Your best bet is to use the nuclear/perinuclear ring region intensity, rather than cytoplasmic region. This corrects for cell-to-cell variations in area and thickness which can throw off the cytoplasmic mean intensity measurement. Whether this makes a difference will depend on the shape and size of your cells.
Also make sure to use the log(ratio)! This is imperative - otherwise your averages and SDs will always skew toward the nuclear intensity side. i.e. 2:1, aka 2, is the equivalent of 1:2, aka 0.5 … the average of those ratios should be 1, not 1.25. Using the log(ratio), a value of 0 means equal intensity in nucleus and cytoplasm/perinuclear region, negative numbers mean more in cytoplasm, and positive numbers mean more in the nucleus.
I discuss this in https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5289939 and https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4380925/ … Cite early, cite often!
Attached is an example pipeline and images to illustrate the protocol.
003015-8-CH1.tif (632.1 KB) 003015-8-CH2.tif (632.1 KB)
Nuclear translocation with ring region.cpproj (1.0 MB)