Fact or Artifact?

Prior to last week, my only exposure to "artifacts" was something like this:

This artifact is an ancient Chinese pot, the kind of piece you would see at Chicago's Field Museum, or in a similar collection.

But last week at the lab, Vasily, my research supervisor Olga's husband, told me about biological artifacts. Here is a definition from Biology Online:

Artifact. Any visible result of a procedure which is caused by the procedure itself and not by the entity being analyzed. Common examples include histological structures introduced by tissue processing, radiographic images of structures that are not naturally present in living tissue, and products of chemical reactions that occur during analysis.

Vasily told me that the researcher has to be very careful to distinguish between what might seem like fantastically interesting results and an artifact. I was intrigued. So I looked up some articles on PubMed about these so-called artifacts. Here are some of my findings, along with some pretty cool images that illustrate the concept.

One type of article that kept coming up was about radiological artifacts. Basically, the idea is that the quality of a CT scan or MRI can be compromised, leading to "image artifacts" that can result in the improper diagnosis of a disease. For example, according to one paper I read*, there are several categories of radiological artifacts. The artifact type that I understood the best (the others had to do with rather sophisticaed nuclear technology) results from motion. If there is either involuntary (i.e., sneezing, heart beating) or voluntary (i.e., swallowing) movement, there could be distortions or shadowing on the film. Above is an example of a artifacts pictured in the paper.

On a subject that I understand a bit better (and that results in prettier pictures), there are artifacts on the biological side of research as well. Another article I found** was about a very specific aspect of pancreatic epithelial cells. Its authors reported that a type of cell transition that had been observed in their laboratory was likely an artifact. They hypothesized that their cell isolation procedure might have introduced some type of genetic changes in the cells that caused the artifactual result. Below is an image from this paper, representing the artifact that they observed: the coexpression of MSC antigens CD29 and vimentin in a two-day cultured pancreatic digest. (Don't worry if that's Greek to you. But I know some of you out there are research folks, so I figured I would mention a few details.)



So what's the point, other than looking at cool cell stainings (which is fun on it's own, in my opinion)? The point is that as a scientist, and as a clinician, you have to be cognizant of what is normal. And then when results come back that are abnormal, you take a close look to make sure that there is not an alternate explanation other than true abnormality (or perhaps an amazing discovery, in the case of research). It's always about questioning things. That's the nature of science, isn't it? I think so, at least.





* Popilock, R., Sandrasagaren, K., Harris, L., and Kaser, K.A. (2008). CT artifact recognition for the nuclear technologist. J. Nucl. Med. Technol. 36, 79-81.

**Seeberger, K.L., Eshpeter, A., Rajotte, R.V., and Korbutt, G.S. (2009). Epithelial cells within the human pancreas do not coexpress mesenchymal antigens: epithelial-mesenchymal transition is an artifact of cell culture. Lab. Invest. 89, 110-121.