Imagetwin

What is Imagetwin?

Imagetwin is AI-powered software that improves the quality and trustworthiness of scientific research by spotting integrity issues in figures from life science articles. It flags inappropriate manipulations and duplicated images across figure types such as western blots, microscopy images, and general photography. The tool scans documents or image files with its algorithms and presents results on a web interface, helping researchers, peer reviewers, publishers, and academic institutions maintain the credibility of their scholarly articles.

Who created Imagetwin?

Imagetwin was founded by Markus Zlabinger (CTO) and Patrick Starke (CEO). Markus holds a Ph.D. in computer science and leads technology and product development, with expertise in image processing and machine learning. Patrick, a graduate of Vienna University of Economics and Business, handles cooperations and business development. The company, ImageTwin AI GmbH, is based in Vienna, Austria, and has received funding from Seedfinancing of the BMDW, managed by aws.

What is Imagetwin used for?

• Automated plagiarism detection: Checks figures against a database of over 50 million images to find content reused across articles.
• Manipulation detection: Identifies duplicates and potential data fabrication within articles.
• Figure reuse detection: Specifically looks for instances where figures are reused across various articles.
• Efficiency and accuracy: Detects integrity issues quickly and with a high degree of precision.
• Privacy and security: Uploaded content is not stored, and all server communication is encrypted.
• Research quality: Helps improve the overall quality, rigor, and credibility of scholarly publications.

Who is Imagetwin for?

Imagetwin serves researchers, peer reviewers, and publishers, as well as data scientists, annotation experts, business development managers, and software developers.

How to use Imagetwin?

1. Upload your content: Upload a PDF file or several image files in common formats such as JPG, PNG, and GIF.
2. Start the scan: A single button press initiates the AI-based scan of your selected content.
3. Review the results: Findings are presented clearly on the web interface within seconds.