What is AIMS?
This is the Artificial Intelligence (AI) in Mammography Screening study. AIMS is an implementation study designed to appraise the benefits of a Deep Learning AI tool in breast cancer screening mammography. It is run though a clinical-academic-industrial partnership between Imperial College London, St George’s University Hospital London, OPTIMAM (OMI-DB) - Royal Surrey NHS Foundation Trust, and Google Health, having received funding from NIHR-NHSx-AAC in 2021.
What is the study about?
This research is evaluating how artificial intelligence (AI) can be used to improve the NHS Breast Screening Programme by helping to identify breast cancers in x-ray images (mammograms) of the breast.
About Part C of the study:
Our research has already shown that this technology is as good as an expert radiologist at identifying cancers in breast scans. Now, we want to see if we can use this technology in a real hospital. We will install the AI system at St George's and Imperial NHS Trust as part of a study to check if it works well in a hospital setting.
This study will not have any impact on patient care, and AI results will not be available to the attending clinical team. The study is primarily intended to test technical feasibility and identify workflow adaptations that will be required. We won't be testing how well the system performs in this study.
AIMS Study - Part C
This study is proposed to be an “opt-out” type study, where the AI system will run on all cases screened within specific study times. Women will be informed of the study through multiple avenues, such as information sheets before and during screening, screening centre posters, and videos, and will be given the opportunity to decline to participate.
Why research Artificial Intelligence (AI) for breast cancer screening?
Breast cancer is the leading cause of death in women over the age of 50 in the UK, with 1 in 8 women being diagnosed with the disease in their lifetime. Routine breast screening is therefore important for the early detection of breast cancer, where treatment is more likely to be successful.
Currently, NHS breast screening involves two radiologists assessing X-ray images of the breast (mammograms). If the radiologists disagree about whether cancer is likely present, the mammogram is then reviewed by a third expert.
Screening does not always find a cancer that is there. So some people with breast cancer can sometimes be missed. On the other hand, some women may receive a false positive result, which would mean they may have to undergo further treatment they would not need and may lead to anxiety.
There is a shortage of radiologists and immense pressure on staff in radiology departments to deal with the backlog of people waiting for scans. To help reduce the burden on resources and staff, new ways of screening need to be investigated.
Our previous research found that artificial intelligence can be used to identify cancer in mammograms with greater accuracy than radiologists. This highlights the potential for this system to reduce radiologist workload, improve accuracy of results and improve the rate of cancer detection, which this research is exploring.
How will we assess AI in NHS breast screening services?
We are using AI technology created by Google Health to evaluate how it can be used in NHS breast screening services to support clinicians. There are four stages to this research:
- Determine if the AI system in breast screening generates accurate results on NHS images.
We are comparing the AI system to radiologist screenings in large, diverse patient populations. This uses historical information from screenings to determine the accuracy and fairness of the AI system.
- Safely test the AI system in hospitals (without affecting current standard of care).
This will help us design a strategy to introduce the system in two hospitals within St George’s University Hospitals NHS Trust and Imperial College Healthcare NHS Trust and propose a framework for future use in NHS breast screening services.
-
Assess patient and public perceptions of an AI based mammographic screening tool to improve the quality, efficiency and experience of breast cancer screening. Workshop themes to range from exploring key issues surrounding patient consent in mammography studies, the ethical implications of the use of deidentified data for the purpose of the AIMS study and the design and role of printed and video media to communicate information to the wider public regarding our study.
- Examine how radiologists and clinicians interact with the AI system. Using mammograms from 100,000 women across two hospitals in the UK, we are observing how radiologist panels interact with the AI system when used in place of a second radiologist. This allows us to explore the factors that influence the radiologist’s clinical decision-making when reviewing the mammograms and making a diagnosis.
When is it taking place?
2022-2024
Who is included?
Women undergoing routine breast cancer screening (age 50-70) as part of the national breast screening programme, from January 2016 onwards.
Where is the study taking place?
Clinical centres in Imperial College Healthcare NHS Trust and St George’s Hospitals NHS Foundation Trust.
Our research partners
This research is a partnership between Imperial College London, Google Health, Imperial College Healthcare NHS Trust, St George’s Hospitals NHS Foundation Trust, and the Royal Surrey NHS Foundation Trust.
We’re involving patients in all our original research. We have recruited lay partners who are on our steering committee, and they will be involved in the decision-making, design, and dissemination.
Who is funding the study?
We are funded by the the AI Award which is one of the programmes that make up the NHS AI Lab, led by NHSX and delivered in partnership with the Accelerated Access Collaborative (AAC) and National Institute for Health Research (NIHR).
Read our news story to find out more.
Resources
Our research findings
International evaluation of an AI system for breast cancer screening - Nature 2020
Contact details
For more information about the AIMS study, please contact the Clinical Trials Manager – Aminata Sy on AIMStrial@imperial.ac.uk
