NamesAndTypes loading & indexing

Hello,

I am loading three .tif files per one image to analyze (image + two corresponding mask images) following the same indexing and a clear naming pattern. While loading them on NamesAndTypes, I noticed that indexing is a bit mixed up between 1 and 10, 11, 12 (anything that starts with 1). Why is this the case? I see that I am able to avoid this issue by naming images _010 instead of _10, but I’m wondering if there’s a simple way to correctly load matching images based on index without changing image name itself.

Also, while trying to figure out a way to correctly load corresponding images, I realized that the NamesAndTypes module does not load images (gives me the “no image sets available” message) if I have the Groups module checked to “yes”. Why does the Groups module affect the image loading on NamesAndTypes module?

I’ve attached the pipeline.

Thank you!
EM Axons.cpproj (1.4 MB)

Hi @suminkim,

There are two ways I can think of,

  1. Use regular expressions in Metadata extraction method (capture the image name) & further use that information as one way to catogarise the in “Names & types”
  2. Use the external .csv or excel file with just the image names & import the file in the Metadata data & use that information in “Name& Types” similar how you currently use (may be image name from the external file. IMAGENAME_innermask & IMAGENAME_outmask to match them.

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

Hello,
Thank you for the suggestions!

Any explanation for why the Groups module affects NamesAndTypes module?

Best,
Sumin

Groups affects NamesAndTypes because, as you’ve found, those modules “cheat” a bit- rather than being 4 entirely independent steps, they are all working closely together to set up your image sets and groups in such a way that CellProfiler understands what to do downstream. It’s not actually NamesAndTypes that determines if you have valid image sets, it’s the sum of all 4, but as the most downstream always-required input module, that’s simply where it gets displayed.
It’s (part of) why they’re physically separated by a line- they aren’t quite the same as what comes downstream.

It’s worth noting that, once you have Groups set up correctly, it won’t block your ImageSets being created, it just needs configuration.

Ah I see. Thanks! Indeed turning on metadata and setting groups to use metadata information works fine for loading images. :slight_smile:

2 Likes