Staff at St Luke’s Hospital in Bradford say the introduction of an artificial‑intelligence system for skin‑cancer screening has markedly reduced waiting times and cut the number of unnecessary procedures. The AI, known as DERM – Deep Ensemble for the Recognition of Malignancy – was deployed in the clinic in April to evaluate suspicious skin lesions.
AI implementation
Under the new workflow, a healthcare assistant photographs a mole or lesion three times and uploads the images to DERM. The system analyses the pictures within minutes and returns a risk assessment. Every result is then double‑checked by a clinician before any further action is taken. The technology is supplied by Skin Analytics, which claims a 99.9% accuracy rate for ruling out melanoma.
Impact on patient care
According to the dermatology team, the rapid scanning capability has increased the number of patients that can be seen in a single session from 24 to 32. Consultant plastic surgeon Zakir Shariff described the change as a “big impact” on waiting times and on reducing unnecessary investigations. He added that the system quickly provides non‑cancerous diagnoses, allowing patients to be discharged with advice rather than undergoing further appointments.
The trust reports that unnecessary biopsies have fallen by roughly 10% since DERM was introduced. With around 5,000 referrals for suspected skin cancer each year in Bradford, only about 8% – roughly 400 patients – are ultimately diagnosed with a malignant tumour. Faster triage therefore helps clinicians focus on the most urgent cases.
Patient example
Seventy‑three‑year‑old Lawrence Patten, who has multiple myeloma and a prior squamous cell carcinoma, was urgently referred after his GP flagged a mole on his back. DERM marked the lesion as suspicious, prompting a specialist review. Dr Nader Ghaderi, an associate specialist in dermatology, noted that without the AI flag the patient might have undergone an immediate biopsy or a follow‑up appointment, even though the lesion appeared benign. Patten’s mole will be removed and tested, but he welcomed the swift assessment, saying the prompt attention eased his anxiety.
Dr Ghaderi affirmed that the AI correctly highlighted the case, allowing a more measured clinical decision. Shariff echoed this sentiment, stating that the technology has alleviated bottlenecks across the referral pathway, from initial waiting times to the reduction of needless investigations.
Wider adoption and future outlook
The National Institute for Health and Care Excellence (NICE) has recommended DERM for NHS use over the next three years while further evidence is gathered. The system is now active in 40 NHS hospital trusts, with Bradford Teaching Hospitals Trust being the first in Yorkshire to adopt it. Tom White, the trust’s general manager overseeing dermatology, called the early evidence a potential “game changer,” emphasizing that the AI could significantly lower urgent referrals requiring dermatologist review.
With the ability to process images quickly, double‑check results, and spare patients from unnecessary biopsies, the DERM AI platform appears to be reshaping skin‑cancer pathways in Bradford and may set a precedent for other NHS trusts seeking to streamline dermatology services.
Mitchell Landsberg is a Senior Technology Correspondent at News Raise. He covers consumer electronics, artificial intelligence, software developments, and digital privacy trends.
