ImmuVis: Hyperconvolutional Foundation Model for Imaging Mass Cytometry

2026-02-04

AI Summary

This work addresses the challenge of limited generalizability in existing visual models for imaging mass cytometry (IMC), which arises from heterogeneous molecular marker subsets used across studies. To overcome this, the authors propose ImmuVis, a convolutional foundation model pretrained via self-supervised masked reconstruction. ImmuVis introduces a novel marker-adaptive hyper-convolution mechanism that dynamically generates convolutional kernels from marker embeddings, enabling inference with arbitrary marker combinations without retraining. Furthermore, it incorporates a heteroscedastic likelihood objective to calibrate predictive uncertainty. Evaluated on virtual staining and downstream classification tasks, ImmuVis substantially outperforms current state-of-the-art methods while maintaining significantly lower computational costs than Transformer-based architectures. Its efficiency and strong generalization capability are validated on the IMC17M dataset.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Data Mining & Knowledge Management: Mining of Visual, Multimedia & Multimodal Data

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

We present ImmuVis, an efficient convolutional foundation model 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 to-date dataset, IMC17M (28 cohorts, 24,405 images, 265 markers, over 17M patches), using self-supervised masked reconstruction. ImmuVis outperforms SOTA 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, efficient foundation model for real-world IMC modeling.

Problem

• imaging mass cytometry
• variable marker channels
• foundation model
• multiplex imaging
• channel heterogeneity

Innovation

• hyperconvolution
• marker-adaptive
• imaging mass cytometry
• foundation model
• self-supervised pretraining