Around 2014-2015, I screened 20,000 biomedical articles for image duplication in 2014-2015. About 4% of those papers, 782 to be exact, contained inappropriate duplications. Together with my coauthors Arturo Casadevall and Ferric Fang, I published that work in mBio in 2016 here.
Unfortunately, many of these problematic papers are still not addressed by the journals now, ten years later, even though the concerns in some of them are far beyond what could be blamed on an innocent error. Two of those examples were published in Cytokine, a journal that seems to care little about the problematic papers they published in the past.
Science Integrity Digest 