Use of artificial intelligence (or computer-assisted technology)

EMRAD is helping to develop and test software which could make mammography (breast screening) services run more efficiently and address the clinical workforce shortages which exist in breast screening.

The consortium is working with Faculty AI and Kheiron Medical, two UK artificial intelligence companies, on their software products.

Faculty AI are focusing on the operational aspects of breast screening services to see whether improvements can be made on how clinic sessions are scheduled and booked.

Kheiron Medical have developed a CE-marked mammography intelligent assessment (MIA) tool which can help detect signs of breast cancer, running alongside the human-led process.

EMRAD is one of seven national NHS sites working with the companies on an 18-month project to explore the viability of their technical solutions.

A key driver for the projects is that an increasing number of women are accessing breast screening services every year but there are not enough staff across the East Midlands to universally support the high quality care expected. There are unfilled vacancies within radiology departments in the region with nearly one in four radiologists (imaging doctors) planning to retire in the next five years.

It is hoped that the ‘artificial intelligence’ (or ‘computer-assisted technology’) being developed will build capacity in NHS services and enhance the care that every woman attending breast screening receives. Their successful deployment could also help build confidence in the potential benefits of artificial intelligence for health services.

EMRAD is working with the Nottingham Breast Institute at the City Hospital Campus of Nottingham University Hospitals NHS Trust and the Breast Screening Services of United Lincolnshire Hospitals NHS Trust in progressing the initiative.

The process involves examining anonymised historic mammograms (breast x-rays) and clinical diagnosis results. This data is then combined with information on the efficiency of the breast screening service. All this information is then used to help develop the artificial intelligence programmes. The work is all based on historic data and does not involve current users of breast screening services in the region.

The project is due to run until April 2020. Depending on evaluation, it could be rolled out elsewhere across the NHS and developed into a full clinical trial.

Funding for this is via Innovate UK, a non-departmental public body responsible for supporting innovation.