Digital pathology is by no means a new practice; pathologists have had the ability to scan and digitise an entire pathology glass slide for over twenty years. The proffered advantages of digital pathology are many; allowing for a more efficient and collaborative diagnosis and yet hospitals haven’t made the switch from using microscopes and glass slides.
The Northern Pathology Imaging Co-operative (NPIC), led by the University of Leeds and Leeds Teaching Hospitals NHS Trust received an investment of £17.1m from UK Research and Innovation and involved industry partners to roll out a programme of digital pathology and artificial intelligence across the north of England.
Step one of the transformative programme is already underway; from September 2018 Leeds Teaching Hospital NHS Trust began to digitise all pathology, and over the course of the project all of the hospitals in the West Yorkshire Association of Acute Trusts will follow.
The benefits are clear; speeding up and improving collaboration between pathologists and researchers. Instead of sending pathology glass slides from one lab to another, a link to an image can be shared amongst teams, wherever they are based, with markers plotting areas to study like a longitude and latitude on online maps.
Changing these ways of working is no mean feat, without considering the practical changes for the workforce, the sheer quantity of data this will produce is huge. To put this in context, if you were to print out a full resolution digital pathology image it would be the size of a tennis court.
NPIC will generate an average of 760,000 images per year, about 1.2 petabytes of data. If one byte the size of a grain of rice, one petabyte is the Island of Manhattan covered in rice.
Geoff Hall, Professor of Digital Health and Cancer Medicine and Chief Clinical Information Officer for Leeds Teaching Hospitals is leading this area of work and how to securely store and process data at that scale.
Another key part of the project is to consider the ethics of data sharing to ensure NPIC partners abide by the highest professional standards when images are used for research purposes.
The work will also look at applying artificial intelligence and machine learning to digital pathology, transforming how cancer and other diseases are diagnosed. In the near future AI can be trained recognise patterns in pathology images, on a massive scale and potentially predict patient outcomes.
Professor Hall is also considering how to link the pathology images with patient clinical information to gain further insight into the disease trajectory.
“This is a huge opportunity for Yorkshire to lead in this new area and further enhance our position as a hub for medical technology. We can explore how to use digital pathology as part of precision medicine to ensure patients receive treatments tailored to their disease.” Professor Geoff Hall
Dr Darren Treanor, a Pathologist at the University of Leeds and Leeds Teaching Hospitals NHS Trust, is leading the project, said: “Digital pathology is a technology with a huge potential to improve healthcare.
“NPIC will allow us to use digital pathology to help patients across the region, and provide a platform on which we will develop artificial intelligence tools for pathology diagnosis to be used around the world.”
The consortium includes a network of nine NHS hospitals, seven universities and 10 industry-leading medical technology companies.