Whole slide imaging

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

xy313

New Member
7+ Year Member
Joined
May 16, 2014
Messages
10
Reaction score
1
The topic of digital pathology (I am using that to mean both whole-slide imaging and computer-assisted diagnosis) comes up sporadically on SDN, but it's been a long time and I think a new thread might be useful. I was curious what people's experiences with whole slide imaging systems were, and what you see as the future of digital pathology. I've met some attendings who are really enthusiastic about it, but what I hear on this forum sounds less so and I was surprised by that and interested in hearing peoples' thoughts/experiences.

It's a subject I find really interesting since long ago I had a strong interest in engineering and computer science.

I don't have experience with slide imagers myself...one of the hospitals I rotated at has one, but nobody uses it.

(Edited to add: I know there are a handful of pathology informatics fellowships...if you know anything about those it would be interesting to hear, too.)

Members don't see this ad.
 
Last edited:
I love WSI. I hope that it becomes the standard of care one day.

The problem with it is that, unlike radiology PACS and the DICOM format, there is no universal file format for WSI. Each company, be it GE, Aperio, or Leica, has a different file format. Nobody wants to take the million-dollar plunge and invest heavily into the hardware and software only for it not to be widely adopted.

The classification of WSI as a class 3 device also stalls the adoption of this interesting advancement. We as pathologists have really distanced ourselves from the consultative nature of what we do, instead preferring to be considered "the lab" - a faceless, automated enterprise where stuff goes in, and a machine spits out "a result" for the "clinician" to interpret. As such, "lab tests" are either accurate or inaccurate, and the ruling of the FDA regarding WSI is from such a perspective.

Computer-aided diagnosis is another tool that WSI allows for the use of. I can't recall the study, but I know that there are some patterns that the human eye cannot discern but CAD can, and this can have a significant impact in diagnosis. Will it replace pathologists? No. But it will certainly modernize the repertoire of our ancillary studies, of which IHC, a figuratively ancient technology, is the most recent.

Pathology informatics fellowships are useless. To me it is just academia trying to be erudite about something new - they have to turn everything into a "study" - rather than making true efforts to optimize the technology and get things running, such as consulting radiology vendors etc.

I've tried GE Omnyx, Leica and Aperio and I have to say that the Omnyx system was the most user-friendly, with better imaging quality and a strong consideration for making things efficient. Aperio is more widely adopted but inferior in all ways, and I wasn't much of a fan of Leica either.

A reason to not adopt WSI is the job market. It might make the field even more of a commodity (!). Telerads has a bit of an issue with this.

I don't think that there'll be any concern over foreign reads off domestic soil, due to legal ramifications. Plus, we already have plenty of foreign reads on domestic soil without WSI. Why worry about something that's already here?
 
  • Like
Reactions: 1 user
Thanks Over9000; I not only learned much from your response but enjoyed reading it. Your thoughts on the pathology informatics fellowships were helpful.

Regarding CAD not being able to do some things that the human eye can, I too have seen some papers along those lines. I once attended a talk where the speaker put up a photo of a house with an upside-down plane flying overhead and said that even a small child could pick out what was wrong with the picture, but a computer would have difficulty. The literature I am seeing suggests that CAD is pretty good for some small, well-defined clinical tasks in pathology.

It was very interesting to hear your perspective on the FDA's Class III decision. FDA does seem to have cleared (via Class II) some technologies related to WSI, though I haven't had a chance to see what yet.
 
Members don't see this ad :)
I think whole slide imaging will become a great tool for consults, tumor boards, perhaps research, and teaching. Daily diagnosis it is still a problem because the issue is in order to scan in the slide you still have to make the normal slide. And instead of putting it under a scope and taking a few seconds to evaluate it now you have to take the time to scan it then pull it up and evaluate it. Adds way too much time. It can be useful for offsite tasks and such (like reference labs who want an expert to see cases that are at a different location), but for routine diagnosis the efficiency is not there.

It might be helpful for "saving" slides, digitizing them instead of storing them. And the technology might change in the future so that you can do different types of evaluations (other than simple morphology) with scanned in slides.

So I think it is a work in progress but still for the near future only useful as a niche technique. You can foresee a type of consult practice where the consult center provides outlying labs/groups with a scanner so that they can provide more consults on cases. I am not sure how the billing would work on this though.
 
  • Like
Reactions: 1 user
The new fee schedule bundles of all TC services. Essentially the hospital gets an average of 100 per patient for all labs including pathology starting 2015. If your are administrator are you going to fork out the dough with no return on your investment.

We need new technology that makes thing cheaper not more expensive.
 
  • Like
Reactions: 1 users
Top