Closed for comments This consultation ended on at Request commenting lead permission
More research is needed on using 2 artificial intelligence (AI) technologies in teledermatology services to assess and triage skin lesions in people within the urgent suspected skin cancer pathway, before they can be used in the NHS. The technologies are:
Deep Ensemble for Recognition of Malignancy (DERM)
Moleanalyzer pro.
Access to the technology should be through company, research, or non-core NHS funding, and clinical and financial risks should be appropriately managed.
More research is needed on:
how accurate AI technologies used in teledermatology services are at detecting cancer and non-cancer skin lesions compared with teledermatology services alone
how accurate AI technologies are at detecting non-cancer and cancer skin lesions in people with black or brown skin
the effect of using AI technologies in teledermatology services on the number of referrals for face-to-face dermatology appointments compared with teledermatology services alone
the proportion of lesions referred from primary care that would be eligible for assessment by automated AI technologies used in teledermatology services and by teledermatology services alone.
Research should compare use of the AI technologies with and without healthcare professional review.
The research 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.
Why the committee made these recommendations
Staff shortages in dermatology services and an increasing number of urgent referrals for suspected skin cancer are causing delays in diagnosing and treating all skin conditions in secondary care. AI technologies (DERM and Moleanalyzer pro) could be used within a teledermatology service (a secondary care service that uses digital images to remotely assess skin conditions) to identify non-cancer skin lesions that do not need a further face-to-face dermatologist assessment. DERM can be used alone (automated) or with a virtual review by the company's dermatologists (a second read). Moleanalyzer pro is designed to be used only with dermatologist review.
Some of the clinical evidence on using automated DERM in teledermatology services raises concerns about the risk of missed or delayed cancer diagnoses. Using DERM with a second read could reduce this risk, but it is uncertain if this approach would help increase staff capacity in dermatology services. It is unclear whether DERM (either automated use or with a second read) is more accurate at detecting non-cancer skin lesions than using teledermatology alone.
The clinical evidence on the benefits and safety of using Moleanalyzer pro in teledermatology services is uncertain. It is also uncertain if its use could lead to fewer face-to-face referrals and biopsies compared with using teledermatology alone. Studies on Moleanalyzer pro did not include non-melanoma skin cancers and were not based in the NHS, so it was unclear how a melanoma-only tool would be used in NHS practice. So, the potential benefits from adding it to an established teledermatology service are uncertain.
The clinical evidence suggests that automated DERM is likely to be diagnostically accurate in people with black or brown skin. But, the evidence for both AI technologies is mostly for skin lesions in people with white skin. The amount of data in people with black or brown skin remains small, at around 3% of all data collected, so more data is needed to be certain that these technologies do not incorrectly detect or miss skin cancer in people with black or brown skin.
More research is needed on DERM and Moleanalyzer pro to better understand their clinical and cost effectiveness and whether their use could benefit teledermatology services and lead to more timely care for people with all skin conditions.
How are you taking part in this consultation?
You will not be able to change how you comment later.
You must be signed in to answer questions