The NHS Launched the World’s Largest AI-Led Breast Cancer Screening Trial Amid Staffing Crisis
The UK National Health Service (NHS) launched the Early Detection using Information Technology in Health (EDITH) trial, aiming to screen over 700,000 women for breast cancer. The trial’s use of AI was implemented to address the UK’s radiologist shortages. Backed by £11m of governmental support from the National Institute for Health and Cancer Research, the trial aims to transform global healthcare and explore the broader use of AI in clinical diagnostics.
Radiologist Shortages
Since the COVID-19 pandemic, the NHS has been overburdened, facing an increasing number of patients and a declining number of clinicians. Radiology is one of the fields most severely affected. In 2023, the Royal College of Radiologists in the UK reported a 30% shortfall in available radiologists in the country, a number that is expected to rise to 40% by 2028 if actions are not taken promptly.
In 2023 the UK government responded by spending over £276m outsourcing, insourcing, and ad-hoc locums to manage the excess demand, a sum that could have paid for 2,690 consultants’ salaries. However, this investment did not yield much success, as numbers continue to rise; more patients are waiting for treatments, and clinicians are overexerted.
Unfortunately, this crisis extends beyond the UK. It was reported that over 80% of medical systems worldwide are facing radiologist shortages. The COVID-19 pandemic appears to have led to widespread physician burnout globally, prompting a significant percentage to retire early or transition to a different career after the pandemic. In 2021, a survey found that 21% of physicians planned to retire early because of COVID-19.
Can AI Replace Human Radiologists?
In the UK, there are currently around 600,000 reported cases of breast cancer, and this number is expected to rise to 1.2 million by 2030. It is now the fourth most common cause of cancer death in the UK and the leading cause of death for women under 50. This high prevalence not only puts a toll on patients and their families but also puts a toll on the economy. In 2024, breast cancer cost the UK economy £2.4 to £2.6 billion.
Breast cancer accounts for 55,000 UK cases yearly, but a 30% staffing shortfall delays diagnoses. Due to this AI is being actively explored in its use for screening analysis. Thus, this raises the question: can AI effectively replace radiologists?
Medical imaging is a field that AI fortunately excels at. AI tools, such as BCR-Net, Atraxis AI, and Transpara AI, all aim to analyse mammograms and detect breast cancer, potentially allowing for early detection. These tools have gained millions in funding from various investors and have shown great promise in pre-clinical studies.
The UK, as a nation, is keen to join in this endeavour and be a leader in this new age of AI, having invested hundreds of millions of pounds into AI research and development. One example of this endeavour is the EDITH trial.
However, the Royal College of Radiologists emphasizes that while AI has the potential to transform diagnostics and improve health outcomes, it must complement—not replace—investment in digital infrastructure, functional IT systems, and staff training to ensure safe and effective implementation.
“Artificial intelligence has the potential to transform the diagnostic landscape, improving health outcomes and shortening waiting lists. While the advent of AI in pathology is very exciting, and the NHS could be a world leader in the development and use of AI in pathology, a great deal of work is required to get to the point where AI is fully developed and used safely in the NHS. Investment in digital pathology, joined up functional IT systems, which facilitate information sharing across organisations, as well as training for pathologists to understand and use AI, will all need to be put in place.”
Dr Bernie Croal, President of the Royal College of Pathologists.
The EDITH Trial
The EDITH trial targets a critical bottleneck: whereby breast cancer incidence and radiologist shortages are expected to rise. The trial is planned to span four years, involving 30 screening centres around the UK.
Currently, the NHS requires double readings of each mammogram by two different radiologists. The EDITH trial aims to assess if AI can safely replace one of the two specialists needed to analyse one mammogram, a move that could free up over 100,000 clinician’s hours annually. Moreover, if successful, AI could help meet the NHS target of giving 80% of patients results within 28 days, up from 77% today.
Five competing AI systems will be tested across the 30 screening sites starting in April 2025. Early pilots suggest AI could detect subtle tumours humans might miss while reducing false alarms. A Nature Medicine study states that integrating AI into breast cancer screening programs can enhance the detection of interval cancers, identifying 20–40% of cases that human readers might overlook. Additionally, a systematic review published in Breast Cancer Research and Treatment highlighted that AI applications in mammography could lead to fewer false-positive findings.
Health Secretary Wes Streeting, The UK Secretary of State and Social Care, and a bladder cancer survivor himself states that:
“With record numbers of people diagnosed with cancer, and Lord Darzi finding that cancer survival is worse in this country than our peers, I know that urgent action is needed to save lives and improve patient care.”
Concluding Remarks
The EDITH trial represents a pivotal moment in the UK’s efforts to modernize breast cancer screening and address the growing radiologist shortage. While AI offers promising advancements in medical imaging, its role should be to enhance, not replace, human expertise.
The integration of AI into clinical practice must be done with rigorous oversight, ensuring accuracy, safety, and equitable access to care. As the NHS moves forward with this initiative, it is crucial to invest not only in AI development but also in the recruitment and training of healthcare professionals.
If implemented successfully, AI-assisted screening could significantly improve early detection rates, reduce diagnostic delays, and alleviate pressure on an overstretched workforce. This would ultimately save lives and reshape the future of cancer diagnostics in the UK and beyond.