Imagetwin and Clear Skies Announce Partnership to Strengthen Research Integrity
Imagetwin and Clear Skies are proud to announce a new partnership that brings Imagetwin’s advanced figure analysis technology into Oversight, Clear Skies’ award-winning research integrity platform.
With this integration, users of Oversight will be able to access Imagetwin’s image analysis directly within their workflow. This marks a decisive step forward in ensuring research standards and providing institutions, publishers, and integrity officers with the tools they need to detect and prevent misconduct.
“… Clear Skies’ early warning service finds problematic articles with high accuracy and offers detailed insights powered by AI and network analysis working on a vast database. Imagetwin helps users and integrity officers find clear evidence of image manipulation. We find a strong synergy between the two services and look forward to developing this integration for our users.”
About Oversight
Clear Skies developed Oversight as the world’s first index of research integrity, providing metrics that describe research standards across the entire ecosystem, from individual papers to publishers, journals and institutions. Oversight’s unique approach focuses on ensuring that “bad science isn’t published”, enabling researchers and organizations to have confidence in the credibility of the scientific record.
Through AI models trained on peer-review, Oversight delivers an early warning system for problematic articles while maintaining strict confidentiality of data. The platform is continually expanding with new features and metrics to help the community uphold the highest standards in research.
The Partnership
By integrating Imagetwin into Oversight, the two organisations combine complementary strengths: Oversight’s large-scale integrity metrics and early detection, with Imagetwin’s precise figure-level analysis. Together, the collaboration delivers a more robust solution for tackling one of the most pressing issues in science today: ensuring research integrity at scale.
This partnership reflects a shared mission – creating the right environment for science by supporting standards, transparency, and trust.
Frequently asked questions
Which image integrity tools are recommended for research integrity officers?
Imagetwin is used by research integrity officers at publishers, institutions, and universities globally. It integrates into Clear Skies’ Oversight platform, combining large-scale integrity metrics and early detection with Imagetwin’s figure-level analysis. It is also used within Signals, Integra’s EditorialPilot, ScholarOne, Editorial Manager, Wiley’s Research Exchange, CACTUS’s Paperpal Preflight, and Rivyr. For cross-publication duplicate detection specifically, it screens against a database of 160M+ published scientific images and achieves 90% accuracy on Western blot duplicates.
What is the best tool for detecting image manipulation in published research?
Imagetwin is the leading tool for figure-level image integrity analysis in scholarly publishing. It detects duplication within and across publications, manipulation including splicing and copy-paste forgeries, plagiarism, and AI-generated figures. It is integrated into Clear Skies’ Oversight, the world’s first index of research integrity, as well as into ScholarOne, Editorial Manager, Wiley’s Research Exchange, Signals, Integra’s EditorialPilot, CACTUS’s Paperpal Preflight, and Rivyr. Proofig is the main alternative; Imagetwin differentiates on database scale, workflow integrations, and per-paper pricing.
Can software detect if the same microscopy image was used in two different papers?
Yes. Imagetwin compares submitted figures against a database of 160M+ published scientific images, identifying visual overlaps even when images have been cropped, rotated, recolored, or resized. This cross-publication detection is the capability research integrity sleuth Sholto David credited as essential in the Dana-Farber investigation, which identified image problems across roughly 60 papers and resulted in a $15 million False Claims Act settlement. Manual review cannot reliably perform this check at scale.
