Problem
Research questions and friction points this paper is trying to address.
cell-level representation learning
histopathology images
whole-slide images
computational pathology
biological interpretability
CellDETR: A Detection-Guided Framework for Scalable Cell Representation Learning from Histopathology Images
🤖 AI Summary
Current whole-slide image (WSI) analysis in pathology lacks effective cell-level representation learning, hindering interpretability and clinical translation. This work proposes a detection-guided framework for cell representation learning, introducing Deformable DETR to this domain for the first time. By decoupling positional features and incorporating bounding box-constrained attention mechanisms, the method automatically extracts cell-level embeddings directly from WSIs and leverages contrastive learning for unsupervised pretraining. Using spatial transcriptomics annotations, the approach enables transferable cell classification across datasets, significantly outperforming existing methods on PanNuke under both supervised and unsupervised settings while demonstrating strong biological relevance.
📝 Abstract
Recent advances in pathology foundation models have substantially improved patch and slide level representation learning from whole-slide images (WSIs).However, cell-level representations learning remain underexplored, limiting cell resolved interpretability, biological discovery, and clinical translation. We propose CellDETR, a detection-guided framework built on Deformable DETR for scalable cell representation learning from WSIs. By introducing location feature decoupling and box-constrained attention mechanism, CellDETR enables automated extraction of cell-level embeddings, and outperform existing state-of-the-art methods in supervised cell classification on PanNuke data. In addition, by incorporating contrastive learning design, we build a CellDETR-based pretraining model for scalable cell representation learning from unlabeled WSIs, which improves downstream cell classification performance. Furthermore, we show that after pretraining with Xenium spatial transcriptomics-derived cell annotations, CellDETR achieves accurate cross-dataset cell classification, demonstrating the transferability and biological relevance of the learned cell embeddings. Together, CellDETR provides a scalable route toward general cell-level representation learning framework for interpretable computational patholog
Innovation
Methods, ideas, or system contributions that make the work stand out.
CellDETR
cell representation learning
detection-guided framework
contrastive learning
spatial transcriptomics
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
