Global first as NHS hospital uses AI for instant skin cancer checks

An NHS hospital is pioneering the use of artificial intelligence to diagnose skin cancer, enabling patients to get lifesaving checks for the disease without seeing a doctor.

In a global first marking a “new era” of cancer care, staff at Chelsea and Westminster Hospital are now using an iPhone with a magnifying lens to take photos of any suspicious-looking moles, with the image analysed in seconds by an AI app.

Nearly half of patients get the all clear and are automatically discharged with no further appointments, while those with cancer or who need further investigation are booked in to see a specialist doctor and begin treatment.

Thousands of NHS patients have had urgent cancer checks using the AI tool, freeing up doctors to focus on the most serious cases and helping to bring down waiting lists.

The system conducts the checks in five minutes, compared with around 20 minutes required for face-to-face examinations with a consultant dermatologist.

Trials show the device called Derm, developed by UK firm Skin Analytics, is 99.9 per cent accurate at ruling out melanoma — the most serious type of skin cancer.

It is the first AI technology in the world to be approved to make clinical decisions on cancer treatment autonomously, without requiring a doctor to double-check its findings.

In December, the Chelsea and Westminster Hospital NHS Foundation Trust took the “groundbreaking” decision to start relying solely on the tool as standard practice for urgent skin cancer checks. More than 20 patients a day are seen by a team of specialist medical photographers, who use the AI system to provide a quick assessment, without requiring dermatologists to review the images.

The tool has also been rolled out at 20 other NHS hospitals to assist doctors in their decision making, and in total it has helped to detect 13,000 cases of cancer in the UK.

Half a million people are referred for urgent skin cancer checks each year on the NHS — more than for any other type of cancer — and hospitals are struggling to manage soaring demand.

Dr Lucy Thomas, a consultant who leads the teledermatology service at Chelsea and Westminster, said: “It’s about freeing up dermatologists to spend time with the patients most in need. If patients who are coming in have just got a harmless freckle, mole or wart, they don’t need to see a dermatologist to be told that.”

The hospital gets 7,000 urgent skin cancer referrals each year, and using AI means these patients get the all-clear much quicker, helping to “reduce unnecessary anxiety”.

Thomas said: “Only 5 per cent of referrals actually turn out to be an urgent cancer. Patients are happy to be seen more quickly and assessed by an AI than wait weeks to be seen by a doctor.”

Thomas said that the use of AI in skin cancer can serve as a blueprint for the rest of the NHS, helping Wes Streeting, the health secretary, to deliver on his key pledge of embracing technology to bring down waiting lists. “This is a solution which could be immediately replicated across the UK to benefit patients on all pathways.”

Thomas said that, eventually, AI diagnostic tools can be used by GPs or directly by patients, to rule out cancers — helping to free up hospital capacity and manage waiting lists.

Dr Eirini Merika, a consultant dermatologist, said: “The use of AI in this setting has revolutionised the way we deal with skin cancer and patients as a whole. In the last few years, we have seen an exponential growth in suspected skin cancer referrals, the use of technology has helped us deal with demand.”

Merika added that “no technology, and no human eye, is completely foolproof”, adding that it is important the clinical expertise of doctors is combined with AI. “This is the start of a new era. AI is here, and it’s here to stay, and I think it’s a matter of finding the best way to combine it early into clinical practice.” The NHS Trust has been in partnership with Skin Analytics since 2022, and the project got off the ground with support of CW Innovation, an innovation programme run jointly by the trust and its official charity CW+.

‘AI found my cancer’

Jane Todman, 54, was diagnosed with melanoma after being assessed using the AI technology. Her GP had referred her to the dermatology team at Chelsea and Westminster in November, after she noticed a change in a mole on the back of her thigh.

Todman, from Putney, said: “The process was seamless — I had an appointment with medical photography and, before I knew it, was called in to have it removed. It was a total shock when I was told it was melanoma, and I struggled to take in just how lucky I had been to have it caught before it spread. The diagnosis made me hyper-aware of every mole on my body and the importance of getting it checked as soon as possible when you have a concern. The care I received was truly outstanding, and within two weeks I had it removed.”

Susan Smith, 78, was delighted to be “in and out” of her appointment within 15 minutes when The Times visited the AI dermatology service at Chelsea and Westminster last week. Smith, who was having a sore on her leg assessed, said she is “all for the use of artificial intelligence” to help reduce waiting times.

Click here to explore this article's original source for more.