Artificial intelligence to help improve fetal ultrasound scanning
| ISRCTN | ISRCTN65824874 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN65824874 |
| IRAS number | 292223 |
| Secondary identifying numbers | CPMS 52243, IRAS 292223 |
- Submission date
- 07/06/2022
- Registration date
- 07/06/2022
- Last edited
- 06/06/2024
- Recruitment status
- No longer recruiting
- Overall study status
- Completed
- Condition category
- Pregnancy and Childbirth
Plain English summary of protocol
Background and study aims
All pregnant women in the UK are offered an ultrasound scan at roughly halfway through their pregnancy to try and identify certain health issues in the baby. These scans are performed because babies are more likely to survive after birth if health issues are identified before birth, compared to after birth. This is because they can be treated properly as soon as they are born, rather than after a delay. Unfortunately, not all health issues are identified before birth. This is because ultrasound scans are difficult to perform and interpret, and there is a shortage of people able to do these scans. Some areas of the UK are much worse at diagnosing these babies than others. The aim of this study is to find out if AI can help the people doing these scans, by assisting them and telling them when the baby has a health issue. The research group has already used computers to help recognise different parts of the baby and to automatically measure the baby’s size. They would like to develop these computer systems to also tell which babies have health issues.
Who can participate?
1. Pregnant women whose fetus is either thought to have no health problems, or whose fetus has been identified as having a health problem.
2. Sonographers who regularly perform fetal anomaly ultrasound scans as part of their usual work.
What does the study involve?
Each pregnant woman will be scanned twice, once in the usual way and once using artificial intelligence assistance. The sonographers will be randomised to perform scans using one of these methods, and will each perform three scans.
What are the possible benefits and risks of participating?
This study places only a relatively small burden on participants, involving two extra ultrasound scans on a single day for the pregnant women, and the performance of three scans in a single day for the sonographers. Ultrasound in pregnancy has been shown to be safe and is in mainstream clinical use. The use of AI will not affect the safety of the scan. The main risk to pregnant women would be the detection of a fetal medical problem that had previously been overlooked, and there will be an 'incidental findings' procedure in place to manage these appropriately. Ultrasound scanning can be uncomfortable during pregnancy, and the scans will be stopped immediately if the participant requests. In addition, the pregnant participant will be able to undergo the scan in whatever position they find most comfortable. The researchers will contact the women after delivery for those included as normal controls, to ensure that the antenatal diagnosis of a normal fetal heart was correct. This could be upsetting if the baby had become unwell or died during delivery or in the newborn period.
There will be no direct benefit to either pregnant participants or volunteer sonographers for taking part.
Where is the study run from?
Guy’s and St Thomas’ Hospital, King’s College London (UK)
When is the study starting and how long is it expected to run for?
December 2019 to October 2023
Who is funding the study?
National Institute for Health and Care Research (NIHR) (UK)
Who is the main contact?
Dr Thomas Day
thomas.day@kcl.ac.uk
Contact information
Scientific
4th Floor North Wing
St Thomas’ Hospital Campus
KCL
Westminster Bridge Road
London
SE1 7EH
United Kingdom
 ORCID ID |
0000-0001-8391-7903 |
|---|---|
| Phone | +44 (0)7944326254 |
| thomas.day@kcl.ac.uk |
Study information
| Study design | Randomized; Interventional; Design type: Screening, Diagnosis, Device, Imaging |
|---|---|
| Primary study design | Interventional |
| Secondary study design | Randomised controlled trial |
| Study setting(s) | Hospital |
| Study type | Screening |
| Participant information sheet | Not available in web format, please use the contact details to request a patient information sheet |
| Scientific title | PROMETHEUS: Prospective tRial of Machine lEarning To Help fEtal Ultrasound Scanning |
| Study acronym | PROMETHEUS |
| Study objectives | Artificial intelligence assistance will improve the detection rate of fetal anomalies compared to standard manual scanning, and will result in a faster scan with a lower cognitive load. |
| Ethics approval(s) | Approved 19/04/2022, London Dulwich REC (The Old Chapel, Royal Standard Place, Nottingham, NG1 6FS, UK; +44 (0)2071048089; dulwich.rec@hra.nhs.uk), ref: 22/LO/0163 |
| Health condition(s) or problem(s) studied | Fetal ultrasound scanning |
| Intervention | 1. Ultrasound scan using artificial intelligence assistance 2. Ultrasound scan without using artificial intelligence assistance Each pregnant woman will be scanned twice, once in the usual way and once using artificial intelligence assistance. The sonographers will be randomised to perform scans using one of these methods, and will each perform three scans. |
| Intervention type | Other |
| Primary outcome measure | Detection rates of fetal anomaly between the two groups, assessed by written report by the sonographer, at baseline |
| Secondary outcome measures | 1. Time taken to complete ultrasound scan and written report, measured using a timekeeping device, at baseline 2. Cognitive load of sonographer after each scan, measured by NASA TLX scale, at baseline |
| Overall study start date | 12/12/2019 |
| Completion date | 02/10/2023 |
Eligibility
| Participant type(s) | Mixed |
|---|---|
| Age group | Adult |
| Lower age limit | 18 Years |
| Sex | Both |
| Target number of participants | Planned Sample Size: 87; UK Sample Size: 87 |
| Total final enrolment | 80 |
| Key inclusion criteria | 1. Cases: pregnant women with a fetus diagnosed with structural malformation between 12+0 and 27+6 weeks’ gestation. 2. Controls: pregnant women with a fetus shown not to have a structural malformation, between 18+0 and 27+6 weeks’ gestation. 3. Sonographers: professional staff who regularly and routinely undertake fetal anomaly screening ultrasound scans |
| Key exclusion criteria | For pregnant participants (cases/controls): 1. Any identified fetal extracardiac structural abnormality 2. Any fetal genetic or chromosomal abnormality 3. Participant withdrawal 4. Refusal of consent 5. Insufficient English language skills to provide informed consent For sonographers: 1. Any previous involvement in the iFIND research project 2. Previous involvement in the research leading up to this trial |
| Date of first enrolment | 15/11/2022 |
| Date of final enrolment | 02/10/2023 |
Locations
Countries of recruitment
- England
- United Kingdom
Study participating centres
Westminster Bridge Road
London
SE1 7EH
United Kingdom
De Crespigny Park
Denmark Hill
London
SE5 8AB
United Kingdom
Sponsor information
University/education
1.4 Hodgkin Building
Guy's Campus
London
SE1 1UL
England
United Kingdom
| Phone | +44 (0)2078486981 |
|---|---|
| richard.trembath@kcl.ac.uk | |
| Website | http://www.kcl.ac.uk/index.aspx |
"ROR" |
https://ror.org/0220mzb33 |
Funders
Funder type
Government
No information available
Results and Publications
| Intention to publish date | 01/10/2024 |
|---|---|
| Individual participant data (IPD) Intention to share | Yes |
| IPD sharing plan summary | Published as a supplement to the results publication |
| Publication and dissemination plan | Planned publication in a peer-reviewed journal |
| IPD sharing plan | Current IPD sharing statement as of 09/01/2023: The datasets generated and/or analysed during the current study will be published as a supplement to the results publication. Previous IPD sharing statement: The data-sharing plans for the current study are unknown and will be made available at a later date. |
Study outputs
| Output type | Details | Date created | Date added | Peer reviewed? | Patient-facing? |
|---|---|---|---|---|---|
| HRA research summary | 28/06/2023 | No | No | ||
| Preprint results | 25/05/2024 | 31/05/2024 | No | No |
Editorial Notes
06/06/2024: The recruitment start date was changed from 06/06/2022 to 15/11/2022.
31/05/2024: The following changes were made to the trial record:
1. The total final enrolment was added.
2. Preprint results added.
09/01/2023: Individual participant data (IPD) sharing statement and summary have been changed.
07/06/2022: Trial's existence confirmed by the NIHR.
