BACKGROUND: In April 2022, Skin Analytics has achieved Class IIa UKCA Mark for the device DERM (Deep Ensemble for the Recognition of Malignancy) which is the UK’s first and only Class IIa certified artificial intelligence (AI) tool that identifies skin cancer by analysing dermoscopic images of skin lesions taken with smartphones with an attached dermoscopic lens. It is designed to accelerate patient diagnosis and relieve pressure on the healthcare system.

DERM is currently used in five NHS sites across the country and with their partners, have assessed more than 23,000 NHS patients, found 1,473 skin cancers, and avoided 4,713 dermatology appointments.

OBJECTIVES: We aim to carry out an economic evaluation to assess the clinical and economic impact of DERM assuming full deployment in the UK under alternative clinical and referral pathways.

METHODS: Different community and primary care pathways and one hospital pathway for initial use of DERM will be compared in terms of costs, resource us, and clinical effectiveness. The level of trust in the use or DERM by clinicians, who may overrule the DERM diagnosis, and economies of scale and scope from increased use will be considered. The statistical analysis will use a simulation model with input parameters from trust-level Hospital Episode Statistics and published data to measure impact, including from service reorganisations, using difference-in-differences and synthetic group statistical methods.

RESULTS: Preliminary results based on Montecarlo simulations project a decrease in unnecessary referrals of false positive melanoma results from 92% to 28% and cost-savings up to 47% compared to the current practice.

CONCLUSIONS: DERM is being used across the NHS and the certification is expected to unlock alternative pathways to diagnosis. This study informs on potential economic and clinical impact at national level assuming different route to diagnosis and acceptability or trust in the DERM device.