Stronger regulation, better data, and training needed as AI diagnostics near mainstream use. 

A new report published today by the British Association of Dermatologists outlines the need for urgent action to harness the promise of diagnostic AI in dermatology. With the technology on the cusp of mainstream use in many NHS dermatology departments, the report calls for the government and regulators to make changes to keep patients safe, support the adoption of appropriate AI technology, and boost innovation in the field.

1 in 4 people in England and Wales see their GP about a dermatological condition every year. This results in about 3.5 million dermatology hospital outpatient or day surgery attendances annually. Demand for dermatology services is also increasing year-on-year. As of January 2025, the waiting list for dermatology services was 118% above its July 2020 level, one of the highest increases among all clinical specialties.

Our report calls for national policymakers to take decisive action to unlock the benefits of AI in dermatology. The UK must shape an environment that supports genuine innovation and healthy competition, underpinned by improved patient outcomes.

To achieve this, the BAD outlines five key priorities for policymakers:

• Establish a transparent regulatory framework to provide clarity for developers and build confidence among clinicians.
• Invest in digital infrastructure to ensure seamless data flow and integration across care pathways.
• Signal clear market needs to guide AI innovation towards practical, patient-centred solutions.
• Improve access to high-quality, diverse datasets so that AI tools are safe, effective, and equitable.
• Support workforce training and AI literacy to ensure clinicians and commissioners can safely evaluate and use emerging technologies.

One of the key challenges for the sector is the rising rate of urgent skin cancer referrals, as a result of the increased incidence of skin cancer, the most common type of cancer in the UK*.

Over the past 10 years, there has been a 170% increase in the number of people referred urgently for assessment by dermatology specialists. NHS England data shows that only 6% of these result in a confirmed diagnosis of skin cancer; the remaining 94% are either non-urgent or non-cancer cases. 

Developments in diagnostic AI technology have the potential to improve dermatology care, but their implementation must do so without exacerbating existing pressures or inequalities. 

Dr Tamara Griffiths, President of the British Association of Dermatologists, said: 

“A key theme of the report is the need to harness innovation. We are fortunate to have an innovative and dynamic AI industry in the UK, but it needs support and unambiguous signals from the NHS about the type of tools which are required. In return, the NHS needs a clear regulatory framework and appropriate clinical input to ensure it is adopting tools which are relevant and safe. Most importantly, we need to ensure that when AI is adopted, it improves the overall patient experience and outcomes in terms of safety, convenience, and waiting times. 

“Behind every urgent cancer referral is a person facing weeks of anxiety waiting for a consultation or a diagnosis. What’s more, every unnecessary referral means other patients may end up waiting longer to access care.” 

Dr Rubeta Matin, Chair of the British Association of Dermatologists’ AI Working Party Group, said:

“Diagnostic AI in dermatology is already here, and soon its usage will be widespread. It has huge promise across many disease areas in dermatology, but we are keen to see if it can safely triage and reduce the huge number of patients referred with a suspected skin cancer who don’t need to be seen. If it can, this would significantly ease capacity pressures and help ensure that those who do need care are seen more quickly.

“However, it is not inevitable that diagnostic AI in dermatology will improve current services. Decision makers in the government, the NHS, and regulatory bodies need to act decisively to make this happen. This report provides concrete recommendations for these decision makers across a range of policy areas, which we think would go a long way to improving the AI diagnostic landscape in dermatology and the wider NHS.”

-ENDS-

Notes to Editors

*There are 17,500 new cases of melanoma skin cancer every year. In addition, it is estimated that there are around 250,000 new cases of non-melanoma skin cancer (NSMC) every year. NMSC is excluded from official statistics due to how common it is. The BAD has partnered with the National Disease Registration Service to improve the quality of skin cancer data and encourage the inclusion of skin cancer statistics in national publications.

For more information, please contact the media team: comms@bad.org.uk, 07769000415.

Report Details

Title of report: Integrating Innovation: Charting the course for AI in dermatology

Link: here