London’s Chelsea and Westminster Hospital is employing a smartphone app and AI to identify patients with skin cancer without requiring appointments with the specialist clinical dermatology team. The app allows clinicians to prioritise patients with potentially harmful skin lesions. Patients with harmless lesions can be discharged far faster, reducing anxiety and providing faster reassurance.

Staff are using smartphones equipped with special attachments to take images for submission to the AI system. In addition to smartphones, staff are also equipped with digital SLR cameras with special lens attachments for very high resolution images where needed.

The AI-driven teledermatology service uses dermoscopic imaging to identify cancerous (malignant), pre-cancerous, and benign moles and lesions on the skin. The AI system, called DERM, developed by Skin Analytics, works by analysing high-quality, magnified images taken with a specialised camera attachment.

DERM is the only AI medical device approved in the UK to make clinical decisions in cancer care. It performs at a high standard, with 97% effectiveness for detecting skin cancer and 99.96% for ruling out melanoma. This is comparable to dermatologist performance.

Using AI-driven dermatology autonomously to accurately identify and discharge patients with benign skin lesions aims to free up over 30% of suspected skin cancer appointments on the urgent cancer pathway at the hospital. This will create more availability for patients with suspected skin cancer as well as those on waiting lists for chronic inflammatory skin diseases such as eczema and psoriasis, leading to better outcomes for all patients.

Since inception, the rollout of the hospital’s teledermatology service has been supported by the CW Innovation programme, which is led jointly by Chelsea and Westminster Hospital NHS Foundation Trust and its official charity, CW+. The CW Innovation programme aims to identify ideas and innovations that support the everyday care of patients.