There is great potential for the development of AI to support the diagnostic process in pathology, especially image analysis in histopathology. Investment in digital pathology systems with joined up IT systems and information sharing across organisations is vital to begin to understand the potential for AI-assisted diagnostics, and continued investment in digital pathology.

While the advent of AI in pathology is very exciting, and the NHS could be a world leader in the development and use of AI in pathology thanks to investments in this area to date, a great deal of work is required to get to the point where AI is fully developed and used safely in the NHS. As a result, it is very unlikely that AI will address the severe workforce gaps we see in the short term, but rather contribute to future developments in medical safety and efficiency in the medium term.

The purpose of this document is to outline the College’s approach to artificial intelligence (AI) in cellular pathology and to set out the case for further work in this area. The scope of this document is limited to the use of artificial intelligence in cellular pathology.

Key messages:

● Digital pathology and artificial intelligence (AI) are technologies which have the potential to transform the way pathologists work. The College supports the use of these technologies to improve healthcare.

● The use of AI in healthcare must be clinically led.

● AI will not replace the need for pathologists to be involved in diagnosis. They will still be needed to interpret and analyse information to produce an overall pathological assessment. However, AI does have the potential to free up time for pathologists, improve laboratory workflow or enhance the accuracy or consistency of diagnosis.

● For the NHS to start benefitting from AI there needs to be a significant investment in digital pathology and AI in the NHS – only a handful of UK trusts are using digital pathology currently – as well as an investment in training pathologists to understand and use AI.

● There will need to be educational resources to educate health professionals in data governance, ethics, appraisal and interpretation of AI and similar technologies.