Histopathology-centered computational evolution of spatial omics: integration, mapping, and foundation models.

Abstract

Spatial omics (SO) enables spatially resolved molecular profiling, while hematoxylin and eosin (H&E) imaging remains the gold standard for morphological assessment in clinical pathology. Recent computational advances increasingly center H&E images in SO analysis and push resolution toward the single-cell level. We systematically review the computational evolution of SO from a histopathology-centered perspective, organizing methods into three paradigms: integration (jointly modeling of paired multimodal data), mapping (inferring molecular profiles from H&E images), and foundation models (learning generalizable representations from large-scale datasets). We summarize actionable modeling directions and persistent gaps, providing a roadmap for developing, and applying computational frameworks in SO.