AI pathology project awarded £10.1m government funding

A digital pathology and artificial intelligence collaboration, with partners including Roche Diagnostics and Microsoft, has been awarded £10.1 million from a government-backed agency to expand across the north of England.

Diagnosing cancer is an area where artificial intelligence (AI) is becoming increasingly useful, aiding pathologists as they check biopsies for signs of cancer.

Late last year, a study published in the Journal of the American Medical Association (JAMA) showed that AI outperformed a panel of experts in a simulation where they were asked to diagnose breast cancer based on stained tissue samples.

Now a Leeds-based project will receive the investment from UK Research and Innovation, as part of the Industrial Strategy Challenge Fund, spearheaded by Greg Clark, secretary of state for business, energy and industrial strategy.

The public-private partnership is led by the University of Leeds and Leeds Teaching Hospitals and embraces a network of nine NHS hospitals, seven universities, Roche Diagnostics, Microsoft and a further eight medical technology companies, called the Northern Pathology Imaging Co-operative (NPIC).

The investment of £10.1m from UK Research and Innovation is boosted by an initial investment of £7m from industry, including a multi-million pound investment from Roche Diagnostics.

The consortium is now set to become a globally-leading centre for applying artificial intelligence (AI) research to cancer diagnosis, according to Roche Diagnostics.

The project also aims to develop more integrated ways of working across regional clinical pathology services.

Clinicians will work with industry and academic researchers to make new AI systems capable of analysing digital pathology images leading to better diagnoses for diseases like cancer.

The work aims to stimulate AI research locally in academic and business sectors, creating jobs and supporting economic growth across the Leeds city region.

Geoff Twist, managing director for Roche Diagnostics Limited said: “This is a true collaboration with partners from across healthcare delivery and industry, coming together with a common goal to improve diagnostic capabilities to offer cancer patients the best care possible. The ultimate aim is to develop an exemplar model that can be replicated across the UK for the benefit of cancer patients.”

Dr Yvette Oade, chief medical officer at Leeds Teaching Hospitals NHS Trust, said: “This is a really exciting step for patients because computers using artificial intelligence can be trained to recognise the patterns of disease. Machines will support clinically trained pathologists to diagnose cancer faster, better and at lower cost.”