imagetwin

Imagetwin is an AI-based software that automatically detects image integrity issues in life science articles and enhances the peer-review procedure.

[imagetwin @ImageTwinAI]

New in Imagetwin (BETA): duplicate detection for line graphs, spectra, chromatography traces, and XRD plots, within and between figure panels.

[imagetwin @ImageTwinAI]

Imagetwin is now integrated into @wileyinresearch's Research Exchange, used by 1,500+ journals.

Pilot result: 3x more image integrity issues detected vs. human reviewers alone.

Read more: https://t.co/YLxk998sMH

[imagetwin @ImageTwinAI]

🆕 Our Private Repository just got an upgrade.

Build your own image database, scan manuscripts against it, and catch duplicate figures automatically.

👉 https://t.co/Huci6rz4DB

[Spotting @Spottingthespot]

Another one of those amazing @ElsevierConnect #papermill publications where 6 authors were added to the paper! Are the 2 'remaining' authors the ones that got paid by these 6 for becoming an author? #papermill #fraud

[Sholto David @addictedtoigno1]

Surprised to discover that Thermo Fisher appears to show a fake western blot for the validation of one of their p53 antibodies. I've added a diagram to show the very similar bands. This does not appear to be one of the "published figures", but their own internal data.

[imagetwin @ImageTwinAI]

@WCRIFoundation 2026 is off to a great start! 🎉

Alisia and Patrick are in Vancouver connecting with the research integrity community. Find them at Booth 10 or catch them at the talks!

Key takeaways coming soon. 👀

[imagetwin @ImageTwinAI]

@danwalker9999 @rbdamgaard @MicrobiomDigest Thanks for the mention. The "undetectable" framing is not exactly true anymore. Generative outputs leave traces, statistical, structural... Detection won't be one tool forever solving it, it's continuous work, and we're on it🙂

[imagetwin @ImageTwinAI]

🤝 @ImageTwinAI is now integrated into @Cactusglobal.

Automated detection of image duplication, manipulation, plagiarism, and AI-generated figures, embedded directly into research and publishing workflows.

🔗 https://t.co/zt4DVayswe

[Spotting @Spottingthespot]

The #papermill #fraud Balal Yousaf from @polsl_pl, who publishes nothing but fake stuff, got another retraction👇A relative fast retraction, took only 5 months! https://t.co/2dNqrwXrFc

[Spotting @Spottingthespot]

Another retraction for professional #papermill #fraud Talha Bin Emran! Notice this time also Raffaele Capasso! This wannabe has put his name on many paper mill publications! But I doubt his university @UninaIT will care! These frauds continue to publish junk!

[imagetwin @ImageTwinAI]

@KamounLab Good one!😂

[Sophien Kamoun @KamounLab]

Waiting for reviewer comments

[imagetwin @ImageTwinAI]

We're hosting A Researcher's Sleuthing Journey and How It Led to a $15M Case webinar with @addictedtoigno1 and @ImageTwinAI co-founder. Join us! https://t.co/QqOdNxFJjG

[imagetwin @ImageTwinAI]

@addictedtoigno1 @ImageTwinAI

[imagetwin @ImageTwinAI]

One researcher. The right tools. $15M fraud settlement.
What peer review missed, caught by image integrity sleuthing.
April 16 | 17:00 CET | LinkedIn & YouTube Live
🔗 https://t.co/SogAJrpEZs

[Retraction Watch @RetractionWatch]

BMJ retracts cardiac stem cell paper, removes authors months after sleuths flag data ‘mismatch.’
https://t.co/51jebP28aH

[imagetwin @ImageTwinAI]

@Thatsregrettab1 @FrontiersIn wow

[imagetwin @ImageTwinAI]

@ImageTwinAI Western Blot Manipulation Detection just got a major upgrade.

📉 False positive rate: 2.4% → 1.7%
📈 Detection rate: +14 percentage points on splices
✅ Additional gains on copy-paste and horizontal splice detection

[imagetwin @ImageTwinAI]

From our webinar:
▫️Policies ≠ operational
▫️Checks too late
▫️Workflows inconsistent

Shift: reactive → proactive screening at submission

🗒️Webinar recap: https://t.co/4WWhrcit4z
🎥Recording: https://t.co/JYXGBpM3C4

[Cheshire @Thatsregrettab1]

I continue slowly working through one volume of AJCR (w/an Associate Editor-in-Chief in Wafik S. El-Deiry, MD, PhD). So much crap.

This one paper was so bad that I gave up before finishing (I’m only flagging images identified by @ImageTwinAI for reasons).