Optics and Tech

To follow up on my previous post, I want to try and imagine the pathology workplace of the future. When the frozen section was invented in the 1890s, doctors hardly could have imagined the workplace of today (2019) 130 years later. In the last century, many advances were made in the field of pathology, aided by the increased availability and means of communication and exchange of ideas (telephone, radio, television, and most recently, computers and the internet became widespread). A century before that, dyes originally invented for the textile industry were applied to biological specimens. About one to two centuries before that, the microscope was invented.

Now, I would like to discuss the possibilities and logistics regarding what I believe to be an eventual and inevitable transition from traditional microscope viewing of physical slides to what I shall loosely call 'digital pathology'.

First of all - I want to note that pathologists have a central role in shaping how we want the 'workplace of the future' to be.  

Secondly, I believe that this is a process which must evolve organically given new developments in technology which are happening on a daily basis in our time. 

Third, there are crucial mistakes to be avoided if digital pathology is to 'get off the ground' so to speak.  

Additionally, I do know that there are opponents (read, older pathologists) who often harbor mistrust of new technology (i.e. computers) because they are set in their ways. The pathology workplace of the future should be so foolproof (read, solid) that a late-career pathology could walk into this fictional futuristic workplace and be impressed and amazed at its capabilities, and excited about the future of pathology and the possibility of working in one of these places.

While radiologists get a bad reputation for being in 'the dark room' perhaps we as pathologists have something to learn from them, as both are highly visual specialties. By reducing the amount of bright light exposure, radiologists have attempted to create a workplace in which a certain type of eyestrain is reduced (but not due to squinting), the details of the surroundings are obscured and their visual attention is focused on what is being assessed. In a similar manner, pathologists desire (or should I say, require) an immersive experience when we are looking at tissue through our microscopes. This, I believe is why pathologists tell me, books cannot compare to looking under the scope. What is seen is much larger when looking under the microscope compared to a picture in a book, moreover there are fewer distractions. A similar phenomenon is when one holds his or her cell phone close to the face, in effect 'blowing-up' the screen.

The fairly recent invention of VR headsets could be a game-changer in this regard. Or take augmented reality implements (such as google glass). I envision something akin to wearable sunglasses, with digital slides projected on the lens, with built in eye movement tracking in order to navigate a slide. Additionally, there will be no keyboards or controls, a pathologist will use simple hand gestures/movements which are read by sensors in the 'room' which serve as 'shortcuts' to navigate back and forth between slides (or if preferable, voice commands). Alternatively, imagine just thinking about magnifying a certain region of a slide and it happening instantaneously -- this requires nascent technology which I will elaborate on further.

Other key considerations which radiologists have heartily adopted, and pathologists should also, have to do with ergonomics and equipment. For example, on my recent visit to the radiology department, I was impressed by their adjustable table heights and comfortable chairs. An simple adjustment many pathologists can make without any fancy equipment is to obtain an adjustable microscope stand (but one might need an adjustable eyepiece as well depending). Comfort improves concentration and makes already heavy work feel lighter somehow. Importantly, they have also invested in state-of-the-art viewing equipment. Even for teleradiologists, to read at home, they must have a monitor (or monitors) which meet a certain specification in terms of quality. This is analogous to microscopes, which can be very costly, but when the inevitable transition to 'digital pathology' occurs the 'viewing station' may become very different.

Another innovation which radiology has gotten down pat, is the use of a universal file type which enables the easy sharing of information between different hospitals/medical centers. For example, one can obtain the images which were acquired in one location, and easily transfer/read the very same image in a different location because of the very fact that the filetype is the same. A simplified example of this is extensions in computer programs. For example .doc means document and can be opened by any program which can read document files. In the same manner, pathologists should decide and insist on standardization of digital slide filetype which would enable universal sharing of such information (although I wonder and hope the companies commercializing slide scanning have already decided upon this).

The final point which I would like to make is with regard to the speed of formulation of reports, and in a very real sense has to do with communication, and this is again is where the imagination part comes in. It has been shown that the human brain thinks much faster than it speaks. Additionally, I think we can all agree that one in general speaks faster than one types (hence dictation), one types faster than one writes with the hand (hence typing and writing), and that one reads, writes, speaks, and thinks faster and with more accuracy the more one uses each faculty. Regarding the conveyance of information and the formulation of reports, I wonder if in the future pathologists will be able to think up/down a report (this may require the invention of a new word).

Returning to my point of elaboration, which I referred to earlier, research is currently being done on the conscious control of objects, whether real or virtual, using brain waves. This is also called many different names, however I will use the term which I find most fitting, a mind-machine interface. It could be imagined that in the future, mind-machine interfaces will become commonplace, and eventually applied to pathology workflows.

The pathology workplace of the future may require the use of a real life, state-of-the-art 'thinking cap' which includes a built-in viewing station and neural net/hat (similar to an EEG monitor) conjoined, which would allow a pathologist of the future to complete reports much faster than before. Old pathologists will tell you, once pathology transitioned from hand-written reports to typing and eventually dictating, how turn-around times (TAT) decreased and subsequently the expectation of clinicians for same-day/the-day-after results became commonplace. If thought to text conversion is accomplished within our lifetime and applied to pathology, I believe pathologists will become ever more efficient at their jobs. Not one, but all of these in combination, and with the incorporation of yet-to-be invented technologies, all are innovations which would revolutionize pathology as we know it today.

If one makes an another analogy to radiology, one could say radiation exposure is the main limitation in conducting scans of the human body; nevertheless, advances in imaging technology (MRI, low-radiation CT) have allowed less radiating (though more costly) scans to be acquired. Radiology differs from pathology, however, in that a patient is a potentially unlimited source of scans. Moreover, the demand for scans and imaging results has on-the-whole been increasing (part of this is due to greed, and the other, negligence on the part of radiologists to serve as gatekeepers of healthcare resources). Because of these practices, modern radiology is now the practice of normology, in which one sees much more normal than 'bad' as a radiologist. The bottleneck in pathology is that it as a practice requires precious human tissue. Because I do not see resections going away anytime soon, I anticipate pathologists in the future having more time for research and academic pursuits, mentorship, safety/infection control, hobbies or perhaps even an improved lifestyle which would be pretty nice.

A very important cause which I believe all pathologists (who care) must fight for, if these changes do occur, is recognition of our speciality as being the foundation of medicine and the bridge between basic scientific research and other specialties of medicine (see the tree of medicine). And in addition, pathologists must prevent of the creation of 'mills' in which pathologists are made to work tirelessly to enrich others, from outside or within medicine itself. Technology has the power to unite people, under a common cause/interests in order to enact change.

P. S. A disturbing scenario involves the monitoring (Orwell's 1984) or de-privatization of thoughts (a.k.a. mind reading). A very dark episode (no. 1) of the anime Kino no Tabi tells the story of a town in which people automatically know each others thoughts. Would brain-computer interface technology develop to the point in which this could be enabled? Food for thought.

Dermpath and AI

Hi everyone,

This weekend, I'm reading Practical Dermatopathology by Ronald P Rapini. It seems a very well organized book. There is an interesting analogy in the book. It says:

Pathologists may be subcategorized into home-run hitters and hedgers. The home-run hitters try to "force" a diagnosis, and give only one most likely diagnosis. They are either very, very correct, and look very smart, or else they strike out and miss the diagnosis completely. This can be dangerous. For example, they might diagnose a lesion as a definite Spitz nevus, which subsequently is found to be a melanoma when it metastasizes. Most Spitz nevi can be diagnosed with relative certainly, but there are always those difficult cases for which all the experts can have their opinions, using the best of criteria, but for which there remains an element of uncertainly. Simple histology has its limits in predicting biologic behavior. Hedger pathologists, by contrast, seldom make a specific diagnosis, and instead often give a long differential diagnostic list, even to the point of listing histologic possibilities that are ridiculous from a clinical standpoint. They rarely strike out, but they are sometimes not very helpful, and are not appreciated by clinicians. Wise pathologists avoid these two extremes.

This reminds me of a concept brought up by a pathology resident, the idea of spotters or 'instant diagnoses' of entities which are instantly recognizable, versus other types of cases that can be more complex. I suppose the home-run approach would work great with spotters, whereas hedging would be most applicable to more ambiguous situations. Another resident told me, 'the best pathologists consider differentials before honing in on a diagnosis'. Essentially, going from good to great requires finding a happy medium. "The art of pathology is to be dogmatic about the diagnosis as often as possible, while not being afraid to hedge and give a differential diagnosis when the diagnosis is uncertain."

Today, I want to discuss the idea of pathology as art, within medicine, involving subjectivity. The reason for this is because of the strange idea that technology, such as high-powered computers or robots, will someday replace pathologists.

People who say pathologists will disappear as a specialty under the threat of AI do not really understand pathology. When this topic inevitably comes up in clinic, and I try to explain why AI would not be a threat to pathologists, it seems the clinicians become disinterested in the conversation. Why this is the case, I am not really sure, since most of the time they were the ones who brought it up as a topic of discussion in the first place, in response to me telling them that I am interested in going into pathology.

Thus, I want to list my personal opinion regarding why I think AI will not threaten the livelihood of pathologists everywhere.

1) Big tumors and sampling - I watched a TEDx talk where the presenter showed machine learning in which the computer was fed images of biopsies, and was able to stratify them in terms of prognosis with some accuracy. There are many judgements which a pathologist must make regarding biopsies and large cases, which a machine is incapable of. A machine is capable of producing output based on the data given to it previously, however, would it be able to determine whether a biopsy is adequate (in other words, representative of the lesion) [absence of data], or when it comes to larger specimens, whether the tumor has been adequately sampled during grossing? For some reason, I find it hard to believe that a machine would be able to determine whether additional sections of a large tumor need to be put in. These are judgements which an experienced pathologist makes easily and instantly when assessing a case.

2) Ontology and context - a machine does not understand the true meaning of what it sees. Yes, a person (i.e. computer scientist) can assign different values to data within the program/algorithm. However, truly understanding disease processes is something I cannot see a machine doing now or in the future. For example, a pathologist notices that a prior biopsy or sample had changes similar to the one he/she is noticing today, and based on the patient's lab tests and chart, concludes that the patient has a particular disease. The pathologist is able come to this conclusion based on a compilation of disparate sources of information, observation, analysis, and medical knowledge.

This brings me to the next two points.

3) Stains and artifacts - A machine (read, computer) does not know what is plausible or implausible. Preparation artifacts occur on a daily basis. Yes, a machine could be taught to recognize these, but there may be misinterpretations because of the variable presentation of such artifacts, given that they may range to subtle to overt. This highlights the often undervalued role of the pathologist as serving in QC (quality control) of the histology lab where slides are prepared. Pathologists are the arbiter of what is allowable in the quality of histology slides that are made for diagnostic purposes, for only pathologists have the privilege of conferring diagnostic meaning to a slide. Moreover, there are many mimics in pathology. Would a machine be sophisticated enough to distinguish these. Lastly, stains which require the use of a polarized microscope/direct immunofluorescence (DIF), I see a machine struggling to interpret given that it is based on light pattern and intensity.

4) Communication - One of the jobs of a pathologist is to serve as a liaison between clinician and the laboratory given the clinical context, in instances calling the clinician to get more information or recommending the appropriate ancillary testing, or simply communicating results in tumor boards, etc. Appropriate communication is something machines still have not mastered, just note our communication frustration with Alexa, Siri, and Google Home and various other voice activated/dependent products on the market currently. Formulation of reports are also important methods of communicating to clinicians in a concise, accurate fashion our assessment of a particular case. I feel doubtful that a computer or machine would be able to fashion a report which serves to be a useful consult to a clinician.

From my experience in pathology thus far, I feel AI will be a helpful adjunct, which may increase the efficiency of existing pathologists. Where I do see AI playing a role, is in removing some of the tediousness of pathology, and introducing more standardization, for example in the interpretation of stains such as HER-2 IHC, where there is interobserver variability. Just like AI was slotted to make radiologists "extinct" years ago, when radiology was brought from the 'dark ages' of silver-stained films to digitalized imaging and reading rooms, pathology is similarly on the cusp of making the transition from physical microscopes and slides to completely digitized workstations. Just see how they do it in Belgium.

One of the revolutions in pathology which I think will happen in my lifetime is the advent of telepathology, in which pathologists are freed to live and work in whichever location that they choose while reading virtual slides for a laboratory based in another location. This may help alleviate trainees' concern in pursuing a specialization which may not be particularly in demand, or concerns about finding a suitable position in a desirable location. Overall, the advances of modern day (tech revolution) hold a lot of promise as a tool for education as well, and I am excited to see how far the next generation of pathologists take this.

Lung path

Hi guys,

Been going over lung pathology using Foundations series book and Sanjay mukhopadhyay's Youtube channel. I want to correlate with cytology, but not sure about resource to use besides DeMay. What would be really good would be an atlas with cytopath correlates. 

Lung is a deceptively "simple" organ. It looks simple on H&E, but now I know there is way more than meets the eye... Autopsies, I feel, really help in understanding the lung in disease given it is pretty rare to get a full lobectomy specimen on a living patient. In that sense, it is somewhat (but not completely) like the brain. i.e. full brain, you know, person = 100% dead; Whereas, lung-wise person ≈ dead, remember you need two. But yeah... It is nice seeing coronal gross slices of lung, in an Autopsy Atlas, for example. Even better to see multiple slides (but not so nice for the patient). It is kind of unfortunate that there aren't more lung pathologists, but I understand it because of the volume. 

I'm interested in how the discovery of PD-L1 will change the field of lung cancer / pathology. The lab at my home institution is trying to bring PD-L1 IHC (immunohistochemistry) in. Perhaps if there was a method that was faster (molecular anyone?) it might be worth a pretty penny. Essentially, since the technology is so new, I feel like there are many directions to take this... investigation into better IHC markers as surrogate or improved over PD-L1 perhaps, or perhaps even a new technique, or a way to standardize it, would be helpful... though I'm guessing pharma R&D are all over this right now. Just a hunch.

What are some new developments you all have noticed recently?

Side note: Was recommended this book (Diagnostic Pathology: Kidney Diseases, 2nd Edition) by Robert Colvin at MGH. Along with Silva's renal path and Heptinstall's. All seem like great books for learning renal path. It's hard to find these books, but I suggest interlibrary loan if available where you are... another resource is WorldCat, which is a worldwide catalogue for books in libraries around the world.

Books

Hi everyone,

Happy 2019!

Today I wanted to post about books.

For surgical pathology, my top 4 recs

Foundations in Diagnostic Pathology series
AFIP fascicles (Atlas of Tumor Pathology)
WHO Classification
AJCC 8th edition

It may be worth going on WorldCat or requesting via interlibrary loan if one wishes to read the physical copy. However, I don't think it is worth buying immediately as the copies are updated every few years and it is worth waiting until residency to buy and only when you need to (buying WHO when starting hemepath rotation, for example).

I was actually quite surprised to find that the library at my medical school had a few physical copies of the Foundations series in addition to the big DeMay cytology textbooks (also recommend). I think there is a smaller, condensed version (Breviary in Cytopathology).

I do feel it is good to get information from different sources, as pathology changes constantly. Staying up to date is a difficult, but necessary task.

It is very interesting as there are a few entities in pathology in which the name has changed recently to be more PC (politically correct) or less confusing and if you talk to a pathologist who has been around a while, they use the names interchangably or just the 'old' name. From the perspective of a young person entering pathology, I think it is better to use the 'new' name, to minimize confusion, however. It also shows that you are keeping up to date with new trends and developments.

What books do you recommend?

P.S. I hear books by Cambridge (publisher) for pathology are good, but can't get my hands on any copies.