LEMON: a foundation model for nuclear morphology in Computational Pathology

2026-03-26

AI Summary

Single-cell-level representation learning remains underdeveloped in computational pathology, limiting precise characterization of cell types and phenotypes. To address this gap, this work proposes LEMON—the first large-scale self-supervised foundation model tailored for nuclear morphology—trained on millions of single-cell images spanning multiple tissues and cancer types to learn scalable, generalizable, and robust representations. LEMON significantly outperforms existing methods across multiple tasks on five benchmark datasets, demonstrating strong generalization capabilities and substantial potential for large-scale single-cell analysis in computational pathology.

Abstract

Computational pathology relies on effective representation learning to support cancer research and precision medicine. Although self-supervised learning has driven major progress at the patch and whole-slide image levels, representation learning at the single-cell level remains comparatively underexplored, despite its importance for characterizing cell types and cellular phenotypes. We introduce LEMON (Learning Embeddings from Morphology Of Nuclei), a self-supervised foundation model for scalable single-cell image representation learning. Trained on millions of cell images from diverse tissues and cancer types, LEMON learns robust and versatile morphological representations that support large-scale single-cell analyses in pathology. We evaluate LEMON on five benchmark datasets across a range of prediction tasks and show that it provides strong performance, highlighting its potential as a new paradigm for cell-level computational pathology. Model weights are available at https://huggingface.co/aliceblondel/LEMON.

Problem

• computational pathology
• single-cell representation learning
• nuclear morphology
• self-supervised learning
• cellular phenotypes

Innovation

• foundation model
• self-supervised learning
• single-cell representation
• nuclear morphology
• computational pathology