CONCEPT

Bringing Clarity to Research

Efforts to promote research integrity has been garnering a lot of attention in the academic field. Although solutions for detecting plagiarism in text have been around for many years, an automated system dedicated to identifying manipulated and duplicated images is still not available to this date¹. Finding reused and manipulated images not only requires a considerable level of expertise, but is also a laborious, time-consuming, and overwhelming task, especially when handling an enormous collection of images. ImaChek is the only solution that can automatically detect problematic images in scientific papers prior to the publication stage.

¹ Based on internal investigation


Before

After

FEATURE

  • FEATURE 01

    Enhance and maintain the quality of research

    Images can be checked prior to submission to maintain the quality of research and keep risks related to misconduct to a minimum.

  • FEATURE 02

    Protect the reputation of institutions

    Allegations of research misconduct are costly and damage the reputation of the institution. Images can be checked in advance to prevent the onset of research misconduct.

  • FEATURE 03

    Reduce checking time for faster publication

    Automating the image checking process helps to increase the speed and efficiency for publication.

SOLUTION

Automated Detection of Manipulated and Duplicated Images

To this day, visual inspection is the main method for identifying problematic images in scientific papers. This is, however, an extremely painful task. ImaChek helps to streamline and simplify the image screening process.

  • FUNCTION1Duplication Detection

    Nature News released in 2016 reported that the frequency of using duplicated images in one paper has doubled during the first half in the year 2000 (1). As the number of images used in scientific papers increase, authors may unintentionally reuse the same image. As a result, however, both the researcher and the entire institution are at risk for losing their credibility and reputation. ImaChek is capable of automatically detecting duplications in scientific papers.

  • FUNCTION2Manipulation Detection

    Even though the Journal of Cell Biology devised guidelines on the acceptable standards for image processing in 2004 (2), falsified image data in scientific papers continues to be a problem. Based on an article from Nature News in 2015, 20% of manuscripts submitted to EMBO Press was reported to have contained problematic images. In addition to the detection of image duplication, preventative measures must be established. ImaChek's unique algorithms can help to automatically identify signs of image manipulation.

HOW TO USE

Once the PDF containing the images, or original images of high resolution, is uploaded, ImaChek automatically extracts the images and checks for signs of possible image manipulation and reuse. The detection results can be downloaded as a PDF report.

1st STEPUpload images

2nd STEPView detection results

3rd STEPDownload PDF report

NEWS

2019.11.01
Website renewed.

BLOG

  • 2019.11.07

    公正な研究を目指す研究者と研究機関に向けた 科学論文の不正画像自動検出システム 「ImaChek(イマチェック)」をフルリニューアル

    2019年11月7日(木)、科学論文(ライフサイエンス分野の学術・学位論文)における不正画像の自動検出システム「ImaChek(イマチェック)」をフルリニューアルいたしました。エルピクセルでは、創業当初より「研究公正」に向き合い、科学論文に含まれる画像の改ざんや使い回しを検出するシステムを開発してま...

CONTACT

REFERENCES

"For the first time in the publishing community, the Journal of Cell Biology instituted a series of guidelines regarding the acceptable standards for image processing in 2004(2). Since then, many journals have devised submission guidelines for acceptable standards for image integrity. Along with the reason of the retraction, or the retraction note, previous papers that have been retracted are compiled in a database found in Retraction Watch(3).

ImaChek's detection criteria for problematic images is based on retraction notes and the guidelines instituted by journals. "

1
Problematic images found in 4% of biomedical papers
Monya Baker
doi:10.1038/nature.2016.19802

2
What’s in a picture? The temptation of image manipulation
Mike Rossner, Kenneth M. Yamada
The Journal of Cell Biology Jul 2004, 166 (1) 11-15; DOI: 10.1083/jcb.200406019

3
What a massive database of retracted papers reveals about science publishing’s ‘death penalty’ Science Oct 2018; DOI: 10.1126/science.aav8384