This is insane. As a researcher I had to take research ethics course, and obviously we study some cases of research misconduct and fraud. But this kind of stuff is next-level.
I know Sweden launched a national board to investigate research misconduct [1], I hope more countries do the same.
Though many of them look like outright fraud (anytime photoshopping is involved), I think some could be honest mistakes. If you have dozens of images that are small variants of the same study, it's very easy to misname one of the files or pick the wrong one when loading it into the image editor.
Exactly, honests mistakes can be made (happens to me), and it happened that I spotted some in a paper and contacted the authors and they publish an erratum (or not).
But this kind of photoshopping, done over a period, really gets me.
I know Sweden launched a national board to investigate research misconduct [1], I hope more countries do the same.
[1] https://npof.se/en/