A ground-breaking project involving Addenbrooke's has been awarded £1.5 million to develop AI that can help check for breast cancer - part of the government's plan to accelerate cancer diagnosis using artificial intelligence.
The project is one of nine AI healthcare technologies to receive nearly £16 million in government funding.
At the end of the day it's the pathologist making the diagnosis, but the AI can help us do this more efficiently and be more accurate.
Dr Elena Provenzano, consultant histopathologist at CUH
The AI technology has been developed by the start-up company Ibex and uses an algorithm to analyse images of breast tissue extracts, helping pathologists detect cancer, so they can complete diagnoses more quickly.
Its high accuracy rate could also reduce the need for patients to repeat the biopsy process and free up more time for consultants.
Known as Galen Breast, it will be trialled at Cambridge University Hospitals NHS Foundation Trust (CUH) as well as Nottingham University Hospitals, North West Anglia NHS Foundation Trust, Betsi Cadwaladr University Health Board and University Hospitals Birmingham.
Researchers will analyse its findings on 10,000 patients and evaluate improvements in the quality of diagnosis, cost-effectiveness, and quicker turnaround times for patients.
Normally breast tissue is checked for cancer cells by a pathologist looking at a specially prepared slide under the microscope.
Turning these slides into digital images means the new AI algorithm can help identify cancerous cells and speed up the time it takes for a pathologist to make a diagnosis.
Dr Elena Provenzano is a consultant histopathologist and leading the AI research project at CUH. She said:
"The AI technology will screen the digital image of the breast tissue and highlight if it is likely to contain cancerous cells. This helps us prioritise cases and reduce delays in diagnosis, which is really important for patients. The AI can also order extra tests we may need to do.
"Trialling this technology is really important in showing the efficiency savings for the NHS. Hopefully once we can demonstrate that the AI algorithm can help us save time and improve the patient's pathway, it can then be rolled out across the NHS."
We see ourselves, and the UK as a whole, at the cutting edge of innovation in digital pathology and look forward to implementing the Ibex Galen Solutions here in Cambridge as part of our histopathology team’s digital transformation.
Dr Rebecca Brais, consultant histopathologist and clinical lead for digital pathology at CUH
Dr Rebecca Brais is a consultant histopathologist and clinical lead for digital pathology and major projects in diagnostics. She added:
"Our digital pathology team at CUH is involved in a number of other AI research projects, including being one of ten major NHS laboratories creating a hub containing digital images of patients’ cells. This project has also been awarded £2.5 million in this latest round of government funding."
“We are also very proud to be a founding partner of PathLAKE Plus, a vital collaboration between several NHS Trusts across England. The consortium was awarded a £13.5 million grant from the Offices for Life Sciences to scale up digitization NHS patients’ pathology to a population of over 17 million and deliver the benefits of ground breaking AI tools.
"As part of this project we will also be implementing Ibex Galen prostate this year, another AI based tool that will aid the diagnosis of prostate cancer."
Artificial Intelligence has the potential to speed up diagnoses and treatments and free up time for our doctors and nurses so they can focus on caring for patients.
Steve Barclay, Health and Social Care Secretary
The other award winners include AI systems which can help detect diagnose rare diseases, identify women at highest risk of premature birth and support the treatment of neurological conditions like dementia.
The AI and Health and Care Awards is a programme of the NHS AI Lab which was set up to accelerate the safe, ethical and effective adoption of AI in health and care.
The full list of the winners so far can be found here (opens in a new tab).