Alternative ways of handling a huge number of images as input for CellProfiler

Image Analysis
#1

Hi all,

We are handing a large collection of images to be analysed in CellProfiler. We are currently passing a folder as a parameter but that becomes inefficient when the number of images in the folder is very high, so we would like to pass all the images as one single file. We are exploring several ways:

  • Have a zip/tar file as the input, is that currently possible in CellProfiler?
  • Use a multi-page tiff to pack groups of unrelated images (e.g. a particular channel of all the images in a plate into one multi-page tiff). The metadata will be then also unrelated for each one of the tiff files. Would that affect the results of the analysis?
  • Is CellH5 an option in this case?

Any advice would be very much appreciated. Thanks a lot!

3 Likes
#2

Hi @BeatrizSerrano,

Regarding your first option,
Though I am not sure about if CP process zip/tar files, but I have used the shell script (linux based) where I unzip the files just before the Cellprofiler pipeline headless mode process & zip backs once the pipeline is completed. All these commands where in the same script.
Regarding you second option,
Multi-page tiff is a good option. It should not affect your results provided that, not using the metadata information rather give it manually using the regular expression or in an seperate excel sheet.
Third option not sure.
Regards,
Lakshmi
Fujifilm Wako Automation (Consultant)
www.wakoautomation.com
For CellProfiler training or optimised pipeline write to,
lakshmi.balasubramanian.contractor@fujifilm.com

Read more on our site.
Yokogawa CV8000 - The Ultimate in Confocal HCS
https://www.wakoautomation.com/products/yokogawa-high-content-imaging

CellProfiler headless with multipage tif
#3

Thank you @lakshmi! The multi-page tiff seems to work with some adjustments in the Metadata and NamesAndTypes modules.