Academia’s image manipulation epidemic
Last week, we explored the troubling rise of data fabrication in academia, including a high-profile case involving an Ivy League professor and misuse of ChatGPT.
Now, another concerning issue looms: image manipulation. Driven by the pressure to publish, authors are resorting to altering images to fit desired outcomes, with manipulation being found in about 4% of biomedical papers. This obstructs scientific progress, leading to retractions.
Retractions due to image manipulation occur almost every other day, as documented by the Retraction Watch database. These cases persist and show no signs of diminishing. In July 2022, a whistleblower found serious image manipulation concerns in Alzheimer's expert Sylvain Lesné's Western blot photos. In Science Mag, it was noted, "Some Alzheimer's experts now suspect Lesné's studies have misdirected Alzheimer's research for 16 years." Just last year, Stanford's president, Marc Tessier-Lavigne, stepped down following concerns of image manipulation in his research, which led to the retraction of the implicated work.
Given the damage that image manipulation can cause in the scholarly community, let’s explore recent cases and discover what preventative steps Morressier is taking to tackle this issue, alongside imagetwin.
Image deception at the Dana-Farber Cancer Institute
A few weeks ago, allegations of data fabrication hit the headlines involving researchers at the Dana-Farber Cancer Institute, an institution affiliated with Harvard Medical School. The misconduct spans papers published between 1999 and 2017, mostly involving duplications of blots, bands, and plots within images, as flagged by integrity sleuth Sholto David.
In a blog post, David highlighted suspicious images from over 30 papers authored by Dana-Farber scientists. These images contained duplications, meant to boost the credibility of the scientists' findings.
The news of this misconduct closely follows a separate plagiarism investigation involving former Harvard president Claudine Gay, who resigned earlier in the year. These events have led to multiple integrity scandals for Harvard-affiliated institutions within a very brief timeframe, a shock for many considering Harvard's esteemed reputation.
What tech is being used to identify image misconduct?
During the investigation into fraud allegations at Dana-Farber, Sholto David was assisted by imagetwin, an AI-based platform that automatically spots integrity issues within figures of research articles.
More recently, Harvard Medical School neuroscientist Khalid Shah faced accusations of image plagiarism across several papers. Image manipulation expert Elizabeth Bik brought this misconduct to light, uncovering 44 instances of falsification spanning from 2001 to 2023. Bik, just like David, turned to imagetwin's powerful software and employed it alongside her own investigative skills.
Be part of the solution
Our largest focus is equipping our clients with effective tools to uphold integrity. Given imagetwin’s pivotal role in recent controversies and its powerful skill in flagging image manipulation-related integrity breaches, we've joined forces with imagetwin to provide Morressier users with access to its cutting-edge software. Under the partnership, Morressier’s workflows now feature automated analysis of scientific images, providing editors and peer reviewers with tools to detect potential issues such as plagiarism and manipulation, early and at scale.
Morressier's partnership philosophy is grounded in the belief that providing the industry's latest and most advanced technology on a single platform creates a seamless experience for our customers. Our focus on the power of integrations and end-to-end solutions allows publishers to focus on what matters the most: research. Explore more details of our collaboration here.
