AI software to detect skin cancer earlier and cut waiting times for referrals is being rolled out at an NHS trust in West Yorkshire for the first time.
Bradford Teaching Hospitals Foundation Trust has adopted DERM (Deep Ensemble for Recognition of Malignancy) technology at St Luke’s Hospital to rapidly analyse and identify the skin lesions and moles which are most likely to be cancerous.
This comes as only 6% of urgent skin cancer referrals from primary care across the UK, and 8% of those referred to Bradford Teaching Hospitals Trust, tend to be confirmed as malignant cancer cases, contributing to long waits for those with serious cases.
Consultant Plastic Surgeon and Clinical Lead for skin cancer, Zakir Shariff, called the technology “the future of skin cancer diagnosis in this country”.
He said: “Combining this cutting-edge AI will give us the capacity to pick up potentially serious skin lesions quicker and speedier than current processes”.
DERM, which was developed by Skin Analytics and has been adopted as part of a three-year scheme, works by scanning images of lesions and triaging them, with benign cases directed to non-urgent pathways.
It will be used in St Luke’s Hospital's new thrice-weekly teledermatology service, which operates on the basis of a referral from a GP. If a lesion is identified as suspicious by the technology, the patient will be immediately directed to the consultant dermatologist based next door for an excision, which will be sent to a laboratory for further analysis.
Those with skin conditions deemed benign would be reassured and discharged.
Tom White, General Manager for the Musculoskeletal and Therapies Clinical Support Unit, said that the service may be rolled out in community and GP surgeries, meaning that patients would not be required to handle the potentially stressful experience of a hospital visit.
He added: “For many patients, weeks of biopsies and awaiting results can be avoided as the process will reduce prolonged uncertainty and help avoid more invasive surgery.”
White also emphasised that the AI uses a “risk-based logic, designed to ensure that lesions are routed to the safest, most appropriate management outcome”.
The data relating to Bradford’s use of the technology will be scrutinised by the Cancer Alliance for safeguarding and proof of efficacity during its first year of deployment.
DERM is currently being used by 25 other NHS trusts in the UK, and according to Skin Analytics Performance Reports, it is 99.7% accurate at ruling out skin cancers.