Very-Low-Field MRI for the Diagnosis of Acute Stroke
Stroke
What this study involves
The registry does not currently label this study as recruiting. Ask about its status before planning a screening visit.
- Study type
- Interventional
- Phase
- Not applicable
- Age range
- 18 years - Not specified
- Sex eligible
- All sexes
- Healthy volunteers
- Accepted
- Participants
- 230 Estimated
These are starting points, not an eligibility decision. The research team must review the full requirements.
About the research
The goal of this clinical trial is to learn if a very-low-field MRI scanner (0.064 Tesla), combined with artificial intelligence (AI) image enhancement, can accurately diagnose acute ischemic stroke in adults with suspected stroke. Healthy adult volunteers are also included to optimize image quality. The main questions it aims to answer are:
How accurately does AI-enhanced very-low-field MRI detect ischemic stroke (sensitivity and specificity) compared with standard high-field MRI (3 Tesla)? Does AI enhancement improve diagnostic performance compared with very-low-field MRI images without AI?
Researchers will compare very-low-field MRI images, with and without AI enhancement, to the 3 Tesla MRI performed as part of routine care. Two neuroradiologists will read the images independently, without knowing the results of the other scan.
This study has three parts:
Part 1: healthy volunteers (30 participants). Participants will undergo one 3 Tesla MRI and one very-low-field MRI on the same day (45 minutes maximum).
Part 2: patients with recent stroke, 1 to 7 days after onset (40 participants). Participants will undergo one very-low-field MRI within 6 hours of their routine 3 Tesla MRI.
Part 3: patients with suspected stroke within the last 24 hours (160 participants). Participants will undergo one very-low-field MRI (about 30 minutes) within 4 hours of their routine 3 Tesla MRI, after any emergency treatment. About half will have a second very-low-field MRI around 24 hours later.
Who may be able to participate?
The list below is the sponsor’s full registry wording. It can include conditions, prior treatment, test results and other requirements. A match in search does not mean you qualify.
Inclusion Criteria:
All participants:
- Age 18 years or older
- Affiliated with or beneficiary of a social security scheme
Part 1 (healthy volunteers):
- No known neurological disease
- Written informed consent
- Registered in the national file of healthy volunteers (VRB)
Part 2 (subacute stroke):
- Confirmed acute ischemic stroke on 3T MRI
- Symptom onset 1 to 7 days before inclusion
- Very-low-field MRI feasible within 6 hours of 3T MRI
- Written informed consent
Part 3 (suspected hyperacute stroke):
- Clinical suspicion of acute stroke with symptom onset within 24 hours
- 3T MRI performed as part of routine care
- Very-low-field MRI feasible within 4 hours of 3T MRI, during working hours (8 a.m.-6 p.m.)
- Informed consent from the patient, or, if the patient is unable to consent, from a relative, the trusted person or an independent physician (emergency procedure, Article L.1122-1-3 of the French Public Health Code), followed by the patient's consent to continue as soon as possible
Exclusion Criteria:
- Contraindication to MRI (e.g., non-MRI-compatible pacemaker or implanted device, metallic foreign body)
- Pregnancy or breastfeeding
- Clinical instability incompatible with an additional MRI examination
- Inability to remain still for the duration of the examination
- Adults under legal protection (guardianship or curatorship), unless allowed by the protocol
- Participation in another interventional study that may interfere with this study
Treatment, groups and participation
The registry may describe treatments and study duration without a complete visit calendar. Ask the study team for the exact schedule, procedures, risks, costs and alternatives. Keep your usual care plan unless your treating clinician tells you otherwise.
- Assignment
- Not randomized
- Masking
- Single masked
Part 1 - Healthy volunteers
Not reported in this registry record.
Device: Very-low-field MRI
Part 2 - Subacute stroke patients
Not reported in this registry record.
Device: Very-low-field MRI
Part 3 - Suspected hyperacute stroke patients
Not reported in this registry record.
Device: Very-low-field MRI
Very-low-field MRI DEVICE
Brain MRI acquired on a 0.064 Tesla very-low-field scanner, including diffusion-weighted, FLAIR, T1-weighted sequences. Images are reconstructed with and without AI-based enhancement.
What researchers are measuring
Outcome measures describe what is being studied. They are not proof that a treatment works.
Primary measures
Sensitivity of AI-enhanced very-low-field MRI for the diagnosis of acute ischemic stroke
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Sensitivity (%) of AI-enhanced very-low-field MRI (0.064 T) for detecting acute ischemic stroke in patients with suspected hyperacute stroke (Part 3), using high-field 3T MRI as the reference standard. Images are read independently by two neuroradiologists blinded to 3T MRI results, with consensus in case of disagreement. Non-interpretable very-low-field MRI scans are classified as negative for stroke.
Specificity of AI-enhanced very-low-field MRI for the diagnosis of acute ischemic stroke
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Specificity (%) of AI-enhanced very-low-field MRI (0.064 T) for excluding acute ischemic stroke in patients with suspected hyperacute stroke (Part 3), using high-field 3T MRI as the reference standard. Images are read independently by two neuroradiologists blinded to 3T MRI results, with consensus in case of disagreement. Non-interpretable very-low-field MRI scans are classified as negative for stroke.
Secondary measures
Sensitivity and specificity of very-low-field MRI without AI enhancement for acute ischemic stroke
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Sensitivity and specificity (%) of non-enhanced very-low-field MRI for acute ischemic stroke, compared with AI-enhanced very-low-field MRI (McNemar test), using 3T MRI as reference.
Diagnostic accuracy of very-low-field MRI for intracranial hemorrhage
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Sensitivity and specificity (%) of very-low-field MRI (with and without AI enhancement) for detecting intracranial hemorrhage, using 3T MRI as reference.
Diagnostic accuracy of very-low-field MR angiography for proximal arterial occlusion
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Sensitivity and specificity (%) of very-low-field MR angiography for detecting proximal intracranial arterial occlusion, using 3T 3D time-of-flight MR angiography as reference
Diagnostic accuracy of very-low-field MRI for stroke mimics
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Proportion (%) of stroke mimics correctly identified on very-low-field MRI, using the final diagnosis based on 3T MRI as reference.
Agreement in ischemic lesion volume between very-low-field and 3T MRI
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Ischemic lesion volume (mL) measured on very-low-field and 3T diffusion-weighted images, compared with Pearson correlation and Bland-Altman analysis
Diagnostic accuracy of very-low-field MRI for hemorrhagic transformation
Time frame: 24 ± 12 hours after inclusion (patients with follow-up very-low-field MRI)
Detection of hemorrhagic transformation (ECASS II classification) on follow-up very-low-field MRI.
Intra- and inter-reader agreement
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Cohen's kappa coefficients for intra- and inter-reader agreement on the diagnosis of ischemic stroke on very-low-field MRI
Agreement on simulated therapeutic decisions
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Agreement (kappa) between simulated treatment decisions made by four reader pairs based on very-low-field MRI and decisions based on 3T MRI
Image quality metrics of AI-enhanced very-low-field MRI
Time frame: At inclusion (very-low-field MRI performed within 4 hours after 3T MRI)
Peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) of AI-enhanced very-low-field images compared with 3T images.
Review the complete result tables and limitations in the official record.
Open official results ↗Research locations (0)
The overall study status and a local site’s status can differ. Confirm that location’s availability before traveling.
No locations are listed. Use the official study contact.
Study team and timeline
- Lead sponsor
- Centre Hospitalier St Anne
- Study start
- 2026-11 Estimated
- Primary completion
- 2030-03 Estimated
- Study completion
- 2030-09 Estimated
- First posted
- Nov 10, 2026 Actual
- Last updated
- Nov 10, 2026 Actual
Official study contacts
Official study title
Very-Low-Field MRI for the Diagnosis of Acute Stroke
Ask Wellness Clinical Research
Your message goes to WCR, including when this study is run by another organization. WCR may offer general guidance but may not operate this trial, have access to its team or be able to enroll you.
Source: ClinicalTrials.gov API v2. Information is submitted by sponsors and investigators. NLM does not approve the safety or science of all listed studies. Registry disclaimer.
Retrieved: Nov 12, 2026 · Last registry update: Nov 10, 2026
