By Abby S. Van Voorhees, MD, November 01, 2012
In this month’s Acta Eruditorum column, Physician Editor Abby S. Van Voorhees, MD, talks with Ellen Mooney, MD, and Antoinette Foote Hood, MD, about their recent Journal of Cutaneous Pathology article, “Diagnostic accuracy in virtual dermatopathology.”
Dr. Van Voorhees: For those of us not in the know, what is virtual microscopy? Is it being used in dermatopathology? How so?
Dr. Hood: Virtual microscopy is the process whereby, with scanning machinery, you’re able to take a whole mount of a glass slide and get digital imagery of that entire specimen and then, with a computer program, can convert that into an image that can be placed online, sent on a DVD, or put on a computer. The person who is looking at it can go from low power, to see the whole mount, to higher power, to see individual cell detail. So it’s digital imagery of a whole mount of the glass slide, and it can be performed on any tissue. Dermatopathology is particularly well-suited to this process because our pieces of tissue tend to be a bit smaller.
Dr. Mooney: This technique is being used for medical student and resident teaching in the U.S. and Scandinavia. In addition it is regularly used for resident teaching and in conferences in Australia and occasionally in Europe, e.g., Austria, Germany, and the UK. The Biomedical Informatics Research Group in Finland has used virtual images in the European Congress of Pathology and at the Department of Pathology, University of Helsinki, Finland, they host their dermatopathology seminars online in virtual format. It has also caught on quite a bit in CME/continuing professional development, and in external quality assurance. In dermatopathology, it’s mostly utilized in CME. [pagebreak]
Dr. Van Voorhees: Have interpretations of virtual images been compared to those of glass slide readings?
Dr. Hood: Yes, and that’s well-documented in the literature, with one of the earlier articles coming out of our medical school when they were looking at replacing a lot of microscopes for the histology laboratory and the thought was, instead of having to spend all this money on microscopes and old slides, can we digitize our images and teach histology to medical students without microscopes and glass slides? So they did the controlled study and found out that it didn’t really make any difference which way you taught histology.
We followed that study with a comparison of glass slides and virtual dermatopathology images using pathology and dermatology residents. We did a controlled study and found that once again the two techniques, glass or virtual, were interchangeable in the doctor’s ability to make a diagnosis.
Dr. Mooney: There are a couple more studies from Austria that looked at melanocytic lesions and inflammatory lesions. They found, in the inflammatory lesions being studied virtually, that if the clinical information was not available along with the images, the concordance with the original diagnosis was not as good, whereas participants did better with tumors even without clinical information. [pagebreak]
Dr. Van Voorhees: What did your study find out about how virtual images compare to photomicrographs?
Dr. Mooney: There was basically no significant difference observed in the diagnostic accuracy using virtual dermatopathology compared with photomicrographs which, to me, was a bit surprising considering that the photomicrographs were taken to point the physician in the direction of the pathology, whereas with the virtual slides you just had to find it yourself. In one instance, there was almost a significant difference where in the first eight cases the photomicrograph group was doing better.
Dr. Van Voorhees: It seems there’s a very big difference because a photomicrograph is a single picture of something you want someone to see whereas on a virtual image you can move around much as you can on a glass slide.
Dr Mooney: The difference wasn’t as large as you might expect. We usually had four photomicrographs per case, and these were taken at different magnifications. The photomicrographs were taken by the faculty of the self-assessment exam at the Nordic Congress of Dermatology in 2008. These were the photomicrographs that they took in order to review the cases and give the diagnoses in their discussion after the exam was over. So, these were directed at the pathology.
Dr. Van Voorhees: So you thought the photomicrographs should have an advantage and were surprised to see things were more equivalent.
Dr. Mooney: Yes, because with the whole mount you have to go looking, whereas with the photomicrograph they were directed toward the pathology. We had seven dermatologists, seven pathologists, 17 dermatopathologists, and two fellows. And we had 15 cases, with multiple-choice questions, two per case. Sometimes we’d interject a special stain or immunohistochemistry and then ask for the diagnosis again. Or we’d ask for a diagnosis and then ask a second question in relation to the disease or condition in that specific case. [pagebreak]
Dr. Van Voorhees: Were there any differences noted in the ability to interpret a slide using virtual microscopy for different pathologic diagnoses? Were there differences between the type of providers?
Dr. Mooney: Basically there was no difference. Everyone had a hard time with the tough cases regardless of the format. There were some tough lymphomas and inflammatory cases that people were missing, regardless of whether they were photomicrographs or virtual.
Dr. Van Voorhees: Where do you see the benefits of digital microscopy?
Dr. Hood: It will likely replace microscopes in medical school teaching. Possibly you’ll see it replacing microscopes in undergraduate education as well. Just on a financial basis, maintaining the equipment and the study materials is increasingly expensive and difficult. There will always be microscopes because you’ve got to teach people how to do some things, but for educational purposes, a virtual image is in my opinion better than glass quality. I think you’re going to see that happening in a lot of schools and, in certain circumstances, in residency programs.
The other advantage for an educational setting is that it allows you to have an incredible library of permanent images that don’t break, fade, or disappear. It will also be a great thing for distance learning. It’s now possible to share images from Norfolk, Va., with people anywhere in the country and even any place in the world that has the right sort of equipment. So it has potential for libraries, for teaching cases, for distance learning, and for medical school.
In clinical settings, you could very easily put virtual images on a server and have them available to clinicians. I believe the surgeons in particular would like to be able to review original material before doing a second surgery, and the convenience of being able to pull up the original biopsy digitally, as opposed to finding and sending the slide, would be a wonderful and very powerful tool. I predict that that’s going to happen more and more.
I’m currently doing some dermatopathology teaching in both Norfolk and Richmond. What we’re doing at Medical College of Virginia is that the interesting cases that the residents biopsy are being scanned, and they and I are able to look at them on a server and discuss them. This also has huge implications for telemedicine. [pagebreak]
Dr. Mooney: And in CME it’s very useful. In the self-assessment exam in this study, we presented participants with clinical data, a clinical photograph, and the whole mount as a digital slide and asked questions that they could answer online. The software is now such that you can add annotations, e.g., circles, arrows, add little text boxes as you’re studying. Digital microscopy is also useful for storing slides you don’t have to keep retrieving the glass slides, which may break, and there’s less cost involved because you don’t have to pay someone to retrieve them.
Dr. Van Voorhees: Do you think that the use of this technology has contributed to your ability to do studies that have a global focus? Has the world truly gotten smaller?
Dr. Mooney: Yes, it has, and it really was an aid. Once the slides were scanned and digitized, at the Congress in Iceland, I was able to set up the questions through the software company in Ireland, and then connect with Toni and the reviewers one in Switzerland who reviewed them. The EVMS CME office subsequently approved the exam for CME credit. People across the world were participating including our two collaborators, Werner Kempf and Gregor Jemec in Europe, and others were taking the exam in Australia, in various places in Europe, and in the U.S. So yes, it certainly has helped me to do this type of study. [pagebreak]
Dr. Van Voorhees: Should readers expect to see virtual microscopy more and more in their futures, particularly as they consider recertification?
Dr. Hood: The American Board of Pathology is currently converting its glass slides to virtual images to be used on its examinations, and is increasing the number of virtual images on the exams each year. And similarly, the American Board of Dermatology has been using virtual images on the in-training exam as long as five years ago; we’ve been gradually using virtual images on the certifying examination with the idea that there will be a mixture of glass slides and virtual images over time. The American Society of Dermatopathology is currently using virtual images in its online MOC self-assessment. It’s also going to be important for recertification and maintenance of certification.
Given that, I should mention that there is an online library of virtual images at www.virtualdermpath.com that is available to everybody and has hundreds of images; residents use it to study before the exams.
Dr. Mooney: Speaking of exams, the exam we used for this study is still available online if people want to take it and find out how they do. Visit www.nivdp.com/cme.htm to take it.