How to Use Imagetwin

New to Imagetwin? Follow this step-by-step introduction to learn how to run scans, review results, and make the most of the platform.

Contact support

---

To start, log into the Imagetwin app at https://app.imagetwin.ai/

1

Step 1: Scan page

Drag and drop your file(s) into the dedicated area or click the area and choose a file through a file selection dialog.

Supported formats include .pdf, .jpg, .jpeg, .png, .gif, .jfif, .bmp, .tif, .tiff, .svg, .webp, .doc, .docx.

2

Step 2: Start the scanning process

After reviewing the selected file, click “Scan” at the bottom of the page. Scans are automatically saved for 90 days and can be accessed in the History tab (on the left side) later on. If you prefer not to save a scan, simply uncheck the “Save scan for 90 days” option before proceeding. Saved scans are only accessible and visible to you (and/or your institution).

3

Step 3: Congratulations, your scan has been successfully completed.

You will be automatically redirected to the Results page. On the top of the page, you will see a summary bar of all findings, and how many of each Imagetwin has found.

4

Step 4: Review your results

Imagetwin will show you where it has detected duplicates within the paper you have uploaded, across other publications, manipulations, as well as AI-generated images. By clicking on each finding, the detail view opens up, which allows you to investigate the finding further and offers a further forensic analysis.

Analyze the different finding types

Finding type 1: Duplicate across the page

When Imagetwin identifies duplicated images on the same page, the matching regions are marked with corresponding colored outlines. To compare the detected matches more easily, select Area in the upper-right corner. This will display the related images—or specific image sections—next to each other for a side-by-side review.

Finding type 2: Duplicate across an external publication

If Imagetwin detects a match with images published in an external source, it displays relevant publication details along with available reference links, such as DOI or PubMed Central records. The review process remains the same as for internal matches: click Area to focus on the detected regions and compare the images or image segments side by side.

Finding type 3: Manipulations

Imagetwin can identify a wide range of image manipulations commonly found in research publications, including splicing, cloning, concealed edits, and more.

For each detected manipulation, Imagetwin provides three views to support the review process:

• Original Image (top): the unmodified source image.

• Enhanced View (middle): a filtered version designed to make potential manipulations more visible, such as splice boundaries or cloned regions.

• Suspicious Area View (bottom): a detailed close-up of the flagged region, allowing for closer inspection. You can customize the visualization by applying different filters and adjusting the heatmap settings, making it easier to examine potential issues in greater detail.

Finding type 4: AI-generated images

Each paper scanned by Imagetwin will also be scanned using our AI detector for images to help identify AI-generated visuals. Each flagged image includes an “AI Assessment” showing a distribution of which generator was most likely used. For example, you may see a score indicating a higher probability of generation by Firefly (Adobe's family of generative AI tools) compared to DALL-E or Stable Diffusion. This gives you more context to evaluate flagged cases.