Autonomous AI delivered a 62% capacity boost in two UK dermatology services, but six cancers slipped through the net.
Autonomous AI could free dermatologists to deliver thousands more face-to-face appointments by filtering benign lesions out of urgent skin cancer pathways, according to real-world UK data.
A study of 8391 patients across two hospitals found an AI-supported pathway saved an estimated 2851 hours of clinician time over 16 months compared with traditional face-to-face care.
That represented a 62% gain in clinical capacity and based on 20-minute consultations, enough time for more than 8500 additional face-to-face dermatology appointments.
While the autonomous AI cut clinician workload, six false-negative lesions underline the need for ongoing surveillance.
The findings will be presented today at the European Academy of Dermatology and Venereology (EADV) Congress 2026 being held in Vienna this week.
The researchers said the prospective deployment was the first large-scale real-world dataset examining autonomous AI within a cancer pathway.
The pathway used a CE-marked Class III AI medical device to analyse clinical and dermoscopic smartphone images of lesions referred through the urgent suspected skin cancer pathway.
Following an initial validation period, lesions classified as benign could be discharged autonomously without specialist review, while higher-risk lesions were sent to a teledermatologist.
The pathway managed 8391 patients, accounting for 94% of urgent suspected skin cancer referrals across the two hospitals, with 86% of patients consenting to autonomous decision-making.
After exclusions, the AI autonomously discharged 31% of patients at one hospital and 25% at the other without clinician review. Teledermatologists subsequently discharged another 24% and 25% respectively.
The proportion of patients requiring routine follow-up fell from 27% with standard teledermatology to 12% under the autonomous pathway.
Biopsy rates were also lower, at 27% compared with 43% under conventional face-to-face care.
Lead author Dr Lucy Thomas, a consultant dermatologist at Chelsea & Westminster Hospital NHS Foundation Trust and honorary clinical lecturer at Imperial College London, said the potential benefit was in redirecting specialist time rather than replacing dermatologists.
“We believe the greatest value of autonomous AI lies not in the technology itself, but in the specialist capacity it unlocks,” she said.
“Every hour saved reviewing low-risk lesions can be reinvested in patients with skin cancer, helping them access timely treatment to improve prognosis, and in patients with severe inflammatory skin disease, where earlier access to specialist care and effective treatments can transform quality of life.”
Safety monitoring identified six false-negative cases that had been discharged from the pathway, including five basal cell carcinomas and one melanoma in situ.
The cases were detected through post-market surveillance, and researchers reported no adverse outcomes during the available follow-up.
In a national dataset that included the two study sites, sensitivity was greater than 98% for invasive melanoma, squamous cell carcinoma, and basal cell carcinoma, while specificity was 72.1%.
Dr Thomas said the findings reinforced the need for surveillance to continue after an AI system was introduced into clinical practice.
“One of the key lessons for us is that deploying an AI system safely isn’t a one-off exercise,” she said.
“You need to keep monitoring it, understand when things go wrong, learn from those cases, and make sure patients themselves know what to look out for.”
Related
The findings come as UK dermatology services contend with rapidly increasing demand. Urgent suspected skin cancer referrals in England have almost tripled since 2009, while only about 6% result in an urgent skin cancer diagnosis, according to figures cited by the researchers.
Dr Thomas said further evidence across larger populations and different healthcare settings would be needed to establish how widely the approach could be applied.
“If these findings are replicated across larger populations and different healthcare settings, autonomous AI could become an important part of creating a more sustainable dermatology service – not by replacing dermatologists, but by allowing scarce specialist expertise to be focused where it can make the greatest difference to patients’ lives,” she said.
The EADV Congress is being held in Vienna, Austria, and online until 3 October 2026. See here for more information.
