Related Experiment Video
Updated: Aug 5, 2025
Quantitative Analysis of Protein Expression to Study Lineage Specification in Mouse Preimplantation Embryos
Published on: February 22, 2016
A novel ground truth dataset enables robust 3D nuclear instance segmentation in early mouse embryos
Hayden Nunley1, Binglun Shao1,2, Prateek Grover1
1Center for Computational Biology, Flatiron Institute - Simons Foundation, New York, United States of America.
Researchers developed a new mouse model and dataset (BlastoSPIM) for accurate 3D nucleus segmentation in early embryos. This enables advanced studies of cell fate and rearrangement using the Stardist-3D deep learning model.
Area of Science:
- Developmental Biology
- Bioimaging
- Machine Learning
Background:
- Accurate 3D nucleus segmentation is crucial for studying early embryonic development.
- Existing methods struggle with low signal-to-noise, high anisotropy, and dense nuclei in microscopy images.
- Lack of annotated 3D data limits supervised machine learning for this task.
Conclusions:
- The novel reporter line and BlastoSPIM dataset significantly advance 3D nucleus segmentation in developmental biology.
- Stardist-3D, trained on BlastoSPIM, enables detailed studies of cell fate patterning.
- BlastoSPIM serves as a valuable resource for improving deep learning models in bioimaging.
More Related Videos
08:33Author Spotlight: Strategies for Mounting Zebrafish Embryos for High-Resolution Multiview Light-Sheet Microscopy — Techniques for Imaging and Image Reconstruction
Published on: July 19, 2024
12:59Three and Four-Dimensional Visualization and Analysis Approaches to Study Vertebrate Axial Elongation and Segmentation
Published on: February 28, 2021