The AI Medical Device Powering a New Era of Skin Cancer Care Across the NHS
Skin Analytics is a Business Reporter client
A unique technology assesses skin lesions in seconds and allows the NHS to see more patients, reducing cancer waiting times and allowing clinicians to spend more time with patients most at need.
AI-driven technology is revolutionising skin cancer diagnostics and making a major contribution to the NHS and its hard-pressed dermatology services.
Patients are routinely assessed by GPs but, because of limited dermatology training, too many cases are referred to specialists – and this at a time when 25 per cent of consultant dermatology posts are unfilled.
“GPs have an incredible workload and they don’t want to miss a diagnosis, so a huge number of referrals are made even though more than nine out of ten are benign,” says Dr Dan Mullarkey, a practising NHS GP and Skin Analytics Medical Director. “This means specialists are spending time looking at skin lesions that don’t need their expertise.”
Skin Analytics, founded in 2012 and an NHS AI in Health and Care Award winner, is in discussion with more trusts and private providers to roll out its DERM – Deep Ensemble for the Recognition of Malignancy – product across the healthcare system, with a goal to improve patient outcomes, liberate clinicians’ time and generate NHS budget savings.
“What the past few years has shown is that we can’t keep doing what we’re doing and hope to keep the standard of care we expect in the UK,” says Neil Daly, founder of Skin Analytics, which has just landed further funding to expand its services. “We’ve shown this technology can support existing dermatology pathways and have now started to build new pathways with our NHS partners that leverage the scalability of AI through community diagnostic centres. This is what really excites us.”
The technology created by the Skin Analytics team, a machine-learning algorithm, has been developed over a decade through high-level research, and is the UKs first and only UKCA Class IIa certified dermatology AI device. It powers a healthcare solution capable of recognising the most common benign and malignant lesions to help triage patients and build better, more efficient skin cancer pathways for patients, healthcare professionals and the wider NHS.
The algorithm has been trained to recognise the 11 most common lesions – including melanoma, squamous cell carcinoma and basal cell carcinoma – to provide accurate evaluations, and has now been used in AI pathways that have seen more than 33,000 patients.
DERM, which has being deployed successfully in eight healthcare settings, has proved its efficacy in supporting dermatologists by helping West Suffolk NHS Trust see more skin cancer referrals. Together, the partnership has increased the number of skin cancer referrals seen within two weeks from 20 to 95 per cent.
“The power of this technology is to take a large haystack and make it smaller for the dermatology teams to find the cancers,” says Daly. “Better yet, it is so efficient at that, we can use it at the start of the pathway to take pressure off primary care also.”
Healthcare professionals capture high-quality images using a dermatoscopic lens attached to a smartphone. One of these images is then assessed in seconds by the algorithm. The results, which are paired with a medical history questionnaire, are instantly available. Depending on the pathway, this outcome report can be sent onwards to the dermatology teams helping to prioritise those patients most at risk, or help or support a GP make a referral decision, thus helping to reduce the burden on the national dermatology teams.
Another partner, University Hospitals Leicester, saw 36 per cent of patients referred on the urgent skin cancer pathway discharged without needing to attend a hospital appointment after just three months of running the service.
“It’s been really exciting to see this technology being used. It speeds up the diagnostic process for suitable patients and helps to ensure they are only seen in our cancer diagnosis clinics if really necessary,” says Dr Pawan Randev, GP and clinical lead for cancer for the Leicester, Leicestershire and Rutland Integrated Care Board.
Skin Analytics has deep knowledge and experience of NHS clinical pathways and works with partners closely after launch to ensure DERM-enabled pathways solve local challenges.
The service has been operating in the NHS since 2020, when University Hospitals Birmingham deployed it during the pandemic. At this hospital, the UHB dermatology team has outperformed cancer waiting time targets for the past five quarters, despite seeing more than a 30 per cent increase in referral volumes over the past three years.
“From the very start, we have focused on both the ability to identify skin cancer and doing that in a way that is sustainable to the healthcare system,” adds Daly. “Our ambition is to help deal with current challenges in the system and to define new pathways that deliver huge long-term patient and economic value.
“The dramatic improvements DERM can achieve underline the potential we have to create a step-change for the better for the patient, physicians and the healthcare system. They also provide an opportunity to change the culture around skin cancer.
“The incidence of melanoma is increasing faster than any other form of cancer, and it is responsible for the majority of skin cancer deaths. Patients in whom melanoma is diagnosed at stage I have more than a 95 per cent chance of survival, compared with 8 to 25 per cent with a stage IV diagnosis, yet many people put off seeking help because they are worried about wasting the doctor’s time.”
“This device is so easy to use that we can build community diagnostic centres where people could drop in and get the reassurance they need, or to get onto the treatment pathway without delay. This is a golden opportunity for the UK to make a significant improvement in the outcomes of skin cancer and be a global leader.”
“We have been developing and refining this AI technology for ten years and our aim is to be deployed across the UK in community settings so that we can find every skin cancer at Stage 0 or Stage 1 to dramatically improve patient outcomes and reduce the burden on cancer care,” Mr Daly adds.
For further information, visit Skin Analytics at skin-analytics.com.
Originally published on Business Reporter