Imagetwin is software from Imagetwin AI GmbH that detects integrity issues in figures in scientific articles, including western blots, microscopy images and light photography. Users upload a PDF or image files and see results in a web interface within seconds, with confidence scores. It detects duplication within manuscripts, manipulation such as splicing and copy-move edits, and plagiarism by comparing figures with published work, and it flags AI-generated images with model attribution. Imagetwin gives the size of its database of published figures as over 160 million, and it offers API access for peer review and publishing workflows and a private repository for checking new submissions against an organisation’s own images. Taylor & Francis has integrated it through an API on selected journals.
Benefits & outcomes
- Imagetwin reports that the American Society for Microbiology uses it across 17 journals as part of ethics screening since 2023, and that resolving an issue before publication takes about 1.5 hours on average compared with up to 10 hours after publication.
- Imagetwin reports that more than 1 million manuscripts have been scanned.
- Imagetwin reports that 1,546 academic organisations use it.
Notes & quotes
- “Imagetwin has been a fantastic tool to help me find duplicated photo panels in scientific papers.” Elisabeth Bik, research integrity expert.
- “Having the flexibility to check any manuscript at any stage has proved very useful.” Kim Eggleton, Head of Research Integrity and Peer Review, IOP Publishing, which has used Imagetwin since 2022.
Publisher / company: American Society for MicrobiologyImagetwinIOP PublishingTaylor & Francis
Date of case study: Reported in 2026
Status: Live
Built by: Company
Sources: