Cell Detection with Star-convex Polygons

Abstract

Automatic detection and segmentation of cells and nuclei in microscopy images is important for many biological applications. Recent successful learning-based approaches include per-pixel cell segmentation with subsequent pixel grouping, or localization of bounding boxes with subsequent shape refinement. In situations of crowded cells, these can be prone to segmentation errors, such as falsely merging bordering cells or suppressing valid cell instances due to the poor approximation with bounding boxes. To overcome these issues, we propose to localize cell nuclei via star-convex polygons, which are a much better shape representation as compared to bounding boxes and thus do not need shape refinement. To that end, we train a convolutional neural network that predicts for every pixel a polygon for the cell instance at that position. We demonstrate the merits of our approach on two synthetic datasets and one challenging dataset of diverse fluorescence microscopy images.

Conference paper at MICCAI 2018

References in corpus (4)

• SSD: Single Shot MultiBox Detector
• Arbitrary-Oriented Scene Text Detection via Rotation Proposals
• Multiclass Weighted Loss for Instance Segmentation of Cluttered Cells
• Moral Lineage Tracing

Cited by in corpus (25)

• Star-convex Polyhedra for 3D Object Detection and Segmentation in Microscopy
• Understanding metric-related pitfalls in image analysis validation
• The Multi-modality Cell Segmentation Challenge: Towards Universal Solutions
• BiomedParse: a biomedical foundation model for image parsing of everything everywhere all at once
• CPP-Net: Context-aware Polygon Proposal Network for Nucleus Segmentation
• Community-developed checklists for publishing images and image analysis
• Kartezio: Evolutionary Design of Explainable Pipelines for Biomedical Image Analysis
• Nuclei & Glands Instance Segmentation in Histology Images: A Narrative Review
• CellProfiler plugins -- an easy image analysis platform integration for containers and Python tools
• Answering some questions about structured illumination microscopy
• Boosting electrode performance and bubble management via Direct Laser Interference Patterning
• Object-Guided Instance Segmentation With Auxiliary Feature Refinement for Biological Images
• Automated Whole Slide Imaging for Label-Free Histology using Photon Absorption Remote Sensing Microscopy
• Improving Pixel Embedding Learning through Intermediate Distance Regression Supervision for Instance Segmentation
• An Instance Segmentation Dataset of Yeast Cells in Microstructures
• ConsInstancy: Learning Instance Representations for Semi-Supervised Panoptic Segmentation of Concrete Aggregate Particles
• PhagoStat a scalable and interpretable end to end framework for efficient quantification of cell phagocytosis in neurodegenerative disease studies
• Cluster-based human-in-the-loop strategy for improving machine learning-based circulating tumor cell detection in liquid biopsy
• RepSNet: A Nucleus Instance Segmentation model based on Boundary Regression and Structural Re-parameterization
• NCIS: Deep Color Gradient Maps Regression and Three-Class Pixel Classification for Enhanced Neuronal Cell Instance Segmentation in Nissl-Stained Histological Images
• Comparative Evaluation of Traditional and Deep Learning-Based Segmentation Methods for Spoil Pile Delineation Using UAV Images
• Generalist Segmentation Algorithm for Photoreceptors Analysis in Adaptive Optics Imaging
• Learning with minimal effort: leveraging in silico labeling for cell and nucleus segmentation
• A Comparison of Deep Learning Methods for Cell Detection in Digital Cytology
• The TYC Dataset for Understanding Instance-Level Semantics and Motions of Cells in Microstructures