by Harrison Dai, Medical Student, Eastern Virginia Medical School

 

I went into medical school knowing that I wanted to do pathology. Though I am trying to keep an open mind, I still feel that no other field is quite as intellectually stimulating or rapidly developing as pathology is. The more that I am progressing through my pre-clinical years of medical education, the more I realize that my future aspirations will entail being in the field of pathology. I have always appreciated how pathologists frequently utilize cutting edge research to better their diagnoses- whether it be looking for newly discovered protein signatures or genetic mutations that are unique to certain disease processes. As I am about to finish my first year of medical school, I find myself at a very exciting crossroads in pathology, especially with respect to digital pathology.

Prior to medical school, I had only ever seen pathologists make diagnoses based on microscopic examination of whatever they are seeing on glass slides. Albeit on occasion I saw clinical pathologists making diagnoses in the clinic, I had the notion that a vast majority of a pathologist’s work was based primarily on interpreting what was on a glass side- and sometimes clinical correlation if a case was particularly challenging. However, when I started medical school, I was pleasantly surprised to learn about the advances that pathology has been making within the digital realm. I distinctly remember how fascinated I was when virtually previewing a pathologist at my school digitally signing-out cases. More specifically, I was particularly shocked by how clear the digital slides were, even when viewing them on high-power. Having seen too many poor-quality histology images either on the internet or textbooks, I was not aware about how technological advancements in whole-slide imaging have made it possible to produce high enough quality slide images that pathologists can diagnosis from.

 

Coming from a medical student’s perspective, I am awestruck by how far digital pathology has come. It is hard to imagine that is has only been about 20 years since the first commercial whole-slide scanner was introduced. I’ve heard numerous pathologists tell me that, at the time, the concept of digitizing glass slides was virtually unheard of- even scoffed at. Since then, digital slides can almost be found ubiquitously spanning from diagnostic settings, tumor boards to medical education. In my vicinity, it’s hard not to see pathologists utilizing some aspect of digital pathology in their day-to-day routines whether it involved teaching or consulting. It’s also exciting to hear about how artificial intelligence and machine learning are currently being introduced into digital pathology in efforts to improve diagnostic capabilities. Though I am barely a medical student, I find digital pathology playing an ever-increasing role in my medical education. I have lost count of how many times that I have relied on using whole slide imaging repositories to learn about pathology and microscopic features of diseases. No offense to some of my professors, I even sometimes found that learning histology was much easier when looking at digital images of slides rather than listening to lectures describing low-quality images of cells.

 

Looking ahead, I’m very curious to see what type of advancements and improvements are going to be made in the digital pathology realm. I’m especially interested in seeing how certain challenges pertaining to data management and public awareness are going to be addressed. Considering how a scanned image of a slide in one focal plane occupies at least 1 GB of data, there are legitimate questions about data storage and quality assurance, especially if one were to store hundreds if not thousands of digital slide specimens. Not only is it difficult to store such large amount of data, but there are also difficulties ensuring that the stored images of slides are exact replicas of their physical counterparts and contain the exact histological features of said physical slide. On a more personal note, I would also argue that digital pathology has somewhat of an advertising problem. Although digital pathology maybe something that most pathologists or residents know of, I don’t think anyone outside of said sphere is aware of such thing. Anecdotally speaking, even as someone who wants to pursue pathology as a career, I happened to come across digital pathology by chance. Had I not shadowed a pathologist at my institution who happened to sign-out cases digitally, I probably would not have come across the field of digital pathology until much later in my medical education. Despite some of the current challenges that digital pathology currently faces, I remain deeply optimistic about what the field, and pathology in general, will look like when I am done with my medical training. It is my sincere hope that one day digital pathology will be widely adopted and be utilized as part of common practice. Though it is very likely that digital pathology will be very different by the time I hopefully become a pathologist, I am glad to be part of and experience such an interesting time of innovation and advancement within the field.

 

Disclaimer: In seeking to foster discourse on a wide array of ideas, the Digital Pathology Association believes that it is important to share a range of prominent industry viewpoints. This article does not necessarily express the viewpoints of the DPA, however we view this as a valuable point with which to facilitate discussion.