AI03 Steps Towards Safely Deploying the World’s First Autonomous Artificial Intelligence as a Medical Device for Skin Cancer into a National Health Service Teledermatology Pathway
June 2025
in “
British Journal of Dermatology
”
Studysummary This study detailed the implementation of an autonomous AI device in an NHS skin cancer pathway, showing that it achieved a sensitivity of 97.3% for diagnosing skin cancers and exceeded sensitivity targets compared to specialists with a negative predictive value over 99.7%. Our plain-language summary of this paper — not a Tressless recommendation.
In December 2024, an autonomous artificial intelligence as a medical device (AIaMD) was successfully integrated into the NHS's urgent suspected skin cancer pathway, marking a significant advancement in teledermatology. This deployment followed a 2-year collaboration between a teledermatology service and an AI provider, utilizing a UK Conformity Assessed class IIa AIaMD. Since its implementation, the AI-enabled service has evaluated 10,737 cases, achieving a sensitivity of 97.3% for all skin cancers and a benign specificity of 74.0%. The AIaMD's performance was validated through real-world data from over 135,000 NHS patients, consistently meeting high sensitivity targets. The initiative also focused on reducing health inequities by routing patients with darker skin types to a consultant dermatologist for a second opinion. The transition to an autonomous AI pathway has increased capacity for urgent cancer and routine dermatology services, demonstrating the positive impact of AI integration in healthcare.
