An observational study in Stroke, Ischemic Stroke and Intracerebral Hemorrhage, sponsored by Hospital Universitari Vall d'Hebron Research Institute. Recruiting at 7 sites in 6 countries. Open to participants aged 18 Years and older. Per ClinicalTrials.gov, last updated 2026-11-09.
Sponsored by Hospital Universitari Vall d'Hebron Research Institute · Observational
Stroke is a leading cause of death and adult disability. Stroke care is still limited by fragmented care pathways, non-standardised diagnostic work-up and follow-up, and local, non-interoperable data systems across the pre-hospital, in-hospital and post-hospital settings.
UMBRELLA-P is a multicentre, prospective, observational study in 7 university hospitals in 6 European countries (Spain, Italy, Belgium, the Netherlands, Germany and Switzerland). Adults with clinically suspected acute stroke or transient ischemic attack (TIA) are managed according to standard of care. Routine clinical, imaging, laboratory, ECG and patient-reported data are collected, pseudonymised, stored on a dedicated server at each hospital and harmonised to the OMOP Common Data Model.
The harmonised data form a federated real-world data platform (U-Platform). On this platform, artificial intelligence (AI) models are trained and validated through federated learning, so that no patient data leave the participating sites. The models aim to predict short- and long-term outcomes, reduce time to treatment, and support diagnosis, triage, discharge, follow-up and secondary prevention. Five clinical sub-studies address pre-hospital triage, fast track to reperfusion, emergent intracerebral hemorrhage (ICH) management, TIA/minor stroke, and cardiac evaluation with blood biomarkers.
The study is observational. The outputs of the AI models are used for research only and will not influence clinical decisions. Participants are followed for up to 12 months.
Design: multicentre, observational, prospective cohort study. Standard-of-care stroke pathways are followed and no study-specific intervention is performed.
Data flow:
A preceding retrospective phase harmonised historical cohort data across the stroke care pathway. Models developed on retrospective data are validated on the independent, prospectively collected data of this study.
Clinical sub-studies (work packages):
CE-marked digital solutions (e.g. StrokeViewer, Stroke Triage App, Cardiologs/Predict AF, NORA App, JOIN) are used only within their intended purpose as part of routine care. AI model evaluation takes place outside the clinical workflow and does not constitute a performance study within the meaning of Article 62 of the EU MDR.
Recruitment period: October 2026 to February 2029 (28 months). Consecutive eligible patients are included. Written informed consent is obtained from the patient or next of kin; deferred consent is permitted because of the acute setting.
Funding: Innovative Health Initiative Joint Undertaking (IHI JU), grant agreement No 101172825.
2,612 studies on the registry are indexed under Ischemic Stroke; 971 are open to participants now.
This study's planned enrollment of 10,000 is above the median of 300 across 771 observational studies indexed under Ischemic Stroke.
Browse Ischemic Stroke studies →Hospital Universitari Vall d'Hebron Research Institute is the lead sponsor of 276 studies on the registry; 64 are open to participants now.
Compared within Ischemic Stroke, this study's primary condition as ClinicalTrials.gg classifies it; the registry's first condition term is Stroke. The counts are of registry records indexed under each term, refreshed daily. How we classify →
Consecutive adult patients with clinically suspected acute stroke or TIA who are evaluated or admitted at 7 European comprehensive stroke centres, including pre-hospital (EMS) referrals.
Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.
Additional inclusion criteria by sub-study:
Exclusion Criteria:
* None
Underlined terms in the criteria open these too. Explained by clinicaltrials.gg, not part of the registry record. The study team decides who can take part.
A “mini-stroke”: temporary stroke-like symptoms caused by reduced blood flow to part of the brain, spinal cord or eye, without signs of a new stroke on a scan.
Stroke symptoms need emergency care at once, even if they go away.
Before you join, the study team explains what taking part involves — its purpose, what happens, the risks and possible benefits — and answers your questions, so you can decide freely. Consent is usually recorded on a form (the informed consent form, ICF) signed by you or, where the law allows, by someone legally authorised to decide for you.
You can keep asking questions during the study, you are told of new information that matters, and you can choose to leave at any time.
TIA (transient ischaemic attack)
A “mini-stroke”: temporary stroke-like symptoms caused by reduced blood flow to part of the brain, spinal cord or eye, without signs of a new stroke on a scan.
Stroke symptoms need emergency care at once, even if they go away.
Explained by clinicaltrials.gg, not part of the registry record. The study team decides who can take part.
Informed consent
Before you join, the study team explains what taking part involves — its purpose, what happens, the risks and possible benefits — and answers your questions, so you can decide freely. Consent is usually recorded on a form (the informed consent form, ICF) signed by you or, where the law allows, by someone legally authorised to decide for you.
You can keep asking questions during the study, you are told of new information that matters, and you can choose to leave at any time.
Explained by clinicaltrials.gg, not part of the registry record. The study team decides who can take part.
Patients with suspected stroke according to ambulance personnel, or confirmed LVO stroke, with symptom onset or last seen well less than 24 hours earlier.
Other: Standard of care (no study intervention)
Patients with suspected stroke according to ambulance personnel, or confirmed LVO stroke, with symptom onset or last seen well less than 24 hours earlier (same criteria as WP2).
Other: Standard of care (no study intervention)
Patients with intracerebral hemorrhage, regardless of sex, risk factors or etiology.
Other: Standard of care (no study intervention)
Patients with TIA or minor stroke (NIHSS below 7) and premorbid mRS below 4.
Other: Standard of care (no study intervention)
Patients with acute ischemic stroke, with or without atrial fibrillation.
Other: Standard of care (no study intervention)
Routine stroke care according to European Stroke Organisation guidelines. No study-specific intervention is administered. CE-marked digital tools used in routine care (StrokeViewer, Stroke Triage App, Cardiologs/Predict AF, NORA App, JOIN, Stroke Connect) are used within their intended purpose only.
Area Under the Curve (AUC) of the Pre-hospital AI Model for Detection of Large Vessel Occlusion (LVO) Stroke (WP2)
Discriminative performance (area under the receiver operating characteristic curve, with 95% CI) of the pre-hospital AI model for identifying LVO stroke, using the final in-hospital diagnosis as the reference standard.
Time frame: At hospital admission (Day 1)
AUC of the Image-based AI Tool for Prediction of Successful Reperfusion After Endovascular Thrombectomy (WP3)
Discriminative performance (area under the receiver operating characteristic curve, with 95% CI) of the image-based AI tool for predicting successful reperfusion, defined as expanded Thrombolysis in Cerebral Infarction (eTICI) grade 2b-3 on final angiography.
Time frame: Day 1 (at the end of the endovascular procedure)
AUC of the AI Model for Prediction of Early Hematoma Expansion in Intracerebral Hemorrhage (WP4)
Discriminative performance (area under the receiver operating characteristic curve, with 95% CI) of the AI model for predicting hematoma expansion on follow-up imaging compared with baseline imaging.
Time frame: 24 hours after admission
C-statistic of the AI Risk Model for Prediction of Recurrent Stroke After TIA or Minor Stroke (WP5)
Discrimination (C-statistic, with 95% CI) of the AI risk model for predicting recurrent stroke, benchmarked against the ABCD2 score.
Time frame: 12 months
AUC of the Multimodal AI Tool for Identification of a Cardio-aortic Source of Embolism (WP6)
Discriminative performance (area under the receiver operating characteristic curve, with 95% CI) of the multimodal AI tool combining cardiac CT, ECG and blood biomarkers for identifying a cardio-aortic source of embolism, using the final etiological diagnosis as the reference standard.
Time frame: 7 to 30 days after stroke onset
AUC of the Pre-hospital AI Model for Detection of Intracerebral Hemorrhage (WP2)
Discriminative performance (area under the receiver operating characteristic curve, with 95% CI) of the pre-hospital AI model for identifying intracerebral hemorrhage, using the final in-hospital diagnosis as the reference standard.
Time frame: At hospital admission (Day 1)
AUC of the Pre-hospital AI Model for Differentiating Stroke From Stroke Mimic (WP2)
Discriminative performance (area under the receiver operating characteristic curve, with 95% CI) of the pre-hospital AI model for differentiating stroke from stroke mimic, using the final in-hospital diagnosis as the reference standard.
Time frame: At hospital admission (Day 1)
Accuracy of the Image-based AI Tool for Prediction of the Arterial Access Route (WP3)
Percentage of endovascular procedures in which the arterial access route predicted by the AI tool (e.g., femoral or radial) matches the access route actually used.
Time frame: Day 1 (at the end of the endovascular procedure)
Accuracy of the Image-based AI Tool for Prediction of the First-line Thrombectomy Technique (WP3)
Percentage of endovascular procedures in which the first-line thrombectomy technique predicted by the AI tool (e.g., stent retriever, aspiration or combined) matches the technique actually used.
Time frame: Day 1 (at the end of the endovascular procedure)
AUC of the AI Model for Prediction of Early Neurological Deterioration in Intracerebral Hemorrhage (WP4)
Discriminative performance (area under the receiver operating characteristic curve, with 95% CI) of the AI model for predicting early neurological deterioration, assessed with the National Institutes of Health Stroke Scale (NIHSS).
Time frame: 24 hours after admission
AUC of the AI Model for Prediction of Poor Functional Outcome After Intracerebral Hemorrhage (WP4)
Discriminative performance (area under the receiver operating characteristic curve, with 95% CI) of the AI model for predicting poor functional outcome, defined as a modified Rankin Scale (mRS) score of 3 to 6.
Time frame: 70 to 90 days after stroke onset
AUC of the Multimodal AI Tool for Prediction of Newly Diagnosed Atrial Fibrillation (WP6)
Discriminative performance (area under the receiver operating characteristic curve, with 95% CI) of the multimodal AI tool combining cardiac CT, ECG and blood biomarkers for predicting atrial fibrillation newly diagnosed during follow-up.
Time frame: 12 months
Percentage of Participants With Recurrent Stroke
Any new ischemic or hemorrhagic stroke occurring after the index event.
Time frame: 12 months
Percentage of Participants With Major Adverse Cardiovascular Events (MACE)
Composite of recurrent stroke, myocardial infarction and cardiovascular death.
Time frame: 12 months
Percentage of Participants With Hospital Re-admission
Any unplanned hospital re-admission after discharge from the index hospitalization.
Time frame: 12 months
All-cause Mortality
Percentage of participants who died from any cause.
Time frame: 12 months
Functional Outcome Measured With the Modified Rankin Scale (mRS)
The mRS ranges from 0 (no symptoms) to 6 (death); higher scores indicate worse functional outcome.
Time frame: 70 to 90 days after stroke onset
Plan to share: No
No publications or documents are linked to this record.
From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗
Eligibility is decided by the study team. Share this record with your doctor or contact the team directly.
Contact study teamGet an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
Hospital Universitari Vall d'Hebron Research Institute →