Digital Pathology for Primary Diagnosis from an IT Manager's perspective
During this presentation, Mark will give an overview of his perspective on integration of new technologies in the laboratory setting.
He will focus on four key elements of Digital Pathology: Image acquisition, Image Storage, Image presentation and User Experience- highlighting where Pathlab are right now and then where to from here?
Webinar Transcription
Thank you for the opportunity to share part of our digital pathology journey with you. Hello, my name is Mark Flood. Currently, I work for Pathlab. Pathlab is a medical laboratory organization in New Zealand who employs 600 staff from the phlebotomists out in their collection rooms to the laboratory staff performing the testing.
I've been a part of the IT team here, there's six of us. I've been here for the last 10 years, and we're responsible for supporting those 600 staff with their IT needs, from managing user accounts and Active Directory to mouse replacement and pretty much everything in between. As time has progressed, I've developed a passion for histology, so I was invited to be part of the implementation of digital pathology into the pathologist's workflow, crafting it into a real weapon for the user against workflow pressures.
I've got a question for you. What's your biggest peeve? What is it that really gets up your nose? One thing that comes to me is time wasting. You know when you're stopped at a red light and there's absolutely nothing coming in either direction, but you still have to wait there until that light turns green?
Life is so precious. We need to make a choice to make the most of every moment and enjoy it. Enjoy every moment. What I want you to do right now is to get out your phone. Come on. And I want you to text someone close to you and tell them that you love them. That doesn’t mean text a person close as in a person sitting right beside you, because it could be a work colleague and that could get awkward. I mean close as in your spouse, maybe your parent if they're still around, or maybe your child, if you have them. Three simple words that have so much power. Let’s go do that now. I’ll give you a minute. In fact, I’m going to do it myself.
All right, if you've got your phone out, I need you to turn it on silent because all those people you just sent a message to will be texting you back with, “Are you okay?” Once you put your phone on silent, I’d like you to find a notepad app or a memo app on your phone. I don’t know about you, but I really struggle with my short-term memory, people’s names, that kind of thing. The way I used to get around this is I used to carry an A6 notebook and a pen in my pocket, and I’d write things down all the time. Now, I use the memo app on my phone. It allows me to electronically search the entire app for specific words and abbreviations.
Back to the topic, making the most of every moment. With this in mind, I’m going to try not to drag this talk out. I’m just going to quickly go through the nuts‑and‑bolts take‑home points. If you have any questions, quickly write them down or type them out in your newly discovered memo app, and hopefully we’ll have a discussion about them after. Sound okay? Great.
One other thing before I start: I like to encourage drinking water, so I’m going to play a little drinking game. Every time I say the word “massive,” I want you to take a sip of water. All right, let’s get into it.
Technology Integration
Over time, we as a human race have made massive advancements in technology, but not all of our ideas have been widely accepted. Why not? I think it comes down to successful integration. What is integration? Integration is the opposite of differentiation. Differentiation separates things apart. Integration brings them together in an unrestricted and equal association.
The introduction of new technologies happens all the time. We all know that. It’s the integration of that technology that matters. Which technology should we use? Well, for the most part, competitors will provide a similar service. But the real key to success is to be flexible, to be fluid‑like.
What’s this? It’s a rock. It’s a pretty good‑looking rock. It’s not massive. Now imagine this rock is your laboratory environment. Now we’ve also got some colored Play‑Doh representing our new idea. A lot of work has to go into molding the new idea around the laboratory environment. To allow for easier integration, we need to turn our rock into Play‑Doh.
If we can make our laboratory environment more flexible, yellow and blue make green. Yellow and blue create something new. It doesn’t really matter which technology you go with. The biggest impact on its integration is if you don’t change the rock into Play‑Doh.
There are two constants in life: time, which I’ve already referred to, and change. Without change, you might feel a sense of comfort and familiarity. Then change is introduced, resulting in a sense of loss. You must lose one thing in order to gain another. There may also be a grief process that people have to go through to deal with those losses. We’re all different, and because of that, some people won’t be able to make the transition to the new integration.
Change creates challenge. Challenge leads to growth. And the cycle begins again. If we’re not being challenged, we’re not growing.
Digital Pathology
I love this quote from our lead of specialty and histology, Corinne Hill: “If we don’t start now, we never will.”
You can spend your entire life waiting for the right moment, the right people, the right amount of finances to become available. Seems like a massive waste of time to me, and you already know how I feel about time wasted.
Don’t wait for the right people. Go out and find them. Don’t wait for the finances to come. Go out and find it. Make a choice, and start now.
Cancer
No one likes it, no one wants it, but if you’ve got it, you want to know about it as soon as possible. So turnaround times for diagnosis become paramount.
The online industry is booming. It’s massive. COVID‑19 has forced people to do things they’ve never done before, especially in an online capacity. During the most recent COVID‑19 lockdown here in New Zealand, we found ourselves in a position where we could support remote work for our pathologists, enabling them to work from home, review the images, and produce histological reports.
We’ve been playing in the digital pathology world for a while now. We have a pathologist working in Peru who’s been signing out our reports for the past four years.
Multidisciplinary meetings, or MDMs, consist of surgeons, radiologists, pathologists, and nurses all gathered in a room discussing complex cases. With video conferencing technologies, our pathologists can present digital pathology images from the comfort of their own home or office. No travel expenses, no time spent in traffic, and another vehicle off our roads.
That said, it does get complex for the pathologist. They must navigate reporting software, digital image software, video conferencing software, and network challenges. Sometimes it can all be a little bit too much. For the majority of them, they’re trained to diagnose pathology, not trained to use computers.
What I’m going to share with you now are what I believe to be the four key elements of digital pathology and how we here at Pathlab have set up shop: image acquisition, image storage, image presentation, and user experience.
Image Acquisition
We have two scanners, our original Aperio AT2 and an Aperio GT450 DX, and I believe we’ve just had the okay to purchase a second GT450.
In our experience, it takes, on average, 40 seconds to scan a slide over a one‑gigabyte Ethernet network, or four seconds over a 10‑gigabyte Ethernet network. To put that into perspective, if you’ve got 500 slides, it’s going to take 5.56 hours solid over a one‑gigabyte Ethernet network. That’s assuming an average of one gigabyte per slide. So you’ve got one gigabyte of data per slide, 500 slides, and it’s going to take 5.56 hours to get that image data up to your server.
Image storage continues to be a massive headache for us. It needs to be secure. It needs to be recoverable. It needs to be backed up. It’s critical that there’s no data loss or corruption when the storage hardware has component failures.
Local Versus Cloud Storage
Cloud storage is unlimited. It doesn’t matter what you throw at it, it’ll store it. It’s completely backed up with its own built‑in redundancy. It’s easy to share on a global scale, but uploading large amounts of data to and from the cloud takes time and can be quite expensive. Plus, how secure is your data when it’s in the cloud?
Local storage, on‑site storage, is cheaper. It’s fast. It’s easily transferable if you ever wanted or needed to change your hardware. But it’s difficult to share outside your organization. There’s no automated backup system, so you’ll have to create your own. The hardware needs to expand continuously as your data grows.
A 40‑times image averages about a gig of storage, as I’ve mentioned before. We’re scanning approximately 1,000 slides a day, five days a week. That equates to five terabytes per week.
One way to store the images is to use a RAID 6 array. A RAID 6 array allows for two disk failures before any data is lost. So in the current slide, you’ve got 12 drives there. Each drive has 12 terabytes of storage on it. Twelve twelves are 144 terabytes. So in a RAID 6 array, we’ve got 120 terabytes, with two drives redundant. This is equivalent to about six months of scanned slide storage, assuming we’re scanning 1,000 a day and each slide is approximately one gigabyte in size at 40 times.
The cost of these 12 drives is 10,000 New Zealand dollars, and that’s for the drives alone, let alone the server, the controller, the switching, the UPS, the cooling system, and the maintenance and servicing.
RAID arrays provide a backup of sort, but the drive rebuilds, once a disk fails, can take days. Requiring access to slide images reduces dramatically after the first month. The need for slide review diminishes over the life of the slide. At some point, the image is stored for regulatory purposes only.
Regulations are a little bit foggy at the moment. For glass slides, the required retention is 10 years for an adult and 25 years for a pediatric slide. Now, is the image required to be stored for that same period? Some would argue no. We can get another image from the slide. But others would say, “Well, the diagnosis was made from the image, so surely the image should be kept as well.” Then you could go on to argue that the diagnosis was made by a certain pathologist, so do we need to keep them for 25 years?
We archive after six months in an offline capacity. That’s what we’re doing. It’s still available, but it takes a while to retrieve. The management of these offline drives then needs to be established. How are they stored? Where are they stored? On‑site or off‑site? How are they rotated once the drive becomes full?
We have pathologists who are driving this forward. They see the benefits and are embracing the move and willing to invest in it.
Image Presentation
Slide management software within the eSlide Manager allows tools from the image to be read and presented as required by the navigational choices of the user. Whether it’s zooming in, repositioning the tissue orientation on the screen, or focus using Z‑stack, the software pieces together adjacent tiles to create the appearance of a seamless image as the user moves around the slide.
We need to ensure that the image displays correctly. Are the diagnostic features as easily identifiable in the image as they are from the glass slide?
Pathologist Workstation
We decided to buy a single PC setup for proof of concept. There it is at the top there. It was a high‑end PC with a massive graphics card and lots of RAM. We set up three monitors, including a 4K, and it worked.
The image below shows one of our pathologists working from home using a laptop and a 4K screen. Since then, we discovered that we probably only needed two monitors per workstation, and a 4K screen was probably overkill. 2K actually performs better in our experience.
The digital display is superior to the microscope in terms of field of view. In other words, you can see more of the slide on the screen than you can looking down the scope. However, more is not necessarily better. Zooming around a higher‑resolution image puts a massive amount of strain on the image processing software, and in our experience it actually behaves better at a lower resolution. There’s less of the image to display to the pathologist, so it can display it more rapidly.
Another thing to note: the PCs didn’t have enough USB ports once you take into account the mouse, keyboard, camera, microphone, speakers, and barcode scanner. Something to bear in mind when purchasing PCs.
In the slide shown, the 4K monitor is on the right‑hand side, the 2K monitor on the left‑hand side. Both are displaying a 40‑times image, and you can see the difference.
Every diagnostic workstation is now fitted with two screens. One is a 27‑inch, 8.3‑megapixel, 4K high‑definition screen with an IPS panel used as the main image viewer. The other screen displays the digital pathology reporting system and any other applications the pathologist might use, such as email or report forms.
On a 4K screen, the resolution is so high that the images and text become very small and difficult for the pathologist to see. This is another reason we tend to favor a 2K screen rather than a 4K screen.
Here’s another image of the viewing software. The slide tray is on the left‑hand side, thumbnails on the right, and the image itself in the middle.
User Experience
Historically, we’d scan the glass slide and the corresponding report would pop up. With digital pathology, we’re using the report to drive the image, so we’re changing the workflow. We need to ensure the image displayed corresponds with the report we have open, and that association needs to be fail‑safe. If there’s a loss of sub‑components of the system, there must be an ability to rebuild that association.
Integration with our reporting software works. The reporting software creates a URL that takes you to the corresponding case in the image‑viewing software. Each case may have multiple slides, and therefore multiple images, so the link should bring up the case rather than a single slide image.
In our system, the CEREBRO tracking system talks to the Delphic AP reporting system. It determines patient identification details and transfers the part, block, and stain data so the pathologist can see exactly where the process is at.
CEREBRO also produces a barcode label for the slides. The image software reads this barcode, stores it as metadata, and writes it back to CEREBRO. The image software then confirms the image has been scanned, and CEREBRO notifies the reporting system that the image is available. When the pathologist opens the case, the image automatically opens if available.
Using dual or triple screens created a new challenge. There is now massive mouse movement between screens for every specimen. This creates a new ergonomic burden that didn’t exist before. We’ve explored hotkeys, macros, foot pedals, Xbox controllers, and trackball mice, but nothing quite fits. What we really need is a mouse that operates like a microscope stage.
Drag and dictation helps significantly, allowing pathologists to navigate screens and applications by voice. Wireless mice, however, are a real headache. When the batteries die, work effectively stops.
What happens if you lose your network or your scanner? Despite best intentions, there are no guarantees. For that reason, we still keep our glass slides and microscopes as a backup.
The Future
Scanning speeds will increase. Image metadata will continue to grow. As metadata grows, data storage requirements will also grow, creating long‑term challenges around data recovery and presentation on future hardware.
AI is an exciting area. I don’t see it replacing pathologists, but supporting them. It’s like a personal assistant pointing out tumor margins or generating report templates, or automating tasks like mitotic figure counting to speed up diagnosis.
These tools rely on algorithms trained on image data. The more data we generate and store, the better and faster these technologies become.
Standards will need to be created. They’re not quite there yet, but they will need to happen.
In summary, if you’re serious about integrating digital pathology into your environment, you’ll need fast, reliable access to high‑quality, highly available, and secure data. Sounds simple, right?
Good luck. Massive thank you again for the opportunity to share our experiences with digital pathology. It’s been our journey. Your presence here tells me that your journey has already started, and it may be very different from ours.
I’ll leave you with this final quote from Benjamin Franklin:
“If time be of all things most precious, wasting time must be the greatest prodigality.”
About the presenter
Mark is the key IT Application Specialist for Histopathology at Pathlab Bay of Plenty and has been responsible for supporting Pathlab’s IT requirements for the past 10 years.
Before joining Pathlab IT team, Mark had broad medical laboratory experiences in scientific roles at Pathlab Bay of Plenty, Canterbury Health Laboratories, Christchurch and the New Zealand Blood Service.
Mark worked extensively with key stakeholders on the Digital Pathology integration project at Pathlab and was instrumental in putting together an implementation and maintenance IT framework plan ensuring the project’s success. Pathlab was the first Laboratory to go live with adopting digital pathology for Primary Diagnosis in New Zealand.
Mark is passionate about both IT and histopathology and is keen to help laboratories overcome the workload pressures with the adoption of automation and technology.
Related Content
Leica Biosystems content is subject to the Leica Biosystems website terms of use, available at: Legal Notice. The content, including webinars, training presentations and related materials is intended to provide general information regarding particular subjects of interest to health care professionals and is not intended to be, and should not be construed as, medical, regulatory or legal advice. The views and opinions expressed in any third-party content reflect the personal views and opinions of the speaker(s)/author(s) and do not necessarily represent or reflect the views or opinions of Leica Biosystems, its employees or agents. Any links contained in the content which provides access to third party resources or content is provided for convenience only.
For the use of any product, the applicable product documentation, including information guides, inserts and operation manuals should be consulted.
Copyright © 2026 Leica Biosystems division of Leica Microsystems, Inc. and its Leica Biosystems affiliates. All rights reserved. LEICA and the Leica Logo are registered trademarks of Leica Microsystems IR GmbH.