Evaluating an artificial intelligence tool for measuring the aorta from computed tomography (CT) scans
| ISRCTN | ISRCTN12897963 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN12897963 |
| Integrated Research Application System (IRAS) | 356296 |
| Central Portfolio Management System (CPMS) | 71579 |
| Sponsor | AIATELLA Oy |
| Funder | AIATELLA Oy |
- Submission date
- 02/05/2026
- Registration date
- 06/05/2026
- Last edited
- 05/05/2026
- Recruitment status
- No longer recruiting
- Overall study status
- Completed
- Condition category
- Circulatory System
Plain English summary of protocol
Background and study aims
The aorta is the body's main blood vessel, carrying blood from the heart to the rest of the body. It can become enlarged over time, which may lead to serious and life-threatening conditions. Doctors use CT scans to measure the size of the aorta and monitor patients. A CT scan is a test that takes detailed pictures of the inside of your body. Currently these measurements are made by hand by specialist doctors, which takes time and can vary between readers. Aorta AIM is an AI software tool designed to make these measurements automatically. This study aims to test how accurately Aorta AIM measures the aorta from CT scans compared to specialist doctors.
Who can participate?
The study will use CT scan images obtained as part of routine clinical care at three NHS hospitals in England. All patient-identifying information is removed before the data is used for research. No patients are directly approached or recruited. Data may be included unless the patient opted out of NHS data sharing through the National Data Opt-Out programme.
What does the study involve?
Existing CT scans from routine clinical care are identified by staff at each hospital and stripped of any patient-identifying information. Aorta AIM measures the aorta from each scan. Specialist doctors then measure the same scans, without seeing the AI results or each other's measurements. The two sets of measurements are then compared to assess how accurately Aorta AIM performs. No patients are contacted and no changes are made to anyone's care.
What are the possible benefits and risks of participating?
There are no risks to patients as this study only uses existing scan images with no patient contact. All patient-identifying information has been removed from the data. If Aorta AIM performs well, it could in future help doctors measure the aorta more quickly and consistently, benefiting patients with aortic conditions.
Where is the study run from?
The study is led by Northumbria Healthcare NHS Foundation Trust, with research activities also taking place at Newcastle upon Tyne Hospitals NHS Foundation Trust and South Tyneside and Sunderland NHS Foundation Trust.
When is the study starting and how long is it expected to run for?
May 2026 to September 2026.
Who is funding the study?
AIATELLA Oy (Finland).
Who is the main contact?
Jack Parker - research@aiatella.com, jack@aiatella.com
Contact information
Principal investigator
North Tyneside Hospital, Rake Lane
North Shields
NE29 8NH
United Kingdom
 ORCID ID |
0000-0001-8460-9873 |
|---|---|
| Phone | +44 3448118111 |
| david.ripley@nhct.nhs.uk |
Scientific, Public
Lapinlahdenkatu 16
Helsinki
00180
Finland
| ORCID ID |
0000-0003-2253-6079 |
|---|---|
| Phone | +358 4578313729 |
| jack@aiatella.com |
Study information
| Primary study design | Observational |
|---|---|
| Observational study design | Retrospective validation study |
| Scientific title | AI-CARE: Artificial Intelligence for Cardiovascular Analysis and Risk Evaluation |
| Study acronym | AI-CARE |
| Study objectives | |
| Ethics approval(s) | Ethics approval not required |
| Health condition(s) or problem(s) studied | Aortic pathology, including aneurysm and dissection |
| Methodology | Retrospective CT studies from adult patients who underwent routine clinical aortic imaging at three NHS Trusts in England between January 2016 and August 2025 are identified and pseudonymised by the local clinical care team at each site. Pseudonymised DICOM images and demographic data (age, sex) are transferred to a secure research environment. The Aorta AIM v1.0 software processes each scan to produce automated measurements of maximum aortic diameter across standardised anatomical regions. Separately, fellowship-trained radiologists and/or cardiologists independently measure the same scans following a standardised measurement protocol, remaining blinded to AI outputs and to each other's measurements throughout. An independent statistician then compares AI and expert measurements. No patient contact occurs at any stage. |
| Intervention type | Device |
| Phase | Not Applicable |
| Drug / device / biological / vaccine name(s) | Aorta AIM v1.0 |
| Primary outcome measure(s) |
|
| Key secondary outcome measure(s) |
|
| Completion date | 28/08/2026 |
Eligibility
| Participant type(s) | |
|---|---|
| Age group | Mixed |
| Lower age limit | 22 Years |
| Upper age limit | 120 Years |
| Sex | All |
| Target sample size at registration | 225 |
| Key inclusion criteria | 1. Adults ≥22 years at time of imaging 2. Imaging performed 1 Jan 2016 – 31 Aug 2025 3. CT imaging of the whole, thoracic, or abdominal aorta, with or without contrast enhancement 4. Slice thickness ≤3mm |
| Key exclusion criteria | 1. Metal implants <5cm from aorta (e.g. thoracic stent) 2. Prior aortic surgery 3. Severe motion artifacts (e.g. >3mm vessel blurring) 4. Congenital aortic anomalies (e.g. coarctation, vascular rings) 5. Post-traumatic aortic repairs 6. Patient opted-out of having their data used for medical research via the NHS ‘National patient opt-out scheme’ before the date of cross-referencing by the site team |
| Date of first enrolment | 11/05/2026 |
| Date of final enrolment | 31/07/2026 |
Locations
Countries of recruitment
- United Kingdom
- England
Study participating centres
Rake Lane
North Shields
NE29 8NH
England
Freeman Road
High Heaton
Newcastle upon Tyne
NE7 7DN
England
Kayll Road
Sunderland
SR4 7TP
England
Results and Publications
| Individual participant data (IPD) Intention to share | No |
|---|
Editorial Notes
05/05/2026: Trial's existence confirmed by NHS HRA.
