Abstract:Background: Light microscopy is a widely used technique in cell biology due to its satisfactory resolution for cellular structure analysis, prevalent availability of fluorescent probes for staining, and compatibility for the dynamic analysis of live cells. However, the segmentation of cells and nuclei from microscopic images is not a straightforward process because it has several challenges such as high variation in morphology and shape, the presence of noise and diverse contrast in backgrounds, clustering or overlapping nature of cells. Dealing with these challenges and facilitating more reliable analysis necessitates the implementation of computer-aided methods that leverage image processing techniques and deep learning algorithms. The major goal of this study is to propose a model, for instance segmentation of cells and nuclei, applying the most cutting-edge deep learning techniques. Methods: A fine-tuned You Only Look at Once version 9 extended (YOLOv9-E) model is initially applied as a prompt generator to generate bounding box prompts. Using the generated prompts, a pre-trained segment anything model (SAM) is subsequently applied through zero-short inferencing to produce raw segmentation masks. These segmentation masks are then refined using non-max suppression and simple image processing methods such as image addition and morphological processing. The proposed method is developed and evaluated using an open-sourced dataset called Expert Visual Cell Annotation (EVICAN), which is relatively large and contains 4,738 microscopy images extracted from cross organs using different protocols. Results: Based on the evaluation results on three different levels of EVICAN test sets, the proposed method demonstrates noticeable performances showing average mAP50 [mean average precision at intersection over union (IoU) =0.50] scores of 96.25, 95.05, and 94.18 for cell segmentation, and 68.04, 54.66, and 38.29 for nucleus segmentation on easy, medium, and difficult test sets, respectively. Conclusions: Our proposed method for instance segmentation of cells and nuclei provided favorable performance compared to the existing methods in literature, indicating its potential utility as an assistive tool for cell culture experts, facilitating prompt and reliable analysis.

Split and Expand: An inference-time improvement for Weakly Supervised Cell Instance Segmentation

Joint Feature Learning for Cell Segmentation Based on Multi-scale Convolutional U-Net.

Scellseg: a Style-Aware Cell Instance Segmentation Tool with Pre-Training and Contrastive Fine-Tuning

LKCell: Efficient Cell Nuclei Instance Segmentation with Large Convolution Kernels

Look in Different Views: Multi-Scheme Regression Guided Cell Instance Segmentation

DeepSplit: Segmentation of Microscopy Images Using Multi-task Convolutional Networks

CausalCellSegmenter: Causal Inference inspired Diversified Aggregation Convolution for Pathology Image Segmentation

MoE-NuSeg: Enhancing Nuclei Segmentation in Histology Images with a Two-Stage Mixture of Experts Network

Weakly Supervised Cell Instance Segmentation by Propagating from Detection Response

EnNuSegNet: Enhancing Weakly Supervised Nucleus Segmentation through Feature Preservation and Edge Refinement

Point-supervised Single-cell Segmentation via Collaborative Knowledge Sharing

CellSegNet: an Adaptive Multi-Resolution Hybrid Network for Cell Segmentation

Instance segmentation of cells and nuclei from multi-organ cross-protocol microscopic images

Cell segmentation for immunofluorescence multiplexed images using two-stage domain adaptation and weakly labeled data for pre-training

InstanSeg: an embedding-based instance segmentation algorithm optimized for accurate, efficient and portable cell segmentation

A dual decoder U-Net-based model for nuclei instance segmentation in hematoxylin and eosin-stained histological images

BIDCell: Biologically-informed self-supervised learning for segmentation of subcellular spatial transcriptomics data

Scellseg: A style-aware deep learning tool for adaptive cell instance segmentation by contrastive fine-tuning

Separable-HoverNet and Instance-YOLO for Colon Nuclei Identification and Counting

Weakly supervised segmentation via instance-aware propagation

DoNet: Deep De-overlapping Network for Cytology Instance Segmentation