1 Recommendations

1.1

Deep Ensemble for Recognition of Malignancy (DERM, an artificial intelligence [AI] technology) can be used within teledermatology services in the NHS during the evidence generation period as an option to assess and triage skin lesions in adults referred to the urgent suspected skin cancer pathway. It can only be used:

  • if the evidence outlined in the evidence generation plan is being generated

  • once it has appropriate regulatory approval including NHS England's Digital Technology Assessment Criteria (DTAC) approval.

1.2

Mitigate the potential risk of missed or delayed cancer diagnoses when using DERM during the evidence generation period by:

  • doing a healthcare professional review for people with black or brown skin

  • regular monitoring of DERM's performance to maintain accuracy

  • using additional protocols when necessary, such as:

    • a national governance framework to ensure local oversight of use of DERM

    • a healthcare professional review.

1.3

The company must confirm that agreements are in place to generate the evidence. It should contact NICE annually to confirm that evidence is being generated and analysed as planned. NICE may revise or withdraw the guidance if these conditions are not met.

1.4

At the end of the evidence generation period (3 years), the company should submit the evidence to NICE in a format that can be used for decision making. NICE will review the evidence and assess if the technology can be routinely adopted in the NHS.

What evidence generation is needed

More evidence needs to be generated on:

  • how accurate DERM used in teledermatology services is at detecting cancer and non-cancer skin lesions, particularly in people with black or brown skin, compared with teledermatology services alone

  • the effect of using DERM in teledermatology services, compared with teledermatology services alone, on the number of:

    • skin lesions identified as benign and the proportion of these that are redirected to non-urgent dermatology pathways

    • referrals that result in face-to-face dermatology appointments

  • the proportion of skin lesions referred from primary care that would be eligible for assessment by DERM used in teledermatology services and by teledermatology services alone.

    The evidence generation plan gives further information on the prioritised evidence gaps and outcomes, ongoing studies and potential real-world data sources. It includes how the evidence gaps could be resolved through real-world evidence studies.

What this means in practice

DERM can be used as an option in the NHS during the evidence generation period (3 years) and paid for using core NHS funding. During this time, more evidence will be collected to address any uncertainties.

After this, NICE will review this guidance and the recommendations may change. Take this into account when negotiating the length of contracts and licence costs.

Why these recommendations were made

Teledermatology services are secondary care dermatology services that use digital images to remotely assess skin conditions. DERM could be used within a teledermatology service to identify and triage non-cancer skin lesions out of the urgent suspected skin cancer pathway.

Comparative evidence suggests that DERM may be able to identify a cancer lesion with similar accuracy to teledermatology or face-to-face dermatology assessment. Using automated DERM could identify more non-cancer lesions that do not need further review and so redirect more cases to non-urgent dermatology pathways compared with using teledermatology alone.

But, it is unclear whether DERM could free up capacity within dermatology services for diagnosis and care of non-cancer, non-urgent inflammatory skin conditions that need face-to-face assessment. Further evidence is needed to better understand the effect of using DERM on clinical capacity for both urgent and routine dermatology services.

The evidence on DERM is mostly for skin lesions in people with white skin, but a small amount of data suggests that automated DERM is also diagnostically accurate in people with black or brown skin. More evidence is needed to be certain that automated DERM does not incorrectly detect or miss skin cancer in people with black or brown skin.