New AI based tools are capable of generating diagnostic information or predicting health outcomes using digital images.

AI will not replace the need for pathologists to be involved in diagnosis – they will still be needed to interpret and analyse information from AI tools to produce an overall pathological assessment. However, advancements in AI are set to support the pathology workforce across several areas, improving diagnostic accuracy, prioritising cases and guiding treatment. This promises several benefits including alleviating pressure on pathology services, reducing costs and improving patient outcomes. 

AI-based tools must adhere to comprehensive guidelines and undergo rigorous evaluation before being adopted in clinical settings, but assessing the extent to which a new AI-based tool has been tested can be challenging. This has highlighted the need for transparency, openness and scrutiny of this emerging technology.

To help address this Dr Gillian Matthews and colleagues at the National Pathology Imaging Co-operative based at Leeds Teaching Hospitals NHS Trust, have created a public register for digital pathology AI tools, the first comprehensive, up-to-date and independent database where stakeholders can easiily access information about AI-based medical devices.

The open access register provides a central resource on AI-based digital pathology products and outlines product details, regulatory status, and clinical evidence of performance. 

The team identified and reviewed the 26 AI products with regulatory approval for image analysis, the majority of which focused on breast and prostate pathology. Of these, only half had publicly available evidence on clinical validation. These studies were typically funded or authored by the manufacturers of these products, and only 15% included testing on data from patients in the UK.

NPIC hope that the register will enhance accessibility and support informed decision making for pathologists and other stakeholders. By maintaining and expanding this resource, the team aim to support the informed and safe implementation of AI based tools in clinical practice, ultimately enhancing patient care and outcomes.

'There is growing interest in the use of AI-based tools in pathology, but locating information about commercial products can be challenging. We created this register to provide an accessible, centralised source of public information on AI products. We hope this will serve as a useful starting point for potential users to explore products, support informed decision-making within trusts, and drive improvements in transparency and evidence generation.'

Dr Gillian Matthews

Find out more 

You can find out more about the team's findings at Public evidence on AI products for digital pathology | npj Digital Medicine.

NPIC is a unique collaboration between NHS, academic and industry partners to:

• deploy digital pathology systems in hospitals across the country to improve pathology services and patient outcomes
• disseminate best practice in digital pathology
• develop and evaluate AI in pathology
• create a platform for nationwide research and innovation.