Interview with Kevin Sandeman, Head of Department Region Skåne’s pathology laboratory in Malmö, Sweden
Skåne, a southern county of Sweden, recently completed one of the largest adoptions of digital pathology in the world. Four labs, 14 scanners, and an estimated 890,746 slides were used and scanned from May 2019 to July 2020. At the helm of this feat, pushing forward the digitization of these pathology labs along with several colleagues, is Kevin Sandeman. We interviewed him to hear about the process of going digital, what challenges were faced, and ultimately the benefits of adopting new technology.
How long did it take for the Skåne Region’s labs to implement digital pathology?
Kevin: The process of digitization of the units started in May 2019 and finished in digitizing all histopathology in November of that year. We initially had to update all the technical infrastructure, like graphics cards, monitors, everything. At some labs we even had to update their wireless and wired networks. We wanted to make sure all the IT equipment was in place.
What was the biggest challenge in digitizing the labs?
The biggest challenge turned out to be the state of mind of the pathologists, to convince them to go digital. Changing minds is difficult when it comes to making someone change their working methods and workflows completely.
How did you overcome this challenge?
The first goal was to convince everyone that going digital was the right thing to do. We set up a voluntary practice period for the pathologists in Skåne labs to do a trial run. They had the chance to use actual slides versus digitized ones. The pathologists then wrote their feedback on using both methods, documenting the challenges of the digital workflow.
From here we were then able to come up with a plan on how to make the transition to digital more seamless; an internal quality check had to be done. Meaning we had to figure out how to change and optimize certain practices to fit with a new digital workflow such as adjusting the tissue preparation phase to enhance the quality of the scanning.
Learning by doing, then taking on the feedback produced by this and working together to overcome any barriers produced by digitizing our workflows was an effective way to overcome the challenge of the pathologist’s perception of adopting new technology.
What advice would you give someone who is thinking of embarking on a similar journey?
I would advise them to involve the pathologists early on, in the development of the applications and the flow, so they have the possibility to influence the process from the beginning. Pathologists and technicians will then be more eager to adopt digital pathology and adapt their ways of working. This way as well they will be able to fully see the benefits of digital pathology.
What is the biggest benefit of digital pathology?
When you finally have digital pathology implemented, you will find a peace of mind in the lab, because no one is concerned anymore about the sorting and archiving of glasses which was a persistent stress for people before.
There is no more need to do a librarian style job. Now you have the slides instantly at your hands. There is peace for everybody, both technicians and the pathologists. On top of this you have improved ergonomics, more efficient training and collaboration as well as integrated diagnostics.
How do you see artificial intelligence (AI) fitting into the now digitized workflow?
We have been using some basic computer-based applications. Some even involve artificial intelligence; for example, an AI-based Ki-67 counter. The pathologists are really eager to use AI, especially when for example counting positivity in some types of tumors as they don’t like the uncertainty of their own assessment. Using AI as an aid to support their interpretation is very appealing. There is definitely an interest to adopt it further to make objective measurements.
Have you been using AI in your work?
As an independent research project, I have been collaborating with some researchers at the University of Helsinki on developing a Gleason scoring system for prostate cancer using Aiforia’s deep learning AI software. At the moment we are assessing our results and hoping to share some data soon.