search

lopaavol / mihc-suppl-software

Code repository for paper Systems pathology by multiplexed immunohistochemistry and whole-slide digital image analysis
MIT License
7 stars 2 forks source link

Starting images for the program to work #2

Open pauliferni opened 3 years ago

pauliferni commented 3 years ago

Hello to test your pipelines in the cellprofiler, I see that you select the starting images depending on the cell that will appear in each image. Could you provide the folder where the starting images are to test the pipeline, and see how it works?

At the end of the pipeline, put the folder where it is exported in case it helps you (Projects / 080915_export / prostate_cocktail / results / img)

Thanks!

lopaavol commented 3 years ago

Hi!

Note that the CellProfiler version used in the paper was 2.1.2. The pipeline needs modifications for it to run in newer CellProfiler 3.x or 4.x.

The pipeline takes in any number of images but it expects a certain amount of different channels ie. each location has many different imaged channels. Image names are expected to include the channel names. The same pipeline can be used with different channels but it will require changes into the channel definitions (and maybe to later processing steps if the name of the channel is changed).

The output folder could be also just the default output folder.