Personalized Cell Segmentation: Benchmark and Framework for Reference-Guided Cell Type Segmentation

TL;DR

This paper introduces the Personalized Cell Segmentation (PerCS) task, a benchmark, and a novel framework, PerCS-DINO, for segmenting specific cell types based on reference images, advancing cell segmentation capabilities.

Contribution

The paper defines the PerCS task, creates a benchmark dataset, and proposes PerCS-DINO, a new deep learning framework utilizing cross-attention and contrastive learning for reference-guided cell segmentation.

Findings

PerCS-DINO outperforms existing models on the benchmark.

The benchmark includes 1,372 images and over 110,000 annotated cells.

The task highlights new challenges in cell type-specific segmentation.

Abstract

Accurate cell segmentation is critical for biological and medical imaging studies. Although recent deep learning models have advanced this task, most methods are limited to generic cell segmentation, lacking the ability to differentiate specific cell types. In this work, we introduce the Personalized Cell Segmentation (PerCS) task, which aims to segment all cells of a specific type given a reference cell. To support this task, we establish a benchmark by reorganizing publicly available datasets, yielding 1,372 images and over 110,000 annotated cells. As a pioneering solution, we propose PerCS-DINO, a framework built on the DINOv2 backbone. By integrating image features and reference embeddings via a cross-attention transformer and contrastive learning, PerCS-DINO effectively segments cells matching the reference. Extensive experiments demonstrate the effectiveness of the proposed PerCS-DINO and highlight the challenges of this new task. We expect PerCS to serve as a useful testbed for advancing research in cell-based applications.