home-icon
  • checkmark Responsible conduct of research
    • checkmark Design and conduct
    • checkmark Design and methodology
    • checkmark Possible flaws in a study design
    • checkmark Preregistration and registered reports
    • checkmark Reproducibility and replicability of research
    • checkmark Statistics in research
    • checkmark Research funding
    • checkmark Research Data Management (RDM)
    • checkmark FAIR data principles for research data
    • checkmark Data Management Plan
    • checkmark Reporting results
    • checkmark Presenting your data
    • checkmark Image processing
    • checkmark Authorship
    • checkmark Author affiliation
    • checkmark Citation and referencing
    • checkmark Open access to publications
    • checkmark The quality of a journal
    • checkmark Peer review
    • checkmark Preprints
    • checkmark Novelty of your work
    • checkmark The value of negative results
  • checkmark Declaration of conflict of interest
  • checkmark Science communication
  • checkmark Research(er) evaluation and assessment
  • checkmark References for module 3 - Good Academic Practices

Image processing

jumping-icon base

Image processing

In addition to the presentation of categorical and non-categorical data, special care should be taken when presenting digital images to illustrate different experimental conditions. Images have to be considered as data and are more than a simple illustration accompanying the summary statistics. Inconsistencies in the image (e.g. the use of the same image to illustrate different experimental conditions) or selective modification of images may severely decrease the confidence of your peers in your work and may warrant a correction or even a retraction of the publication. This is especially relevant in the field of biomedical and biological sciences, relying on a number of specific technologies (microscopy, western blotting, fluorescence-activated cell sorting (FACS), …) to illustrate the results.

Good academic practices when working with and presenting images

Some general good practices when working with and presenting images are:

  • Modifications such as adjustments of contrast, brightness and/or color balance might be acceptable but only if the adjustments are done on the entire image and do not influence a proper perception of the data.
  • Researchers should be able to trace back and motivate the adjustments.
  • Selective modification of an image, for example to remove or emphasize specific features is generally not acceptable even if the modification is performed to remove an unrelated imperfection (e.g. remove a hair, fingerprint, etc.).
  • Upon combining multiple images into a single field, this should be obvious from the presentation of the image and the text of the figure legend. For example, splicing of bands in case of a Western blot may be acceptable in some cases, but only if properly acknowledged.
  • Always have the original, unaltered images available and only make modifications on a copy of the original image.
  • Be prepared to make the raw image files available to the reviewers and/or readers of your work, for example by providing these as supplementary data or using an online data repository.
  • As many issues arise by the accidental selection of the wrong image, proper data management is a key element in the prevention of mistakes.
  • While generating figures, check and double check whether the correct images have been selected. With this in mind, the use of placeholder images is discouraged.
mindthegap

Further reading:

mindthegap

“I find it tempting to selectively modify images or provide a non-related image in order to have an image that represents my average results or makes the figure more appealing, this is not an acceptable practice.”

The example below illustrates a case in which the same blot image is used 2 times to illustrate 2 different experimental conditions:

(2016) BIK – Prevalence of inappropriate image duplication in biomedical research publications – bioRxiv preprint.

mindthegap

While this minght be an honest error, it also illustrates how easy it is to make a mistake or to use an unrelated image. In the current example, the issue could be detected as the same image was presented twice. However, in case the image was only used to illustrate the unrelated condition, this would not have been picked up.