OCELOT 2023: Cell Detection from Cell-Tissue Interaction Challenge

2025-09-11

AI Summary

Existing cell detection models fail to capture pathologists’ collaborative multi-scale cognitive process—simultaneously evaluating tissue morphology and cellular details—and lack annotated data supporting cross-scale semantic learning. Method: We introduce the first multi-organ whole-slide image dataset featuring overlapping cell- and tissue-level annotations across multiple scales; propose a cell-tissue joint modeling paradigm, empirically validating the critical role of their interaction in detection performance; and design a multi-scale feature fusion module with cross-hierarchical interactive attention to explicitly model biologically grounded dependencies between cells and tissues. Contribution/Results: On the test set, our best-performing model achieves a 7.99-point improvement in F1-score over single-task baselines and significantly outperforms conventional methods. This work establishes a novel, interpretable, multi-scale paradigm for computational pathology analysis.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Pathologists routinely alternate between different magnifications when examining Whole-Slide Images, allowing them to evaluate both broad tissue morphology and intricate cellular details to form comprehensive diagnoses. However, existing deep learning-based cell detection models struggle to replicate these behaviors and learn the interdependent semantics between structures at different magnifications. A key barrier in the field is the lack of datasets with multi-scale overlapping cell and tissue annotations. The OCELOT 2023 challenge was initiated to gather insights from the community to validate the hypothesis that understanding cell and tissue (cell-tissue) interactions is crucial for achieving human-level performance, and to accelerate the research in this field. The challenge dataset includes overlapping cell detection and tissue segmentation annotations from six organs, comprising 673 pairs sourced from 306 The Cancer Genome Atlas (TCGA) Whole-Slide Images with hematoxylin and eosin staining, divided into training, validation, and test subsets. Participants presented models that significantly enhanced the understanding of cell-tissue relationships. Top entries achieved up to a 7.99 increase in F1-score on the test set compared to the baseline cell-only model that did not incorporate cell-tissue relationships. This is a substantial improvement in performance over traditional cell-only detection methods, demonstrating the need for incorporating multi-scale semantics into the models. This paper provides a comparative analysis of the methods used by participants, highlighting innovative strategies implemented in the OCELOT 2023 challenge.

Problem

• Existing cell detection models lack multi-scale tissue interaction understanding
• No datasets with overlapping cell-tissue annotations across different magnifications
• Models need to incorporate cell-tissue relationships for human-level performance

Innovation

• Multi-scale overlapping cell-tissue annotations
• Incorporating cell-tissue interaction relationships
• Fusing multi-scale semantics for detection