
Gogolewski, Krzysztof
Kupidura, Piotr
Łukasik, Szymon
Giezgała, Jakub
Nocoń, Tomasz
Pietrzyk, Kacper
Pieniuta, Robert
Sulimowicz, Mateusz
Orzyłowski, Michal
Siłkowski, Tomasz
Zagródka, Karol
Staub, Eike
Szczurek, Ewa
Computer Vision and Pattern Recognition
We present ImmuVis, a family of efficient foundation models for imaging mass cytometry (IMC), a high-throughput multiplex imaging technology that handles molecular marker measurements as image channels and enables large-scale spatial tissue profiling. Unlike natural images, multiplex imaging lacks a fixed channel space, as real-world marker sets vary across studies, violating a core assumption of standard vision backbones. To address this, ImmuVis introduces marker-adaptive hyperconvolutions that generate convolutional kernels from learned marker embeddings, enabling a single model to operate on arbitrary measured marker subsets without retraining. We pretrain ImmuVis on the largest dataset to date, IMC17M (28 cohorts, 24,405 images, 265 markers, over 17M patches), using self-supervised masked reconstruction. ImmuVis outperforms state-of-the-art baselines and ablations in virtual staining and downstream classification tasks at substantially lower compute cost than transformer-based alternatives, and is the sole model that provides calibrated uncertainty via a heteroscedastic likelihood objective. These results position ImmuVis as a practical framework for real-world IMC modeling.
title ImmuVis: Hyperconvolutional Foundation Model for Imaging Mass Cytometry
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.04585