Automatic nuclei segmentations in near-isotropic, reconstructed volume electron microscopy (FIB-SEM) of mouse Meissner corpuscle (double-innervated) (jrc_mus-meissner-corpuscle-2) model

Model description: Cellpose 3.0.9 was trained on nuclei from 27 2D slices from jrc_mus-meissner-corpuscle-2.

Architecture: Cellpose 3.0.9

Number of parameters: 6600845

Input voxel size (nm): 128 x 128 x 128 (X, Y, Z)

Output voxel size (nm): 128 x 128 x 128 (X, Y, Z)

Classes trained on: Nucleus

Training steps: 27000

Epochs: 1000

GPU: NVIDIA TITAN RTX

RAM: 251.0 GiB

Model wall time (sec): 567.55

Software: Cellpose 3.0.9

Software DOI: https://doi.org/10.1038/s41592-022-01663-4

Source dataset (EM) ID: jrc_mus-meissner-corpuscle-2

Source dataset (EM) DOI: https://doi.org/10.25378/janelia.23969106

Generated dataset DOI: https://doi.org/10.25378/janelia.26506543

Model URL: https://data.janelia.org/UAlMG7

Github repo: https://github.com/janelia-cellmap/cellmap-models

Visualization website: https://openorganelle.janelia.org/datasets/jrc_mus-meissner-corpuscle-2