Description
Recently a couple of ideas pilled up of what everything else could be measured. Here in particular:
Esti wrote: Cha is not normalised so it's difficult to see if there was an increase or decrease (common to photobleaching). When looking at the data, the fluorescence datasets have a very high Cha value, which probably has to do with that precisely. Moreover, in the graphs it can be seen a decreasing tendency for those datasets. Maybe, normalizing it, the correlation will appear.
Vlado wrote: Divide by number of frames (it view), or by physical time-span (bio view, makes intensity change comparable among experiments)
....
Esti: I think it's critical to work with ratios or normalise the images beforehand (the DL way, Vlado presumes). Otherwise, you will get magnitudes according to the bit depth of the image.
Also, Cha is measuring the difference between the last and first frames (in fact, two last and two first frames, for better stability). It's blind to whatever happens with the intensity in between during the timelapse. This could be also taken into account.