Abstract:Multiplexed imaging technologies have made it possible to interrogate complex tumor microenvironments at sub-cellular resolution within their native spatial context. However, proper quantification of this complexity requires the ability to easily and accurately segment cells into their sub-cellular compartments. Within the supervised learning paradigm, deep learning based segmentation methods demonstrating human level performance have emerged. However, limited work has been done in developing such generalist methods within the label-free unsupervised context. Here we present an unsupervised segmentation (UNSEG) method that achieves deep learning level performance without requiring any training data. UNSEG leverages a Bayesian-like framework and the specificity of nucleus and cell membrane markers to construct an probability estimate of each pixel belonging to the nucleus, cell membrane, or background. It uses this estimate to segment each cell into its nuclear and cell-membrane compartments. We show that UNSEG is more internally consistent and better at generalizing to the complexity of tissue morphology than current deep learning methods. This allows UNSEG to unambiguously identify the cytoplasmic compartment of a cell, which we employ to demonstrate its use in an exemplar biological scenario. Within the UNSEG framework, we also introduce a new perturbed watershed algorithm capable of stably and automatically segmenting a cluster of cell nuclei into individual cell nuclei that increases the accuracy of classical watershed. Perturbed watershed can also be used as a standalone algorithm that researchers can incorporate within their supervised or unsupervised learning approaches to extend classical watershed, particularly in the multiplexed imaging context. Finally, as part of developing UNSEG, we have generated a high-quality annotated gastrointestinal tissue (GIT) dataset, which we anticipate will be useful for the broader research community. We demonstrate the efficacy of UNSEG on the GIT dataset, publicly available datasets, and on a range of practical scenarios. In these contexts, we also discuss the possibility of bias inherent in quantification of segmentation accuracy based on score. Segmentation, despite its long antecedents, remains a challenging problem, particularly in the context of tissue samples. UNSEG, an easy-to-use algorithm, provides an unsupervised approach to overcome this bottleneck, and as we discuss, can help improve deep learning based segmentation methods by providing a bridge between unsupervised and supervised learning paradigms.

Semi-Supervised Semantic Segmentation of Cell Nuclei via Diffusion-based Large-Scale Pre-Training and Collaborative Learning

Instance-aware Self-supervised Learning for Nuclei Segmentation

Evaluation of semi-supervised learning using sparse labeling to segment cell nuclei

Diffusion-based Data Augmentation for Nuclei Image Segmentation

Point-supervised Single-cell Segmentation via Collaborative Knowledge Sharing

Cross Comparison Representation Learning for Semi-Supervised Segmentation of Cellular Nuclei in Immunofluorescence Staining.

Advanced Multi-Microscopic Views Cell Semi-supervised Segmentation

Nuclei Segmentation with Point Annotations from Pathology Images via Self-Supervised Learning and Co-Training

Cascaded Diffusion Models for 2D and 3D Microscopy Image Synthesis to Enhance Cell Segmentation

Auto-CSC: A Transfer Learning Based Automatic Cell Segmentation and Count Framework

Efficient end-to-end learning for cell segmentation with machine generated weak annotations

UNSEG: unsupervised segmentation of cells and their nuclei in complex tissue samples

A novel deep learning-based 3D cell segmentation framework for future image-based disease detection

CellSAM: Advancing Pathologic Image Cell Segmentation via Asymmetric Large‐Scale Vision Model Feature Distillation Aggregation Network

Label Propagation for Annotation-Efficient Nuclei Segmentation from Pathology Images

Learn to segment single cells with deep distance estimator and deep cell detector

Semi-Supervised Neuron Segmentation via Reinforced Consistency Learning

Weakly Supervised Deep Nuclei Segmentation With Sparsely Annotated Bounding Boxes for DNA Image Cytometry