Identifying cells in a DIC image using analyze particles

Hey all,
Anyone know a good pipeline for processing DIC images so that the individual cells can be found using analyze particles? For example, in this image, https://imgur.com/a/FcED5eX, I would like to get each individual cell to be an ROI.
Thanks

DIC can get a little tricky because of the contrast mechanism. I generally have luck messing around with edge filters in these cases (Process -> find Edges). I’d recommend trying that with some thresholding (ctrl+shift+T) and some of the Binary operations (Process -> binary). Analyze Particles should work from there.

Let me know if you have any luck.

1 Like

i have actually done that successfully, the only problem is sometimes fails spectacularly and most of the time misses quite a few cells.
My current pipeline is:

  1. find edges
  2. convert to binary
  3. close or dilate to close some of the holes in the binary
  4. fill holes
  5. watershed algorithm to split up blobs
  6. analyze particles

can you post an example result?

also may be worth considering some attempts from people previously. I found this post on the forum that might be helpful:

Here’s an example of what the current pipeline does:

did you try dilating and closing multiple times before filling holes? That might help a little bit here. Not sure how repeatable that might be.

It doesn’t seem horribly far off. you might be able to get it to work.

I mean it works, its just bad. I would like to improve it somehow

Sorry for the delay, but I had an idea i figured was worth suggesting. Weka segmentation plugin might work well for this. have you given that a shot?