Can we use an artificial intelligence system to improve the quality and efficiency of breast cancer screening?
| ISRCTN | ISRCTN88754382 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN88754382 |
| IRAS number | 307842 |
| Secondary identifying numbers | IRAS 307842 |
- Submission date
- 15/06/2022
- Registration date
- 20/06/2022
- Last edited
- 20/05/2024
- Recruitment status
- No longer recruiting
- Overall study status
- Completed
- Condition category
- Other
Plain English summary of protocol
Background and study aims
1 in 8 women will be diagnosed with breast cancer during their lifetime. Breast screening aims to find cancers early when treatment is more successful. In the UK breast screening programme, two cancer specialists (radiologists and radiographers) review the x-ray images (mammograms) taken during your visit. They decide if the mammogram is normal or whether further imaging or investigation is required. Any disagreements between the two specialists are reviewed further by a third specialist.
This research aims to test the ability of a new computer system developed by Google that uses a technology called artificial intelligence (AI) to help detect potential signs of cancer in the breast images (mammograms) that are taken during your screening visit. We have already completed a lot of research where we have shown that this technology is as good as an expert radiologist at identifying cancers on these scans. We believe that this technology has the potential to improve accuracy, safety, and patient experience of breast screening in the UK, and make the process more affordable for the NHS. It may also have a role in supporting NHS screening services and clinicians directly. We now need to understand how it works in a real-world NHS setting.
Who can participate?
Women aged 50 to 70 years old, undergoing routine breast cancer screening as part of the national breast screening programme at participating centres.
What does the study involve?
We will evaluate the performance of the AI system at Imperial College Healthcare NHS Trust and St. George’s University Hospitals NHS Foundation Trust within the routine breast cancer screening in real time with no disruption or impact on screening appointment visits. This study only uses anonymised data to perform the study analysis. We will assess how the AI system can be integrated into the NHS screening pathway at Imperial College Healthcare NHS Trust and St. George’s University Hospitals NHS Foundation Trust. We will test how quickly the AI can read the images and return results to the clinical site. Findings and lessons learnt from this study will help to design a screening pathway that may include AI as one of the readers.
Taking part in this study will not require any additional time, scans, or procedures, and will not affect your routine clinical care in any way. However, we hope that it may help others in the future.
What are the possible benefits and risks of participating?
We do not anticipate any disadvantages or risks to taking part. We do not anticipate any immediate benefits of taking part in this study. However, the information we get from this study will help us assess if artificial intelligence has the potential to improve future clinical care in the UK breast screening programme and worldwide, by providing more accurate reads, improving breast cancer detection, and by reducing the time to provide results to patients.
Where is the study run from?
Imperial College London (UK)
When is the study starting and how long is it expected to run for?
May 2021 to August 2024
Who is funding the study?
National Institute for Health Research (NIHR) (UK)
Who is the main contact?
Clinical Trial Manager, aimstrial@imperial.ac.uk
Contact information
Principal Investigator
Institute of Global Health Innovation (IGHI) & Department of Surgery and Cancer
Imperial College London
10th Floor, Queen Elizabeth the Queen Mother Wing (QEQM)
St Mary's Campus
London
W2 1NY
United Kingdom
 ORCID ID |
0000-0001-7815-7989 |
|---|---|
| Phone | +44 (0)2033121310 |
| a.darzi@imperial.ac.uk |
Scientific
Institute of Global Health Innovation (IGHI) & Department of Surgery and Cancer
Imperial College London
10th Floor, Queen Elizabeth the Queen Mother Wing (QEQM)
St Mary's Campus
London1
W2 1NY
United Kingdom
| ORCID ID |
0000-0003-1668-0672 |
|---|---|
| h.ashrafian@imperial.ac.uk |
Study information
| Study design | Feasibility/pilot study |
|---|---|
| Primary study design | Observational |
| Secondary study design | Feasibility/pilot study |
| Study setting(s) | Hospital |
| Study type | Screening |
| Participant information sheet | https://www.imperial.ac.uk/aiscreening |
| Scientific title | Artificial Intelligence in Mammography Study (AIMS) Part C - Feasibility of an artificial intelligence system to improve the quality and efficiency of breast cancer screening |
| Study acronym | AIMS- Part C |
| Study objectives | A novel AI system for breast cancer screening to demonstrate read-only ‘silent’ integration of the AI system, in a non-interventional manner, at each participating site interfacing with the other technical systems used within the breast screening programme (NBSS and the Trust PACS), explicitly avoiding any potential for AI outputs to influence patient care. |
| Ethics approval(s) | Approved 18/11/2022, East Midlands - Nottingham 1 Research Ethics Committee (The Old Chapel, Royal Standard Place, Nottingham, NG1 6FS, UK; +44 (0)207 104 8115; Nottingham1.rec@hra.nhs.uk), ref: 22/EM/0198 |
| Health condition(s) or problem(s) studied | Decision support in breast cancer screening |
| Intervention | Data will be collected prospectively from the breast screening programme (NBSS) Mammography Image Database, with patient consent. There will be no impact on patient care. The intervention is the AI system, assessed on de-identified prospective breast screening cases and outcomes. To understand how to perform technical integration of an artificial intelligence (AI) system into the standard clinical workflow. |
| Intervention type | Other |
| Primary outcome measure | 1. Time taken for the AI system to return results from mammograph images over the study dataset time period 2. Analysis of number of failure cases ( such as such as model errors, software errors, integration errors, use errors, and hardware errors) for the study dataset time period. Accuracy will be measured as proportion of true results (both true positives and true negatives) among whole instances. Area under the receiver operating characteristic curve (ROC) will be measured for AI 3. Percentage of cases correctly excluded during eligibility checks and reasons do excursion during the study period |
| Secondary outcome measures | 1. Accuracy measures including AI recall rate measured as proportion of true results (both true positives and true negatives) 2. AI sensitivity and specificity with respect to arbitrated recall decisions (measured as the number of positive cases (cases considered positive if they received a biopsy-confirmed diagnosis of cancer within 3 months following the screening visit. Negative cases will require a negative result from the study screening visit) 3. AI sensitivity for biopsy-proven cancer u(true positive rate in percentage(%) derived by ROC analysis) 4. AI specificity for biopsy or diagnostic imaging-proven benign lesions (true negative rate in percentage (%) derived by ROC analysis) |
| Overall study start date | 20/05/2021 |
| Completion date | 29/08/2024 |
Eligibility
| Participant type(s) | Patient |
|---|---|
| Age group | Adult |
| Lower age limit | 50 Years |
| Upper age limit | 70 Years |
| Sex | Female |
| Target number of participants | Up to 14,000 participants |
| Total final enrolment | 10875 |
| Key inclusion criteria | 1. Women undergoing routine breast cancer screening (age 50–70), as part of the national breast screening programme at Imperial College Healthcare NHS Trust and St George’s University Hospital NHS Foundation Trust between the study dates. 2. Mammography images acquired using Hologic/Lorad, Siemens, or GE devices. |
| Key exclusion criteria | 1. Women that opt-out of this study 2. Women who have registered with the NHS national data opt-out |
| Date of first enrolment | 27/11/2023 |
| Date of final enrolment | 28/01/2024 |
Locations
Countries of recruitment
- England
- United Kingdom
Study participating centres
Teddington
TW11 0JL
United Kingdom
Surbiton
KT6 6EZ
United Kingdom
2 Edridge Road
Croydon
CR9 1PJ
United Kingdom
Sutton
SM1 2RJ
United Kingdom
London
SW15 5PN
United Kingdom
Purley
CR8 2YL
United Kingdom
London
W6 8RF
United Kingdom
London
W2 1NY
United Kingdom
Southall
UB1 3HW
United Kingdom
Hounslow
TW3 3LH
United Kingdom
Uxbridge
UB8 1UB
United Kingdom
Sponsor information
University/education
Exhibition Road
South Kensington
London
SW7 2BX
England
United Kingdom
| Phone | +44 (0)20 7594 9480 |
|---|---|
| rgit.ctimp.team@imperial.ac.uk | |
| Website | http://www.imperial.ac.uk/ |
"ROR" |
https://ror.org/041kmwe10 |
Funders
Funder type
Government
Government organisation / National government
- Alternative name(s)
- National Institute for Health Research, NIHR Research, NIHRresearch, NIHR - National Institute for Health Research, NIHR (The National Institute for Health and Care Research), NIHR
- Location
- United Kingdom
Results and Publications
| Intention to publish date | 21/03/2025 |
|---|---|
| Individual participant data (IPD) Intention to share | No |
| IPD sharing plan summary | Not expected to be made available |
| Publication and dissemination plan | Planned publication in a high-impact, peer-reviewed journal |
| IPD sharing plan | Not expected to be made available due to contractual agreements with study sites |
Editorial Notes
20/05/2024: The following changes were made to the trial record:
1. The overall end date was changed from 21/05/2024 to 29/08/2024.
2. The total final enrolment was added.
15/01/2024: The recruitment end date was changed from 22/01/2024 to 28/01/2024.
16/11/2023: The following changes were made to the trial record:
1. The recruitment start date was changed from 23/10/2023 to 27/11/2023.
2. The recruitment end date was changed from 23/01/2024 to 22/01/2024.
3. Ethics approval details added.
4. The scientific title was changed from 'Artificial Intelligence in Mammography Study (AIMS) Part C' to 'Artificial Intelligence in Mammography Study (AIMS) Part C - Feasibility of an artificial intelligence system to improve the quality and efficiency of breast cancer screening'.
11/09/2023: The following changes were made to the trial record:
1. The recruitment start date was changed from 01/09/2023 to 23/10/2023.
2. The recruitment end date was changed from 27/12/2023 to 23/01/2024.
21/07/2023: The following changes have been made:
1. The recruitment start date has been changed from 27/07/2023 to 01/09/2023.
2. The recruitment end date has been changed from 27/11/2023 to 27/12/2023.
08/06/2023: The following changes were made to the study record:
1. The recruitment start date was changed from 16/06/2023 to 27/07/2023.
2. The recruitment end date was changed from 16/09/2023 to 27/11/2023.
3. Ethics approval details added.
4. Added link to participant information sheet.
24/04/2023: The following has been changed:
1. The recruitment start date has been changed from 16/01/2023 to 16/06/2023.
2. The recruitment end date has been changed from 16/05/2023 to 16/09/2023.
22/11/2022: The following has been changed:
1. The recruitment start date has been changed from 01/09/2022 to 16/01/2023.
2. The recruitment end date has been changed from 01/11/2022 to 16/05/2023.
3. The trial website was added.
16/06/2022: Trial's existence confirmed by the National Institute for Health and Care Research (NIHR) (UK).
