Why Image Integrity needs Assessment

Most cases of inappropriate image manipulation are unintentional and rather due to a lack of an in depth knowledge of the image processing software used. A related study showed that around 20-25% of accepted papers contain at least one inappropriately modified image [1]. This is where guidelines and specific training of scientists will bring strong improvements. Nevertheless, quality assurance is important regarding these data alterations [2-9].

Unfortunately, there are also cases of image manipulations on purpose which are hard to prevent in a landscape of publication pressure and impact factors. Investigations carried out by the Office of Research Integrity (ORI) constantly show that misconduct due to image data manipulation and fabrication is not only an isolated case. Those data manipulation of any nature will latest be detected when turning out not to be reproducible [2] but this is way too late. Detection prior to publication is the best option for all instances involved.

Fact is, that many scientific journals neither have the capacities, nor the time and in-depth image processing expertise to screen all submissions for inappropriate image editing or image fabrication on a regular bases.

Early detection of image data manipulation is key to…

• ensure publication of high quality and least biased results!
• maintain public trust and credability in science!
• preserve the reputation of research institutions and journals!
• responsibly and sustainably use public funding resources!
• avoid affecting new research by hypotheses based on altered data!
• make medical development faster and safer!

How BioVoxxel addresses Image Integrity

• screening of images for signs of inappropriate image editing (“beautification”) even beyond existing official, high standard guidelines [1-9]
• pre-publication assessment including certification
• in-depth post-publication assessment of potential image manipulation or fabrication
• strict qualitative and quantitative evaluation based on over 30 different methods and algorithms
• context-related impact assessment by an experienced scientific expert.
• detailed reports on findings.

Partner today with a specialist to ensure data integrity in your institution

Contact BioVoxxel for offers and pricing

Teaching

Education is the best prevention against unintentional but inappropriate image manipulation. BioVoxxel communicates commonly accepted strict standards on image processing for scientific images through workshops for scientists

Consultancy and Support

BioVoxxel offers consultancy regarding the development of customized image analysis tools to reduce user bias during analyses and improve and secure reproducibility, documentation, transparency and reporting.

---

[1] Digital Images Are Data: And Should Be Treated as Such

[2] Beautification and fraud

[3] Seeing the Scientific Image

[4] Science journals crack down on image manipulation

[5] Guidelines for Best Practices in Image Processing

[6] Avoiding Twisted Pixels: Ethical Guidelines for the Appropriate Use and Manipulation of Scientific Digital Images

[7] Digital Image Ethics – University of Arizona (by Douglas Cromey)

[8] What’s in a picture? The temptation of image manipulation

[9] Manipulation and Misconduct in the Handling of Image Data