Star-convex Polyhedra for 3D Object Detection and Segmentation in Microscopy

Abstract

Accurate detection and segmentation of cell nuclei in volumetric (3D) fluorescence microscopy datasets is an important step in many biomedical research projects. Although many automated methods for these tasks exist, they often struggle for images with low signal-to-noise ratios and/or dense packing of nuclei. It was recently shown for 2D microscopy images that these issues can be alleviated by training a neural network to directly predict a suitable shape representation (star-convex polygon) for cell nuclei. In this paper, we adopt and extend this approach to 3D volumes by using star-convex polyhedra to represent cell nuclei and similar shapes. To that end, we overcome the challenges of 1) finding parameter-efficient star-convex polyhedra representations that can faithfully describe cell nuclei shapes, 2) adapting to anisotropic voxel sizes often found in fluorescence microscopy datasets, and 3) efficiently computing intersections between pairs of star-convex polyhedra (required for non-maximum suppression). Although our approach is quite general, since star-convex polyhedra include common shapes like bounding boxes and spheres as special cases, our focus is on accurate detection and segmentation of cell nuclei. Finally, we demonstrate on two challenging datasets that our approach (StarDist-3D) leads to superior results when compared to classical and deep learning based methods.

Conference paper at WACV 2020

References in corpus (3)

• scikit-image: Image processing in Python
• Cell Detection with Star-convex Polygons
• Measurement of areas on a sphere using Fibonacci and latitude-longitude lattices

Cited by in corpus (11)

• Nuclei instance segmentation and classification in histopathology images with StarDist
• Community-developed checklists for publishing images and image analysis
• Soft Metamaterials: Adaptation and Intelligence
• Chloroplasts in plant cells show active glassy behavior under low light conditions
• Boosting electrode performance and bubble management via Direct Laser Interference Patterning
• 3D Segmentation of Neuronal Nuclei and Cell-Type Identification using Multi-channel Information
• ConsInstancy: Learning Instance Representations for Semi-Supervised Panoptic Segmentation of Concrete Aggregate Particles
• Intelligent Software System for Low-Cost, Brightfield Segmentation: Algorithmic Implementation for Cytometric Auto-Analysis
• NucMM Dataset: 3D Neuronal Nuclei Instance Segmentation at Sub-Cubic Millimeter Scale
• Generalist Segmentation Algorithm for Photoreceptors Analysis in Adaptive Optics Imaging
• Cell segmentation of in situ transcriptomics data using signed graph partitioning