BioFormats: NDPIS Brightfield+Fluorescence

Hi! I have recently been working with Nanozoomer files that contain both a brightfield image an fluorescent channels, which is great because there is no need to align the extra channels, they are all aligned and can be opened in some software using the NDPIS file. I can provide a sample file, though it is somewhat large.

With BioFormats/FIJI/Bioformats extension in QuPath, the RGB channel is not able to be loaded along with the other fluorescent channels. It seems to be read as a single 8bit channel image. Is there any possibility that these could all be loaded at once?

Snip from an NDPIS: DAPI+a mostly unstained tissue section in brightfield:


I am not sure, but I am guessing they are subtracting the DAPI channel from blue and presenting the information as RGB.

Advantages: Being able to scan a fully aligned tissue slice with brightfield stains that may obscure nuclear segmentation (with hematoxylin), plus the possibility to use other channels for further markers. Autofluorescence might impact other channels (and chromogens like FastRed that are fluorescent), but it might be interesting to see if KI67 would show up against the “green” AF background since you would only be looking for “real” signal when overlapped with a nucleus.

Hi @Research_Associate, would you be able to upload the sample file to https://www.openmicroscopy.org/qa2/qa/upload/ so I can take a closer look at what is going on. Is it an NDPIS file along with 2 NDPI files?

Hmm, I’m not sure my Message went through with the link to download them, but I was able to get the upload to work as well, now, so that is going slowly. The Zip file name starts with “AnaPath”

Editing to add in that the name is completely unrelated to a company of the same name I just became aware of.

We received the sample file and I was able to test and reproduce what you are seeing. When looking at the NDPIS file with NDP View it does appear that a new combined channel is generated from the underlying files. I will experiment to see if we can easily reproduce the same behaviour with Bio-Formats.

1 Like