Adipocytes Pipeline Help



I am a new user to Cellprofiler as previous methods of using ImageJ were unable to generate quality Region of interests for current analysis of adipocytes. I have tried different pipelines that were available within the forum and I happened to chance upon the latest version of Adipocyte Pipeline numbered Rodeheffer Lab. However, upon running an analysis on the image (provided below), there were no outlines generated by the pipeline, despite giving good images of the membrane detection May I know what is wrong with the pipeline or what settings that I haven’t activate such that the analysis failed. Thank you very much for the help provided!

Adipocyte Pipeline numbered Rodeheffer Lab .cppipe (19.5 KB)


There were some issues with the uploading of the image. I have just reuploaded the image again. Thank you very much for looking into the pipeline issues.


Hi there,

The pipeline from Rodeheffer Lab is not recent, it’s actually aimed for specific input.

And if you’re new to CellProfiler, I would recommend to try build some simple modules on your own to understand each module better.

  1. If in NamesAndTypes, you load the image as color image, you first need to convert it to gray, for example adding a module called ColorToGray. If this is a single channel image, in NamesAndTypes, you can load the image as gray image
  2. You then need to invert the image by ImageMath, because you now have the droplet-membrane brighter than the droplets. You’d want to have the droplets brighter than the droplet-membrane.
  3. Then add IdentifyPrimaryObjects, and experiment different algorithms to see which works best for you. I think Global > Otsu and Declumping by using intensity,

Good luck and have fun.


Dear Minh,

Thank you very much for your prompt response to my queries. I have actually played with the software by breaking down each step of the pipeline to understand the commands and I understand that the pipeline provided requires a very specific input of fluorescent images in Texas Red filter, which I have obtained. I have found out that the image is required to be at 10X Magnification and that it has to be clearly defined. Also, I have encountered issues with Java Heap Memory when working with stitched image, which I believe is due to memory overload from the image size itself. I will continue playing around the software to obtain my results. Thank you very much for all the tips you have provided.