Rates of melanoma and non-melanoma skin cancer have been rising significantly during the last three decades, and waiting times for assessment of a suspicious skin lesion are long in both the NHS and the private health sector in the UK.
Cancer Research UK reports that there are more than 16,000 new melanoma cases per annum in the UK and more than 150,000 non-melanoma cases, although it is widely acknowledged that the number of non-melanoma cases is an underestimate. Further, the incidence of melanoma in the UK is predicted to rise by 7% between 2014 and 2035.
The earliest signs of melanoma include change of size and colour of existing moles, while non-melanoma skin cancer often starts with a small red lump or a patch of skin that changes colour or texture. Identifying these early signs mean that cancer can be often removed with a simple operation under local anaesthetic, which significantly reduces the recovery time for the patient and decreases the risk of the patient needing long-term treatment such as radiation therapy.
However, these changes may only be noticed by the human eye once they have progressed to the point of needing further treatment. By developing artificial intelligence-based computer vision algorithms we can improve the early detection rate, which gives a better outcome for the patient, and means resources can be effectively managed for more serious cases.
One of our recent MSc Artificial Intelligence graduates, Mr Suraj Ghuwalewala, is developing this technology under the supervision of Dr Haider Raza, Dr Alba García Seco de Herrera, and Professor John Q Gan.
This project is being run in collaboration with private health organisation Check4Cancer.
This project is funded through the University of Essex business innovation voucher scheme.
Principal InvestigatorSchool of Computer Science and Electronic Engineering, University of Essex
Co-investigatorSchool of Computer Science and Electronic Engineering, University of Essex