Skip to content
Open access

Fine-Grained and Reliable Cell Classification from Pathology Images

Oct 2026 · bioRxiv · 0 citations · 31 references
Biology

TL;DR

CellBridge is introduced, a multi-modal contrastive learning framework pretrained on 16M cells with paired H&E and spatial profiles, for transferring molecular information into morphological representations that consistently achieves smaller prediction sets, while reducing class-wise coverage gaps.

Abstract

Molecular assays such as spatial transcriptomics facilitate fine-grained identification of cell types for analyzing tumor microenvironments. However, they are expensive to acquire in routine clinical practice compared to hematoxylin and eosin (H&E) images. With molecular guidance, such images can distinguish cell lineages, yet often fail to resolve closely related subtypes. To overcome this, we introduce CellBridge, a multi-modal contrastive learning framework pretrained on 16M cells with paired H&E and spatial profiles, for transferring molecular information into morphological representations. CellBridge further incorporates a Cell-Ontology-aware conformal prediction module that assigns labels at different levels of specificity: leaf-level cell types when morphology alone provides sufficient evidence, and broader ancestor categories when distinguishing among more specific cell types would require molecular information. Across in-domain and out-of-domain benchmarks, CellBridge outperforms leading pathology foundation models in macro F1 by up to 16 percentage points, with the largest gains on the most fine-grained label spaces. It consistently achieves smaller prediction sets, while reducing class-wise coverage gaps. We envision CellBridge as a step toward extracting cellular information that currently requires costly molecular assays from routinely acquired pathology images, while explicitly identifying where such assays remain necessary. Our code is available at https://github.com/Lotfollahilab/cellbridge.

Read PDF

Similar papers

Sep 2026

Towards Scalable Context-Aware Single-Cell Spatial Transcriptomics Prediction from Histology Images

Predicting gene expression from H&E-stained histology images offers a scalable alternative to costly spatial transcriptomics, yet most existing methods operate at the spot level, where signals from multiple cells are aggregated and critical cellular heterogeneity is obscured. Extending this paradigm to single-cell reso...

Zi-Jun Gao, Chun-Bin Gu, Jin-Xi Xiang et al. · 1 citation
Review Open access Sep 2026

Integrating AI-Based Digital Pathology with Spatial Transcriptomics for Deeper Insight.

It is argued that the most credible translational goal is not to infer every transcript from H&E, but to identify biologically coherent spatial states with reproducible morphological correlates that can be validated across independent platforms and cohorts before being mapped across large archival collections.

M. A. Rezania, Haechan Yi, Jae-Il Choi et al. · 0 citations
Review Open access Oct 2026

Beyond Segmentation Accuracy: Evaluating Biological Fidelity in Spatial Pathology

It is argued that cell segmentation for spatial pathology should be evaluated not only by boundary and instance accuracy, but also by preservation of the biological relationships that downstream analyses are intended to measure.

Nevenka Dimitrova, Marjan Stoimchev, A. Filliol et al. · 0 citations

Multimodal Pseudo-Ground Truth for Deep Learning-Based Whole-Cell Segmentation in H&E Images with Spatial Transcriptomic Mapping

A multimodal pseudo-ground truth training technique to enable supervised feature learning on H&E images for whole-cell segmentation and adapted into a ST pipeline to create an integrated end-to-end machine learning solution, without additional optimisation on ST data.

A.-A. Zhang, Ulrich D. Kadolsky, A. Saxena · 0 citations
Open access Oct 2026

Single-Cell Multiplex Immune Biomarker Prediction from Histopathology Across Cancer Types and Clinical Endpoints

Background The spatial organization of cellular lineages within the tumor microenvironment governs disease progression and therapeutic response, but multiplexed spatial profiling remains costly, technically complex, and tissue-destructive. Routine hematoxylin and eosin (H&E) histopathology is inexpensive and widely ava...

Usman Afzaal, Usama Sajjad, A. Akbar et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.