Skip to content

Latest commit

 

History

History
37 lines (22 loc) · 1.56 KB

README.md

File metadata and controls

37 lines (22 loc) · 1.56 KB

CellProfiler cellpose detection README

Data

These pipelines were developed for 2/3D FLIM images of adherent, immune and organoid cancer cells. Images are found in "example_images". The pipeline file are called "runcellposeAdherentCells.cppipe, runcellposeOrganoid.cppipe and runcellposeImmuneCells.cppipe" and is built for use with runCellpose using Cellpose 2.0. The pipelines will segment the nuclei, whole cell, link the objects and extract the cytoplasm for each cell. Note: The immune cells example images are not avaiable at this moment.

Instructions

These pipelines require installing cellprofiler from source and downloading and configuring the cellprofiler-plugins repo. Check out the instructions available on Github.

Note that a GPU is required for the example image to be processed in a timely fashion. If you don't have a GPU, Cellpose will use the CPU.

Cellpose models

The models are inside the NewpipelinesCellPose folder:

Adherent Cells: nuclei and whole cell models Immune Cells: Activated layer, nuclei models - for whole cell detection you can use the Adherent Cells model Organoids: nuclei and whole cell models

Example results

Cellpose segmentation of nuclei (RunCellpose):

Cellpose segmentation of whole cell (RunCellpose):

Find the cytoplasm (IdentifyTertiaryObjects):

Final segmentation: