Digital pathology can improve efficiency and enable AI integration, revolutionising cancer diagnostics. However, investment in infrastructure is needed for widespread adoption.
Digital pathology has the potential to improve patient care and support the pathology workforce by making the diagnosis and monitoring of disease much more efficient. Embedding digital pathology lays the foundation for the adoption of artificial intelligence (AI) in diagnostic services.1
What is digital pathology?
Usual practice involves pathologists looking at a tissue removed during a biopsy on a glass slide through a microscope to check for disease. The move to create digital pathology, involves the slides being scanned to generate digital images that can then be viewed by pathologists on a computer screen.
Digital pathology and efficiency
This digitisation provides flexibility and agility, which makes diagnostic workflows more efficient.
Cases are rapidly transferred between organisations and across pathology networks; access to expert pathologist second opinion is sped up, improving turnaround times and diagnostic pathways and ultimately benefiting patients.
Indeed, digital pathology has the potential to revolutionise the way pathology services are delivered, particularly in cancer diagnostics where the volume and complexity of samples continue to increase.
Access to expert pathologist second
opinion is sped up, improving turnaround
times and diagnostic pathways and
ultimately benefiting patients.
Potential for innovation in pathology
The digital images generated have been used to train algorithms to detect cancer and grade it. Early evaluations of AI for prostate cancer detection are being undertaken in several sites in the UK and the USA.
These evaluations are ongoing to establish where and how AI can help the diagnostic process; for example, by helping to triage cases requiring rapid assessment, ordering additional tests to save time and helping the pathologists evaluate the tissue.
Investment required for digitisation
However, investment is needed in the form of scanners so these glass slides can be digitised. There also needs to be investment in IT to ensure systems can support this digitisation and integrate other vital information that will assist pathologists in making a diagnosis, such as a patient’s clinical history. Only with the creation of this modern digital environment will AI be successfully implemented — and its benefits realised.
[1] The Royal College of Pathologists. RCPath Artificial Intelligence position statement (2023). Digital pathology (rcpath.org)
