Fusing Pixels and Genes: Spatially-Aware Learning in Computational Pathology

TL;DR

This paper introduces STAMP, a novel framework that integrates spatial transcriptomics with pathology images to improve multimodal learning, leveraging large-scale spatial gene expression data for better molecular and spatial understanding.

Contribution

The paper presents STAMP, a new spatially-aware multimodal learning method that combines gene expression profiles with pathology images, supported by the largest spatial transcriptomics dataset to date.

Findings

STAMP outperforms existing models on multiple datasets

Spatial context and multi-scale info improve performance

Gene-guided training enhances representation robustness

Abstract

Recent years have witnessed remarkable progress in multimodal learning within computational pathology. Existing models primarily rely on vision and language modalities; however, language alone lacks molecular specificity and offers limited pathological supervision, leading to representational bottlenecks. In this paper, we propose STAMP, a Spatial Transcriptomics-Augmented Multimodal Pathology representation learning framework that integrates spatially-resolved gene expression profiles to enable molecule-guided joint embedding of pathology images and transcriptomic data. Our study shows that self-supervised, gene-guided training provides a robust and task-agnostic signal for learning pathology image representations. Incorporating spatial context and multi-scale information further enhances model performance and generalizability. To support this, we constructed SpaVis-6M, the largest Visium-based spatial transcriptomics dataset to date, and trained a spatially-aware gene encoder on this resource. Leveraging hierarchical multi-scale contrastive alignment and cross-scale patch localization mechanisms, STAMP effectively aligns spatial transcriptomics with pathology images, capturing spatial structure and molecular variation. We validate STAMP across six datasets and four downstream tasks, where it consistently achieves strong performance. These results highlight the value and necessity of integrating spatially resolved molecular supervision for advancing multimodal learning in computational pathology. The code is included in the supplementary materials. The pretrained weights and SpaVis-6M are available at: https://github.com/Hanminghao/STAMP.