Clinical trial
AI-based Algorithm Development for Improved Diagnostics in the Intensive Care and Acute Stroke Setting Using Electroencephalography (SMART-EEG)
NCT07886957 is registered in ClinicalTrials.gov with Academisch Medisch Centrum - Universiteit van Amsterdam (AMC-UvA) as lead sponsor.
Follow this trial by email Email me the PDF
Summary
Recruiting, started March 2026. Observational study, 825 participants (estimated). Lead sponsor Academisch Medisch Centrum - Universiteit van Amsterdam (AMC-UvA); Netherlands. Primary measures: Prognostic performance of a subhairline EEG-based AI algorithm for neurological outcome prediction...; Diagnostic accuracy of a subhairline EEG-based AI algorithm for detection of intracerebral hemorrhage.
How it compares
71 similar trials: same condition (Hemorrhagic Stroke) and phase
Registry record
As filed in ClinicalTrials.gov.
- Title
- AI-based Algorithm Development for Improved Diagnostics in the Intensive Care and Acute Stroke Setting Using Electroencephalography (SMART-EEG)
- Lead sponsor
- Academisch Medisch Centrum - Universiteit van Amsterdam (AMC-UvA)
- Phase
- Observational
- Overall status
- Recruiting
- Start date
- March 26, 2026
- Completion date
- March 26, 2028
- Enrollment
- 825 (estimated)
- Sites
- 1
Conditions
MeSH condition names; the registry's own terms follow.
Hemorrhagic Stroke, Subarachnoid Hemorrhage
Registry terms: hemorrhagic stroke, intracerebral, postanoxic coma, subarachnoid haemorrhage (sah)
Who could take part, where it ran, who ran it
As filed in ClinicalTrials.gov.
- Sex
- All
- Minimum age
- 18 Years
- Healthy volunteers
- No
- Study type
- Observational
- Sites
- 1
- Countries
- Netherlands
- Lead sponsor
- Academisch Medisch Centrum - Universiteit van Amsterdam (AMC-UvA)
Arms
As registered; 3 arms.
| Arm | Type | Interventions |
|---|---|---|
| Postanoxic coma (UC1) | ||
Show registry wordingadults in postanoxic coma after cardiac arrest admitted to the ICU | ||
| Suspected stroke (UC2.1) | ||
Show registry wordingadults with suspected or confirmed stroke presenting to the emergency department | ||
| Suspected SAH (UC2.2) | ||
Show registry wordingadults with suspected or confirmed subarachnoid hemorrhage presenting to the emergency department | ||
Primary endpoints
Registered primary outcome measures; 2 primary.
| Measure | Time frame |
|---|---|
| Prognostic performance of a subhairline EEG-based AI algorithm for neurological outcome prediction in postanoxic coma | EEG at 24 hours after cardiac arrest; neurological outcome at hospital discharge (up to 3 months after CA) |
Show registry wordingPerformance of the AI algorithm, based on subhairline EEG recorded at 24 hours after cardiac arrest, in predicting poor neurological outcome. Poor outcome is defined as CPC 3-5 at hospital discharge. Performance is expressed as the area under the ROC curve (AUC), sensitivity, specificity, positive and negative predictive values, each with 95% confidence intervals. The optimal cut-off for poor outcome is defined as the highest sensitivity at a false positive rate of 0%. | |
| Diagnostic accuracy of a subhairline EEG-based AI algorithm for detection of intracerebral hemorrhage | EEG at presentation (within 24 hours of symptom onset); final diagnosis at hospital discharge (up to 3 months after presentation) |
Show registry wordingDiagnostic accuracy of the AI algorithm, based on subhairline EEG and clinical symptoms, in detecting intracerebral hemorrhage (ICH) in patients with suspected stroke. The reference standard is the final diagnosis based on neuroimaging. Accuracy is expressed as the area under the ROC curve (AUC), and as sensitivity, specificity, positive and negative predictive values at the optimal cut-off, each with 95% confidence intervals. | |
Lead sponsor's other trials
Academisch Medisch Centrum - Universiteit van Amsterdam (AMC-UvA), most recent start first.
+526 more trials
Similar trials
Same condition (Hemorrhagic Stroke) and phase, most recent start first.
+66 more trials See the condition page