Hi, my apologies if this is a basic question, I’m relatively new to QuPath. I’m working with 30-40 channel fluorescent whole-slide images for an upcoming methods paper, and using QuPath for annotations, cell detection, and marker measurements.
I’m trying to classify cells based on multiple marker intensities. The problem is that basic threshold classification or pixel training doesn’t seem ideal when dealing with this many markers. What I would like to do is perform cell detection and marker measurements with QuPath, then export those measurements to perform Phenograph cluster analysis (in R or python, similar to cytof analysis), annotate the clusters with cellular identities based on mean marker intensities, and then load those single cell classifications back into QuPath to continue analysis.
Does anyone have any experience with importing cell classifications, or using any kind of dimensional reduction analysis in their QuPath workflow?