The National Health Service (NHS) clinic in the UK is using artificial intelligence for the early diagnosis of skin cancer, with doctors describing the technology as a \u201cgame-changing method\u201d in the diagnostic process. The dermatology team at St Luke\u2019s Hospital in Bradford introduced the new system in April to assess suspicious lesions, significantly reducing patients\u2019 waiting times and the number of unnecessary procedures.<\/p>\n
The system, called DERM (Deep Ensemble for the Recognition of Malignancy), was developed by Skin Analytics. According to the manufacturer, the artificial intelligence system has a 99.9% accuracy rate in ruling out melanoma.<\/p>\n
The procedure works as follows: medical staff take three photographs of a suspicious mole or lesion and upload them to the system, which analyzes the images within just a few minutes. If the lesion is benign, the patient receives the appropriate recommendation and leaves the clinic, while suspicious cases are immediately referred to a specialist. For safety, each image at Bradford Hospital is also reviewed by a doctor.<\/p>\n
Thanks to the new technology, the clinic\u2019s capacity has increased by one-third \u2014 whereas previously it could see 24 patients per shift, it can now see 32. At the same time, according to hospital data, the number of unnecessary biopsies has fallen by around 10%.<\/p>\n
Plastic surgeon Zakir Sharif says the technology has had a major impact on reducing waiting times and unnecessary interventions: \u201cThe system diagnoses non-cancerous lesions remarkably quickly, within just a few minutes, which has significantly reduced the workload of the department.\u201d<\/strong><\/p>\n The effectiveness of the technology is also illustrated by the case of 73-year-old Lawrence Patten, who visited the clinic because of a suspicious mole on his back. The patient has a history of multiple myeloma and cutaneous squamous cell carcinoma. The AI assessed the lesion as suspicious, but subsequent specialist examination showed that it was most likely benign. According to his treating physician, Nader Ghader, without the AI assessment, the doctor would initially have recommended a biopsy or prolonged monitoring. The patient himself says that the speed of the diagnostic process plays a major role in reducing the stress associated with the possibility of cancer.<\/p>\n According to Tom White, head of dermatology, the innovation allows doctors to focus on genuinely critical cases. Statistics show that Bradford Hospital receives around 5,000 referrals for suspected skin cancer each year, but malignancy is confirmed in only 8% of cases (around 400 patients).<\/p>\n The AI model has been recommended by the UK\u2019s National Institute for Health and Care Excellence (NICE) for use within the NHS over the next three years, during which clinical evidence will be collected. It is currently being used in 40 hospitals across the UK.<\/p>\n