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

Abstract

Digital imaging has provided scientists with new opportunities to acquire and manipulate data using techniques that were difficult or impossible to employ in the past. Because digital images are easier to manipulate than film images, new problems have emerged. One growing concern in the scientific community is that digital images are not being handled with sufficient care. The problem is twofold: (1) the very small, yet troubling, number of intentional falsifications that have been identified, and (2) the more common unintentional, inappropriate manipulation of images for publication. Journals and professional societies have begun to address the issue with specific digital imaging guidelines. Unfortunately, the guidelines provided often do not come with instructions to explain their importance. Thus they deal with what should or should not be done, but not the associated ‘why’ that is required for understanding the rules. This article proposes 12 guidelines for scientific digital image manipulation and discusses the technical reasons behind these guidelines. These guidelines can be incorporated into lab meetings and graduate student training in order to provoke discussion and begin to bring an end to the culture of “data beautification”.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Notes

  1. 1.

    Underlined terms are defined in a glossary that is provided.

  2. 2.

    While that article had many excellent points, several of the examples described image manipulations that were performed on specific areas of the image. Today, many of these manipulations would be considered falsifications or fabrications, unless the figure legend or methods section contained a detailed explanation of how the figures were created. Interestingly, Mr. Hayden has become more conservative with regard to image manipulations since that article was written in the year 2000. (See interview with Jaime Hayden [Couzin 2006].)

  3. 3.

    In at least one instance that he is aware of, Rossner has seen a paper that was rejected by the JCB that was subsequently published in a different journal without corrections to the inappropriate image manipulations (Young 2008).

  4. 4.

    Not all journals employ this pre-publication examination. Interestingly, if pre-publication screening had been applied to the infamous Hwang stem cell cloning paper in Science (Neill 2006), questions might have been raised before its publication, instead of afterwards (Rossner 2006).

  5. 5.

    A survey performed by the American Journal of Respiratory Cell and Molecular Biology found that figures in 23% of the accepted articles in that journal had images that had undergone some alteration, including ‘erasure or filling in of parts of the background, splicing of bands from one gel into another, and “cloning,”’ (Abraham et al. 2008). The Journal of Clinical Investigation has seen some evidence of tampering in 10–20 accepted articles per year, and about 5–10 of those papers warranted a more thorough investigation (JCI publishes about 300–350 articles per year) (Young 2008). A pilot study carried out by Blood found that “approximately 20% of accepted manuscripts contained one or more figures with digital images that had been manipulated inappropriately” (Shattil 2007).

  6. 6.

    The Microscopy Society of America position on Ethical Digital Imaging considers gamma correction to be a “generally, acceptable (non-reportable) imaging operation” (Microscopy Society of America 2003). The Instructions to Authors for the JCB state that “Non-linear adjustments (e.g., changes to gamma settings) must be disclosed in the figure legend” (Journal of Cell Biology 2009). The Nature Publishing Group states “If ‘Pseudo-coloring’ and nonlinear adjustment (for example ‘gamma changes’) are used, this must be disclosed” (Nature 2009).

  7. 7.

    Phillip Sharp, co-chair of the National Academies of Sciences committee that was initially tasked with coming up with general data handling guidelines (a task that was begun in response to a call from the editors of major journals regarding the problem of inappropriate image manipulation), acknowledged in an interview in Science (Kaiser 2009) that coming up with acceptable image manipulation guidelines in the committee became impossible. Sharp said that “The problem was that every time a panelist made a detailed proposal, another member would say it would not work in their field…” (Kaiser 2009). The National Academy’s report, Ensuring the Integrity, Accessibility, and Stewardship of Research Data in the Digital Age (Committee on Ensuring the Utility and Integrity of Research Data in a Digital Age, National Academy of Sciences 2009), ultimately dealt with issues of researcher’s responsibilities for data integrity, data accessibility and archiving.

  8. 8.

    For more information on the “Basic Properties of Digital Images,” see the Molecular Expressions web site at: http://micro.magnet.fsu.edu/primer/digitalimaging/digitalimagebasics.html (Retrieved 12/06/2009).

  9. 9.

    These numbers (Russ 2004) are at the lower end of the scale. There are higher values that are quoted (without references) for the number of grey shades and colors the human eye can perceive. Methodologies for accurately determining these values are difficult. Do trained observers, such as artists and photographers, see more colors than the average person? No one appears to know for sure. The possibility that some women may have a fourth optical pigment (Jameson et al. 2001) complicates this issue even more.

  10. 10.

    John Krueger (2007) Office of Research Integrity, U.S. Department of Health and Human Services, personal communication.

  11. 11.

    Caveat “…using DAB as a chromogen is problematic because a linear relationship between the amount of antigen and staining intensity exists only at low levels of the latter.” (Bernardo et al. 2009; see also Taylor and Levenson 2006).

  12. 12.

    ISO = International Standards Organization, ITU = International Telecommunications Union.

  13. 13.

    JPEG artifacts—http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/jpegcompression/ (Retrieved 12/07/2009).

  14. 14.

    An example of a temporal sampling artifact can be seen when watching the wheels on an automobile. At certain speeds it will appear as if the wheels are rotating in a direction that is opposite of the direction of travel of the vehicle. This artifact, sometimes referred to as the “wagon wheel effect”, has been known for a long time (“Why Movie Wheels Turn Backward; An explanation of the illusion and a suggested method for correcting it,” 1918) and is caused by the sampling rate of the image capture device or the rate at which the eye/brain processes the images (Purves et al. 1996).

  15. 15.

    For resizing tips for Adobe Photoshop, see “Potentially the most dangerous dialog box in Adobe Photoshop™”—http://swehsc.pharmacy.arizona.edu/exppath/resources/pdf/Photoshop_Image_Size_dialog_box.pdf (Retrieved 12/06/2009).

  16. 16.

    Based on a Google search (November 2009), the following (non-exhaustive) list of journals have image submission and manipulation guidelines which are very similar, or identical to, the wording used by the Journal of Cell Biology; J Gen. Physiology, J Exp. Medicine, Biology of the Cell, J App. Physiology, Biochem J, J Invest Derm, Blood, J. Exp. Botany, J Endocrinology, ASN Neuro, European Resp. Rev., and Diabetes. In addition, publisher Springer (http://www.springer.com/authors?SGWID=0-111-7-574914-0) uses similar wording.

References

  1. Abbott, A. (1997). Forged images lead to German inquiry. Nature, 387(6632), 442. doi:10.1038/387442a0.

    Google Scholar 

  2. Abraham, E. (2007). Update on the AJRCCM-2007. American Journal of Respiratory and Critical Care Medicine, 175(3), 207–208. doi:10.1164/rccm.200612-1784ED.

    Article  Google Scholar 

  3. Abraham, E., Adler, K. B., Shapiro, S. D., & Leff, A. R. (2008). The ATS journals’ policy on image manipulation. Proceedings of the American Thoracic Society, 5(9), 869. doi:10.1513/pats.200809-106ED.

    Article  Google Scholar 

  4. Adler, J. (2005). Veracity of raw images can also come into question. Nature, 435(7043), 736. doi:10.1038/435736b.

    Article  Google Scholar 

  5. Adobe Systems. (2002). Adobe Photoshop 7.0, lesson 17-setting up your monitor for color management. San Jose, CA: Adobe Systems Inc.

    Google Scholar 

  6. Adobe Systems. (2005). Adobe Photoshop CS2 user guide for windows. San Jose, CA: Adobe Systems Inc.

    Google Scholar 

  7. Aldhous, P., & Reich, E. S. (2009). Further doubts over stem-cell images. New Scientist, 203(2720), 2.

    Article  Google Scholar 

  8. American Academy of Dermatology. (1997). Position statement on photographic enhancement. Retrieved 12/04/2009, from http://www.aad.org/Forms/Policies/Uploads/PS/PS-Photographic%20Enhancement.pdf.

  9. Anderson, C. (1994). Easy-to-alter digital images raise fears of tampering. Science, 263(5145), 317–318. doi:10.1126/science.8278802.

    Article  Google Scholar 

  10. Archives of the Confocal Listserver. (1995). Subject: Image enhancement. Retrieved 12/04/2009, from http://lists.umn.edu/cgi-bin/wa?S2=CONFOCALMICROSCOPY&q=enhancement&s=&f=&a=95%2F1%2F1&b=95%2F3%2F1.

  11. Archives of the Microscopy Listserver. (1998). Subject: Image manipulation. Retrieved 11/21/2009, from http://www.microscopy.com/MicroscopyListserver/SearchMLArchive.html

  12. Bagley, K. (2009). Immunologist faked data (Blog). The Scientist (Dec 1, 2009). Retrieved 12/04/2009, from http://www.the-scientist.com/blog/display/56192/.

  13. Baird, D., & Cohen, M. S. (1999). Why trade? Perspectives on Science, 7.2, 231–254. Retrieved 12/04/2009, from http://muse.jhu.edu/journals/perspectives_on_science/v007/7.2baird02.html.

  14. Benham, G. S. (2002). Practical aspects of objective lens selection for confocal and multiphoton digital imaging techniques. In B. Matsumoto (Ed.), Methods in cell biology: Cell biological applications of confocal microscopy (2nd ed., Vol. 70, pp. 245–299). San Diego: Academic Press.

    Google Scholar 

  15. Benos, D. (2006). Ethics: Detecting misconduct (Blog). Nature.com Peer-to-Peer (June 20, 2006). Retrieved 12/04/2009, http://blogs.nature.com/peer-to-peer/2006/06/ethics_detecting_misconduct.html.

  16. Bernardo, V., Lourenco, S. Q., Cruz, R., Monteiro-Leal, L. H., Silva, L. E., Camisasca, D. R., et al. (2009). Reproducibility of immunostaining quantification and description of a new digital image processing procedure for quantitative evaluation of immunohistochemistry in pathology. Microscopy and Microanalysis, 15(4), 353–365. doi:10.1017/S1431927609090710.

    Article  Google Scholar 

  17. Castillo, M. (2008). Digital forensics and the American Journal of Neuroradiology. American Journal of Neuroradiology, 29(2), 211–212. doi:10.3174/ajnr.A0914.

    Article  Google Scholar 

  18. Coburn, M. (2008). Farid founds ‘digital forensics’. The Dartmouth (March 27, 2008). Retrieved 12/04/2009 from http://thedartmouth.com/2008/03/27/news/farid.

  19. Committee on Ensuring the Utility and Integrity of Research Data in a Digital Age, National Academy of Sciences. (2009). Ensuring the integrity, accessibility, and stewardship of research data in the digital age. Washington, DC: National Academies Press.

    Google Scholar 

  20. Council of Science Editors. (2009). CSE’s white paper on promoting integrity in scientific journal publications, 2009 update. Retreived 12/04/2009 from http://www.councilscienceeditors.org/editorial_policies/whitepaper/entire_whitepaper.pdf.

  21. Couzin, J. (2006). Scientific publishing. Don’t pretty up that picture just yet. Science, 314(5807), 1866–1868. doi:10.1126/science.314.5807.1866.

    Article  Google Scholar 

  22. Cromey, D. W. (2001). Digital imaging: Ethics. Retrieved 12/04/2009, from http://swehsc.pharmacy.arizona.edu/exppath/micro/digimage_ethics.php.

  23. Cutrone, M., & Grimalt, R. (2001). The true and the false: Pixel-byte syndrome. Pediatric Dermatology, 18(6), 523–526. doi:10.1046/j.1525-1470.2001.1862009.x.

    Article  Google Scholar 

  24. David, P. (1998). News concreteness and visual-verbal association: Do news pictures narrow the recall gap between concrete and abstract news? Human Communication Research, 25(2), 180–201.

    Article  Google Scholar 

  25. Foster, B. (2000). Adobe Photoshop—the surprise scientific image processing software of choice? Advanced Imaging, November 2000, 49–50.

  26. Frankel, F. (2002). Envisioning science: The design and craft of the science image. Cambridge, MA: MIT Press.

    Google Scholar 

  27. Frankel, F. (2004). The power of the ‘pretty picture’. Nature Materials, 3(7), 417–419. doi:10.1038/nmat1166.

    Article  Google Scholar 

  28. Furness, P. N. (1997). The use of digital images in pathology. Journal of Pathology, 183(3), 253–263. doi:10.1002/(SICI)1096-9896(199711)183:3<253:AID-PATH927>3.0.CO;2-P.

    Article  Google Scholar 

  29. Gravitz, L. (2006). Biology’s image problem. Rockefeller University Scientist, 2006(1), 10–12.

    Google Scholar 

  30. Greenberg, D. S. (1996). US genome chief withdraws five papers over fraud. Lancet, 348(9037), 1303. doi:10.1016/S0140-6736(05)65768-6.

    Article  Google Scholar 

  31. Guneri, P., & Akdeniz, B. G. (2004). Fraudulent management of digital endodontic images. International Endodontic Journal, 37(3), 214–220. doi:10.1111/j.0143-2885.2004.00780.x.

    Article  Google Scholar 

  32. Hayden, J. E. (2000). Digital manipulation in scientific images: Some ethical considerations. Journal of Biocommunication, 27(1), 11–19.

    Google Scholar 

  33. Horn, J. W., & Sterchi, D. L. (2005). One perspective: Regulatory impact on digital imaging in the electron microscopy laboratory. Microscopy and Microanalysis, 11(Suppl 2), 1238–1239. doi:10.1017/S1431927605501466.

    Google Scholar 

  34. Image Analysis Lab. (ca. 1995). Image editing ethics. Seattle, WA: Fred Hutchinson Cancer Center (no longer available online).

  35. Jameson, K. A., Highnote, S. M., & Wasserman, L. M. (2001). Richer color experience in observers with multiple photopigment opsin genes. Psychonomic Bulletin and Review, 8(2), 244–261.

    Google Scholar 

  36. Joint Photographic Experts Group. (2007). Applications: Scientific and industrial. Retrieved 11/6/2009, from http://www.jpeg.org/apps/scientific.html.

  37. Journal of Cell Biology. (2009). Instructions for authors. Journal of Cell Biology. Retrieved 12/04/2009, from http://jcb.rupress.org/misc/ifora.shtml.

  38. Kaiser, J. (2009). Scientific publishing. Data integrity report sends journals back to the drawing board. Science, 325(5939), 381. doi:10.1126/science.325_381.

    Article  Google Scholar 

  39. Katsnelson, A. (2007). Former UPenn postdoc faked images. The Scientist (Aug 7, 2007). Retrieved 12/07/2009, from http://www.the-scientist.com/news/home/53469/.

  40. Krueger, J. (2002). Forensic examination of questioned scientific images. Accountability in Research: Policies and Quality Assurance, 9(2), 105–125. doi:10.1080/08989620212970.

    Google Scholar 

  41. Krueger, J. (2005). Confronting manipulation of digital images in science. Office of Research Integrity Newsletter, 13, 8–9.

    Google Scholar 

  42. Krueger, J. (2009). Incidences of ORI cases involving falsified images. Office of Research Integrity Newsletter, 17, 2–3.

    Google Scholar 

  43. Long, J. (1999). Ethics in the age of digital photography. Retrieved 12/04/2009, from http://www.nppa.org/professional_development/self-training_resources/eadp_report/.

  44. MacKenzie, J. M., Burke, M. G., Carvalho, T., & Eades, A. (2006). Ethics and digital imaging. Microscopy Today, 14, 40–41.

    Google Scholar 

  45. McCabe, H., & Wright, J. (2000). Tangled tale of a lost, stolen and disputed coelacanth. Nature, 406(6792), 114. doi:10.1038/35018247.

    Article  Google Scholar 

  46. McInnes, S. J. (2001). Is it real? Zoologischer Anzeiger—A Journal of Comparative Zoology, 240(3–4), 467–469. doi:10.1078/0044-5231-00055.

    Article  Google Scholar 

  47. McNamara, G. (2006). Crusade for publishing better light micrographs—light microscope publication guidelines. Retrieved 12/04/2009, from http://home.earthlink.net/~geomcnamara/CrusadeBetterMicrographs.htm.

  48. Microscopy Society of America. (2003). Position on ethical digital imaging. Microscopy Today, 11, 61.

    Google Scholar 

  49. Microsoft Corporation. (1997). The Microsoft Press ® computer dictionary (3rd ed.). Redmond, WA: Microsoft Corporation.

    Google Scholar 

  50. Mullin, L. (1998). Truth in photography: Perception, myth and reality in the postmodern world. (Master’s thesis) Retrieved 12/04/2009, from http://etd.fcla.edu/UF/amd0040/Leslie.pdf.

  51. National Press Photographers Association. (1990). Digital manipulation code of ethics. Retrieved 12/04/2009, from http://www.nppa.org/professional_development/business_practices/digitalethics.html.

  52. National Press Photographers Association. (2004). Mission & by laws: Code of ethics standards. Retrieved 12/04/2009, from http://www.nppa.org/about_us/governance/bylaws.html#ethics.

  53. Nature. (2009). Editorial policy: Image integrity and standards. Nature. Retrieved 12/04/2009, from http://www.nature.com/authors/editorial_policies/image.html.

  54. Nature Cell Biology Editorial. (2004a). Gel slicing and dicing: A recipe for disaster. Nature Cell Biology, 6(4), 275. doi:10.1038/ncb0404-275.

    Article  Google Scholar 

  55. Nature Cell Biology Editorial. (2004b). Images to reveal all? [Editorial]. Nature Cell Biology, 6, 909–909. doi:10.1038/ncb1004-909.

    Article  Google Scholar 

  56. Nature Cell Biology Editorial. (2006a). Appreciating data: warts, wrinkles and all. Nature Cell Biology, 8(3), 203–203. doi:10.1038/ncb0306-203a.

    Google Scholar 

  57. Nature Cell Biology Editorial. (2006b). Beautification and fraud. Nature Cell Biology, 8(2), 101–102. doi:10.1038/ncb0206-101.

    Article  Google Scholar 

  58. Nature Cell Biology Editorial. (2007). Imagine. Nature Cell Biology, 9(4), 355. doi:10.1038/ncb0407-355a.

    Google Scholar 

  59. Nature Cell Biology Editorial. (2009). Accurately reporting research. Nature Cell Biology, 11(9), 1045–1045. doi:10.1038/Ncb0909-1045.

    Article  Google Scholar 

  60. Nature Editorial. (2006). Not picture-perfect. Nature, 439(7079), 891–892. doi:10.1038/439891b.

    Google Scholar 

  61. Neale, A. V., Dailey, R. K., & Abrams, J. (2010). Analysis of citations to biomedical articles affected by scientific misconduct. Science and Engineering Ethics, 16, 251–261. doi:10.1007/s11948-009-9151-4.

    Article  Google Scholar 

  62. Neill, U. S. (2006). Stop misbehaving!. Journal of Clinical Investigation, 116(7), 1740–1741. doi:10.1172/JCI28824.

    Article  Google Scholar 

  63. NIH. (1994). Reminder and update: Requirement for instruction in the responsible conduct of research in national research service award institutional training grants. (NOT-94-200). Retrieved 12/04/2009 from http://grants1.nih.gov/grants/guide/notice-files/not94-200.html.

  64. NIH. (2009). Update on the requirement for instruction in the responsible conduct of research (NOT-94-200). Retrieved 12/04/2009 from http://grants.nih.gov/grants/guide/notice-files/NOT-OD-10-019.html.

  65. Normile, D. (2009). Scientific misconduct. Science retracts discredited paper; bitter patent dispute continues. Science, 324(5926), 450–451. doi:10.1126/science.324_45.

    Article  Google Scholar 

  66. North, A. J. (2006). Seeing is believing? A beginners’ guide to practical pitfalls in image acquisition. Journal of Cell Biology, 172(1), 9–18. doi:10.1083/jcb.200507103.

    Article  Google Scholar 

  67. Nouraei, S. A., Frame, J., & Nduka, C. (2005). Uses and abuses of digital imaging in plastic surgery. International Journal of Surgery, 3(4), 254–257. doi:10.1016/j.ijsu.2005.10.004.

    Article  Google Scholar 

  68. Oliver, W. R. (1998). Image processing and scientific misconduct. Microscopy Today, 6, 12–13.

    Google Scholar 

  69. Paalman, M. H. (2000). Scientific misconduct: The ultimate negative career move. Anatomical Record, 261(6), 219–220. doi:10.1002/1097-0185(20001215)261:6<219:AID-AR1001>3.0.CO;2-Y[pii].

    Article  Google Scholar 

  70. Pawley, J. B. (2000). The 39 steps: A cautionary tale of quantitative 3-D fluorescence microscopy. Biotechniques, 28(5), 884–886, 888.

    Google Scholar 

  71. Pawley, J. B. (2006). Points, pixels, and gray levels: Digitizing image data. In J. B. Pawley (Ed.), Handbook of biological confocal microscopy (3rd ed., pp. 59–79). New York, NY: Springer.

    Google Scholar 

  72. Pearson, H. (2005). Image manipulation: CSI: Cell biology. Nature, 434(7036), 952–953. doi:10.1038/434952a.

    Article  Google Scholar 

  73. Pearson, H. (2006). Forensic software traces tweaks to images. Nature, 439(7076), 520–521. doi:10.1038/439520b.

    Article  Google Scholar 

  74. Pearson, H. (2007). The good, the bad and the ugly. Nature, 447(7141), 138–140. doi:10.1038/447138a.

    Article  Google Scholar 

  75. Pritt, B. S., Gibson, P. C., & Cooper, K. (2003). Digital imaging guidelines for pathology: A proposal for general and academic use. Advances in Anatomic Pathology, 10(2), 96–100.

    Article  Google Scholar 

  76. Purves, D., Paydarfar, J. A., & Andrews, T. J. (1996). The wagon wheel illusion in movies and reality. Proceedings of the National Academy of Science USA, 93(8), 3693–3697.

    Article  Google Scholar 

  77. Rasband, W. S. (1997–2009). ImageJ (Software). Retrieved from http://rsb.info.nih.gov/ij/.

  78. Revel, J.-P. (1993a). Finger painting or digital imaging. Microscopy Today, 1(5), 2.

    Google Scholar 

  79. Revel, J.-P. (1993b). The truth in imaging. Microscopy Today, 1(4), 2.

    Google Scholar 

  80. Richardson, M. L., Frank, M. S., & Stern, E. J. (1994). Digital image manipulation: What constitutes acceptable alteration of a radiologic image? [opinion]. American Journal of Roentgenology, 164, 228–229.

    Google Scholar 

  81. Rolph, A., & McNerthney, C. (2007). UW: AIDS researcher falsified data, news. SeattlePI.com (November 28, 2007). Retrieved 12/07/2009 from http://www.seattlepi.com/local/341421_research28.html.

  82. Rossner, M. (2006). How to guard against image fraud. The Scientist, 20, 24.

    Google Scholar 

  83. Rossner, M. (2008). A false sense of security. Journal of Cell Biology, 183(4), 573–574. doi:10.1083/jcb.200810172.

    Article  Google Scholar 

  84. Rossner, M., Held, M. J., Bozuwa, G. P., & Kornacki, A. (1998). Managing editors and digital images: Shutter diplomacy. CBE Views, 21(6), 187–192.

    Google Scholar 

  85. Rossner, M., & O’Donnell, R. (2004). The JCB will let your data shine in RGB. Journal of Cell Biology, 164(1), 11–13. doi:10.1083/jcb.200312069.

    Article  Google Scholar 

  86. Rossner, M., & Yamada, K. M. (2004). What’s in a picture? The temptation of image manipulation. Journal of Cell Biology, 166(1), 11–15. doi:10.1083/jcb.200406019.

    Article  Google Scholar 

  87. Russ, J. C. (1998). The image processing handbook (3rd ed.). Boca Raton, FL: CRC Press.

    Google Scholar 

  88. Russ, J. C. (2004). Seeing the scientific image (parts 1–3). Proceedings Royal Microscopy Society, 39(2), 97–114; (3), 179–194; (4), 267–281.

  89. Sacchi, D. L. M., Agnoli, F., & Loftus, E. F. (2007). Changing history: Doctored photographs affect memory for past public events. Applied Cognitive Psychology, 21(8), 1005–1022. doi:10.1002/acp.1394.

    Article  Google Scholar 

  90. Schekman, R. (2008). Charting the course for PNAS. Proceedings of the National Academy of Sciences of the United States of America, 105(8), 2755–2756. doi:10.1073/pnas.0800528105.

  91. Scientific Working Group Imaging Technology. (2004). Best practices for documenting image enhancement, section 11 (Version 1.2 2004.03.04). Retrieved 12/04/2009 from http://www.theiai.org/guidelines/swgit/guidelines/section_11_v1-2.pdf.

  92. Shattil, S. J. (2007). A digital exam for hematologists [Editorial]. Blood, 109(9), 2275. doi:10.1182/blood-2007-01-070128.

    Article  Google Scholar 

  93. Spring, K. R., Fellers, T. J., & Davidson, M. W. (2006a). Resolution and contrast in confocal microscopy. Retrieved 11/6/2009, from http://www.olympusconfocal.com/theory/resolutionintro.html.

  94. Spring, K. R., Parry-Hill, M. J., Long, J. C., Fellers, T. J., & Davidson, M. W. (2006b). Spatial resolution in digital images. Retrieved 11/9/2009, from http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/spatialresolution/.

  95. Spring, K. R., Russ, J. C., Parry-Hill, M. J., Fellers, T. J., Zuckerman, L. D., & Davidson, M. W. (2006c). Digital image sampling frequency. Retrieved 11/6/2009, from http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/samplefrequency/index.html.

  96. Spring, K. R., Russ, J. C., Parry-Hill, M. J., Long, J. C., Fellers, T. J., & Davidson, M. J. (2007). Convolution kernel mask operation. Retrieved 12/04/2009, from http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/kernelmaskoperation/index.html.

  97. Steneck, N. H. (2007). ORI introduction to the responsible conduct of research. Retrieved 12/04/2009, from http://ori.dhhs.gov/publications/ori_intro_text.shtml.

  98. Suprock Technologies. (2009). Signal processing & data analysis: Rigour. Retrieved 11/21/2009, from http://www.suprocktech.com/solutions/signal-processing-data-analysis.aspx.

  99. Suvarna, S. K., & Ansary, M. A. (2001). Histopathology and the ‘third great lie’. When is an image not a scientifically authentic image? Histopathology, 39(5), 441–446. doi:10.1016/S0002-9394(02)01550-7.

    Article  Google Scholar 

  100. Taubes, G. (1994). Technology for turning seeing into believing. Science, 263(5145), 318. doi:10.1126/science.8278803.

    Article  Google Scholar 

  101. Taylor, C. R., & Levenson, R. M. (2006). Quantification of immunohistochemistry—issues concerning methods, utility and semiquantitative assessment II. Histopathology, 49(4), 411–424. doi:10.1111/j.1365-2559.2006.02513.x.

    Article  Google Scholar 

  102. Tengowski, M. W. (2004). Image compression in morphometry studies requiring 21 CFR Part 11 compliance: Procedure is key with TIFFs and various JPEG compression strengths. Toxicologic Pathology, 32(2), 258–263. doi:10.1080/01926230490274399.

    Article  Google Scholar 

  103. Vogel, G. (2006). Developmental biology. Fraud investigation clouds paper on early cell fate. Science, 314(5804), 1367–1369. doi:10.1126/science.314.5804.1367.

    Article  Google Scholar 

  104. Vogel, G. (2008). Scientific misconduct. Falsification charge highlights image-manipulation standards. Science, 322(5900), 356. doi:10.1126/science.322.5900.356.

    Article  Google Scholar 

  105. Wager, E., Fiack, S., Graf, C., Robinson, A., & Rowlands, I. (2009). Science journal editors’ views on publication ethics: Results of an international survey. Journal of Medical Ethics, 35(6), 348–353. doi:10.1136/jme.2008.028324.

    Article  Google Scholar 

  106. Waters, J. C. (2009). Accuracy and precision in quantitative fluorescence microscopy. Journal of Cell Biology, 185(7), 1135–1148. doi:10.1083/jcb.200903097.

    Article  Google Scholar 

  107. Weissmann, G. (2006). Science fraud: From patchwork mouse to patchwork data. FASEB Journal, 20(6), 587–590. doi:10.1096/fj.06-0401ufm.

    Article  Google Scholar 

  108. Wheeler, T. (2002). Phototruth or photofiction?: Ethics and media imagery in the digital age. Mahwah, NJ: Lawrence Erlbaum Associates.

    Google Scholar 

  109. Why Movie Wheels Turn Backward. (1918). An explanation of the illusion and a suggested method for correcting it. New York Times (Jul 21, 1918). Retrieved 12/07/2009, from http://query.nytimes.com/gst/abstract.html?res=9A05E6DA143EE433A25752C2A9619C946996D6CF.

  110. Wright, D. E., Titus, S. L., & Cornelison, J. B. (2008). Mentoring and research misconduct: An analysis of research mentoring in closed ORI cases. Science and Engineering Ethics, 14(3), 323–336. doi:10.1007/s11948-008-9074-5.

    Article  Google Scholar 

  111. Xin, H. (2006). Scientific misconduct. Online sleuths challenge cell paper. Science, 314(5806), 1669. doi:10.1126/science.314.5806.1669a.

    Article  Google Scholar 

  112. Young, J. R. (2008). Journals find fakery in many images submitted to support research. The Chronicle of Higher Education (May 29, 2008). Retrieved 12/04/2009, from http://chronicle.com/article/Journals-Find-Fakery-in-Man/846/.

  113. Zwier, J. M., Van Rooij, G. J., Hofstraat, J. W., & Brakenhoff, G. J. (2004). Image calibration in fluorescence microscopy. Journal of Microscopy, 216(Pt 1), 15–24. doi:10.1111/j.0022-2720.2004.01390.x.

    Article  Google Scholar 

Download references

Acknowledgements

This essay began as a brief two-page newsletter article in February of 2001 that was intended primarily for graduate students and staff. As the guidelines have been refined and revised over the last several years, I have benefited greatly from the insight and feedback of colleagues at the University of Arizona, with specific thanks to: Carl Boswell, David Elliott, Patty Jansma, R. Clark Lantz, Claire Payne, Dana Wise, and Jeb Zirato. Additional feedback from John Krueger of the Office of Research Integrity, and Sara Vollmer of the University of Alabama—Birmingham, is appreciated. The author would like to specifically thank Michael W. Davidson and his colleagues at the Molecular Expressions website (Florida State University) for developing the online resources that carefully explain some of the technical concepts referred to in this article. Adobe and Photoshop are registered trademarks of Adobe Systems Incorporated, San Jose, CA. Microsoft, Powerpoint, and Windows are registered trademarks of the Microsoft Corporation, Redmond, WA. Apple and Macintosh are registered trademarks of Apple Computer, Inc., Cupertino, CA. Corel and Photo-Paint are registered trademarks of the Corel Corporation, Ottawa, Ontario, Canada. This work was supported in part by the Southwest Environmental Health Sciences Center (SWEHSC), a National Institute of Environmental Health Sciences (NIEHS) funded center (ES006694). The views, opinions, and conclusions of this essay are not necessarily those of the SWEHSC, the NIEHS, or the University of Arizona.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Douglas W. Cromey.

Glossary

Glossary

Term Definition
Aliasing Because pixels are square and biological structures rarely have straight edges, there are many approximations performed when a digital image is acquired. If an edge falls in the middle of a pixel, the average of the light and dark parts of the edge are reported as the intensity value of the pixel. This creates a pixel with a value that is intermediate between the light and dark intensities in the original (see Fig. 3). Aliasing is the stair-step artifact seen when these intermediate values are not created. Anti-aliasing, sometimes referred to as dithering, is when these intermediate pixels smooth out an edge to create an image that better represents the appearance of curved edges and is generally more pleasing to the eye. See: http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/undersampling/index.html (Retrieved 12/06/2009).
Background subtraction (Also referred to as: flat-field correction or normalization) Microscope optics can be dirty and/or misaligned and CCD image sensors can have unequal sensitivities across the chip (e.g., “hot” or “dead” pixels). By collecting a background image under the same conditions as the specimen image, the background can be subtracted from the specimen image to correct for many of these problems. The use of background subtraction should be acknowledged in the figure legend or the methods section. See: http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/backgroundsubtraction/index.html (Retrieved 12/06/2009).
Bit depth Describes the number of grey shades or colors in an image. Most greyscale images are 8 bit (28 = 256 shades). Using a higher bit depth, like 16 bit, yields a much higher number of greyscales (216 = 65,536). Color is often 24 bit: 8 bits each of red, green and blue (224 = 16.7 million).
Black level The threshold at which a signal will be detected. If the signal for a given pixel is below the threshold, that particular pixel will be displayed as black (a value of 0 in an 8 bit greyscale image). By adjusting the black level, the amount of background electronic noise (and low level signal) in a detection system can be reduced.
CCD Charge-coupled device—a light-sensitive semi-conductor chip that is used in most scientific digital cameras, as well as in many consumer digital cameras and digital video recorders. See: http://learn.hamamatsu.com/articles/ccdanatomy.html (Retrieved 12/06/2009).
Contrast stretch (Also known as a histogram stretch) A technique used to improve the contrast in an image without adding any additional data. Involves remapping the brightness of all pixels (so that the brightest intensity in the image is defined as white and the darkest intensity is defined as black) to maximize the use of the available dynamic range in the image. After using this technique, the intensity histogram typically shows gaps where there was once (usually) a continuous range of intensities. The general consensus seems to be that performing this procedure on an image does not need to be reported in the figure legend or the methods section. See: http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/histogramstretching/index.html (Retrieved 12/06/2009).
Dodging and burning Darkroom techniques where a small portion of a photographic print is exposed to less or more light (respectively), than the rest of the print. Dodging would be used to reduce the intensity of a selected area. Burning would be used to increase the intensity of a selected area. This technique was rarely admitted in the past, however, performing similar techniques today must be acknowledged in either the figure legend or the methods section.
Gamma A non-linear technique that preferentially adjusts the mid-tones in an image. The curves and levels adjustment tools in Photoshop can be used to change the image gamma. The manipulation of image gamma should be acknowledged in the figure legend or the methods section. See: http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/gamma/index.html (Retrieved 12/06/2009).
Histogram equalization A useful, but non-linear, technique for improving the apparent contrast in an image that can alter the relationship between brightness and structure (Russ 1998). This technique is often the basis for the auto-contrast tool in many imaging programs. The use of histogram equalization should be acknowledged in the figure legend or the methods section. See: http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/histogramspecification/index.html (Retrieved 12/06/2009).
Intensity histogram A graph provided in most image processing programs. In an 8 bit greyscale image the X axis displays the greyscale intensity and the Y axis displays the number of pixels at the particular intensity value. For 24 bit color images there are typically three separate intensity histograms, each representing the 8 bit values in the red, green and blue channels.
JPEG An acronym for the Joint Photographic Experts Group. An International Standards Organization (ISO), International Telecommunication Union (ITU) standard for storing bitmapped images in a compressed form using a discrete cosine transform. The JPEG file format uses lossy compression. Users can adjust the degree of compression when the file is saved (Microsoft Corporation 1997).
Interpolation The estimation of intermediate values between two known values in a sequence (Microsoft Corporation 1997).
Loss-less file compression “The process of compressing a file such that, after being compressed and decompressed, it matches its original format bit for bit. Text, code, and numeric data files must be compressed using a loss-less method; such methods can typically reduce a file to 40 percent of its original size.” (Microsoft Corporation 1997)
Lossy compression “The process of compressing a file such that some data is lost after the file is compressed and decompressed. Video and sound files often contain more information than is apparent to the viewer or listener; a lossy compression method, which does not preserve that excess information, can reduce such data to as little as 5 percent of its original size.” (Microsoft Corporation 1997)
LZW Lempel–Ziv–Welch—A loss-less file compression algorithm that makes use of repeating strings of data in its compression of character streams into code streams (Microsoft Corporation 1997).
See also: http://en.wikipedia.org/wiki/LZW (Retrieved 12/06/2009).
Metadata Data about data (Microsoft Corporation 1997) that can include information regarding the conditions under which the image data were acquired.
Moiré Derived from the French, “to water”. A visible wavy distortion or flickering in an image that is displayed or printed with an inappropriate resolution. Several parameters can cause moiré patterns, including the size and resolution of the image, resolution of the output device, and halftone screen angle (Microsoft Corporation 1997). Moiré artifacts can regularly be seen in broadcast television due to the incorrect sampling of clothing with tight repeating patterns.
Oversampling See: sampling.
Over-saturate Exceeding the maximum capacity of the detector to measure light, sometimes referred to as clipping. (see also—Truncate) See: http://micro.magnet.fsu.edu/primer/digitalimaging/concepts/ccdsatandblooming.html (Retrieved 12/06/2009).
Raster (bitmap) image A rectangular array of picture elements (pixels), with each pixel representing a discreet color or greyscale.
Resolution As defined by the Rayleigh criterion. The ability to discern two adjacent objects as distinct and separate objects. In microscopy resolution can be calculated based on a number of optical factors, but is most strongly influenced by the wavelength of light used and the numerical aperture of the objective lens. Not to be confused with printer or monitor resolution, which is typically given in dots-per-inch (dpi). See: http://micro.magnet.fsu.edu/primer/java/imageformation/rayleighdisks/index.html (Retrieved 12/06/2009).
Sampling The process of turning an analog signal into its digital representation. Sampling refers to the frequency of data points used to represent a continuous analog signal. The Nyquist/Shandon criterion states that analog signals should be sampled using at least twice the frequency of the highest frequency item in the signal (Pawley 2006). As an example, music CDs are created by sampling an analog signal at 44,000 Hz, which is twice the highest frequency that humans can hear, thus just satisfying the Nyquist/Shandon criterion. Oversampling refers to acquiring samples in excess of the criterion, and undersampling does not meet the criterion.
Sub-resolution point object An object that is smaller than the diffraction-limited resolution of a microscope. In fluorescence microscopy, this is often a fluorescent bead of a size >0.2 μm.
TIFF (also Tiff or Tif) Tagged image file format. A raster or bitmap image file format that incorporates embedded tags to include selected metadata. This format was originally developed by the Aldus corporation, which was subsequently acquired by the Adobe Corporation. This is the only image file format that is recommended by the Microscopy Society of America (MacKenzie et al. 2006). See: http://en.wikipedia.org/wiki/TIFF and http://partners.adobe.com/public/developer/tiff/index.html (Retrieved 12/06/2009).
Truncate To cut off the beginning or end of a series of characters or numbers (Microsoft Corporation 1997). In this context, the term is used to refer to pixel data that are beyond the dynamic range displayed in the image and as such the intensity value of these specific data have been truncated to either the brightest or darkest values possible in the image.
Undersampling See: sampling.
Voxel (VOlume piXEL) “A three-dimensional pixel. A voxel represents a quantity of 3D data just as a pixel represents a point or cluster of points in 2D data. It is used in scientific and medical applications that process 3D images.” From: http://www.pcmag.com/encyclopedia_term/0,2542,t=voxel&i=54113,00.asp (Retrieved 12/06/2009).
White-level balancing Digital cameras are not equally sensitive to the three main colors of light (red, green, blue). To compensate for the differences in sensitivity and the different colors of illumination sources used, software can be used to adjust the balance of the colors so that the whites in the image are correctly displayed and, by extension, all the other colors as well. The use of white-level balancing should be acknowledged in the figure legend or the methods section, particularly if the balance was set automatically or differently for different images. See: http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/whitebalance/index.html (Retrieved 12/06/2009).

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Cromey, D.W. Avoiding Twisted Pixels: Ethical Guidelines for the Appropriate Use and Manipulation of Scientific Digital Images. Sci Eng Ethics 16, 639–667 (2010). https://doi.org/10.1007/s11948-010-9201-y

Download citation

Keywords

  • Digital image
  • Ethics
  • Manipulation
  • Image processing
  • Microscopy