powerful-fish alpha

3D UNet Mouse Embryo Live

A. Wolny, V. Bondarenko

10.5281/zenodo.6384845/7774490

Example image containing both original image data and after applying the model to it.

Model Information

Description: A 3D U-Net trained to predict the cell boundaries in live light sheet images of developing mouse embryo. Voxel size: 0.2×0.2×1.0 µm^3
Task: 3D cell segmentation
Licence: MIT
Tags: 3D, cell membrane, mouse embryo, live tissue, light sheet microscopy, semantic segmentation, torchscript, ilastik, U-Net, plantseg, deepimagej, pytorch, animal

Benchmarking

This model was not tested on datasets with reference annotations, so we cannot produce benchmarking metrics.


Model predictions

Study Dataset Original image Prediction Analysis parameters
S-BIAD1490 Source data
Download

Download
z_slices_analysed: 6,22
xy_image_crop_size: 512,512
t_slices_analysed: 0,1
S-BIAD915 Microscopy data
Download

Download
z_slices_analysed: 53,69
xy_image_crop_size: 1024,1024
input_channel_analysed: 1
S-BIAD827 Experiment A
Download

Download
z_slices_analysed: 24,40
input_channel_analysed: 0
S-BIAD1241 Colocalization N. benthamiana
Download

Download
z_slices_analysed: 9,25
input_channel_analysed: 0