AI-assisted Transcranial Duplex Sonography for Early Detection of Intracerebral Haemorrhage: HYPER-AI-SCAN

HYPER-AI-SCAN: HYPER-acute AI-assisted Sonographic Cerebral Hemorrhage Assessment Network

The goal of this observational study is to evaluate whether transcranial Doppler ultrasound, combined with artificial intelligence (AI), can help identify intracerebral haemorrhage (ICH) in people with acute stroke (both men and women, adults of all ages) within 48 hours of symptom onset.

The main questions it aims to answer are:

Is it feasible to perform standardized protocol transcranial ultrasound in acute stroke patients? Can AI models trained on ultrasound images accurately distinguish haemorrhagic stroke ("ICH suspected") from non-haemorrhagic stroke? There is no comparison group, because all participants will undergo both CT (as standard care) and ultrasound (research imaging), and the AI models will compare their ultrasound-based predictions against CT-confirmed diagnoses.

Participants will:

undergo a non-invasive transcranial ultrasound scan after CT confirms the type of stroke allow researchers to collect coded ultrasound images for AI model training provide clinical and imaging information (already collected as part of routine care) to help evaluate factors related to diagnostic accuracy No treatments or changes to clinical care will be introduced as part of the study.

Study Overview

• Status: Recruiting
• Conditions: Stroke; Intracerebral Hemorrhage; Intracerebral Haemorrhage; Intracerebral Hemorrhage Basal Ganglia

Detailed Description

Stroke is a medical emergency that can be caused either by a blocked blood vessel (ischaemic stroke) or by bleeding inside the brain (haemorrhagic stroke). These two types of stroke require very different treatments, and identifying which one is occurring as quickly as possible is essential.

Currently, the only reliable way to distinguish between these two types of stroke is with a brain scan such as a CT scan. However, CT is not always available immediately, especially in prehospital settings or in hospitals without 24/7 imaging access. As a result, patients may experience delays before receiving the correct treatment.

This study aims to explore whether a simple ultrasound scan of the brain, performed through the skull, can help identify haemorrhagic stroke more quickly. This technique is called transcranial Doppler ultrasound (TCD). It is fast, non-invasive, and uses no radiation.

A total of 500 patients with suspected stroke within 48 hours of symptom onset will be included. After the standard CT scan confirms the diagnosis, each participant will undergo a brief ultrasound scan following a structured protocol.

The ultrasound images will then be used to train and test artificial intelligence (AI) models, which will learn to recognize patterns associated with haemorrhagic stroke. These AI models will compare the ultrasound images with CT results and try to predict whether a bleed is present ("ICH suspected") or not.

The main goals of the study are:

To determine whether portable ultrasound can be performed reliably and consistently in real stroke patients.

To evaluate whether AI can support clinicians by interpreting these ultrasound images and distinguishing between haemorrhagic and non-haemorrhagic strokes.

All other clinical information-such as symptoms, timing of arrival, and medical history-will also be collected to understand which factors may influence the performance of ultrasound and AI.

Importantly, the ultrasound does not replace standard medical care and will not influence the treatment that patients receive. It is performed only for research purposes. The CT scan remains the reference test for diagnosis.

By combining ultrasound with AI, this project hopes to pave the way for future systems capable of assisting paramedics or physicians in identifying haemorrhagic stroke earlier, especially in settings where CT is not immediately available. Earlier recognition may help reduce delays in blood pressure management or treatment reversal for patients taking anticoagulants.

• Study Type: Observational
• Enrollment (Estimated): 500

Contacts and Locations

• Study Contact: Name: RENATO SIMONETTI, MD; Phone Number: 6660 +34934893000; Email: renato.simonetti@vallhebron.cat

Study Locations

• Spain
  • Catalonia
    • Barcelona, Catalonia, Spain, 08035
      • Recruiting
      • Hospital Universitario Vall d'Hebron
      • Contact: RENATO SIMONETTI, MD; Phone Number: 6660 +34934893000; Email: renato.simonetti@vallhebron.cat

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

The study population consists of adult patients (≥18 years) presenting with suspected acute stroke, either ischaemic or intracerebral haemorrhage. All participants undergo standard-of-care CT imaging to confirm diagnosis. Only supratentorial strokes are included, as patients with infratentorial haemorrhage or isolated subarachnoid haemorrhage are excluded.

Eligible patients must be able to undergo transtemporal transcranial Doppler ultrasound using a standardized acquisition protocol. The cohort includes both EMS-transported and in-hospital stroke cases. Participants represent a typical real-world acute stroke population, including individuals with varying stroke severity, vascular risk factors, and clinical profiles.

No modifications to clinical management occur as part of the study. Ultrasound acquisition and data collection are purely observational and performed after CT confirmation.

Description

Inclusion Criteria:

• Adult patients (age ≥18 years).
• Clinical diagnosis of acute stroke (ischemic or intracerebral hemorrhage).
• Able to undergo transtemporal transcranial ultrasound according to the standardized protocol (no clinical instability)
• Informed consent obtained from the patient or legally authorized representative, per local regulations.

Exclusion Criteria:

• Infratentorial hemorrhage (e.g., cerebellar or brainstem hemorrhage), due to limitations of transtemporal insonation.
• Isolated subarachnoid hemorrhage without parenchymal involvement.
• Hemodynamic instability or medical conditions requiring immediate life-saving intervention that preclude safe ultrasound recording.
• Known skull defects or prior craniectomy on the side required for contralateral insonation.
• Any condition that, in the opinion of the investigators, would interfere with protocol adherence or data accuracy.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Acute stroke patients; Single cohort of adults with suspected acute stroke (<48 hours from symptom onset), including both ischaemic and intracerebral haemorrhagic stroke. All participants will undergo standard diagnostic evaluation with computed tomography (CT). A research transcranial Doppler ultrasound (TCD) examination will then be performed using a standardized acquisition protocol. Coded sonographic data will be used to train and evaluate artificial intelligence (AI) models for the classification of intracerebral haemorrhage ("ICH suspected" vs. "No ICH"). No intervention or change in clinical management is introduced as part of the study.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Feasibility of standardized transcranial ultrasound acquisition in acute stroke	Feasibility will be measured as the proportion of patients in whom a diagnostic-quality transcranial ultrasound window is obtained (window quality grade 1 or 2). All examinations will follow the same standardized acquisition protocol, ensuring methodological consistency across operators. Diagnostic window success rate (%) will serve as the primary outcome.	At enrollment time (T0)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Exam acquisition time under a standardized protocol	Time (in seconds) from probe placement to acquisition of the first diagnostic-quality sonographic frame, obtained using the standardized sonographic protocol. Results will be reported as median, interquartile range (IQR), and distribution.	At enrollment time (T0)
Accuracy of AI-based classification of intracerebral haemorrhage using standardized TCD acquisitions	Evaluation of artificial intelligence models (CNN and transformer-based architectures) trained on ultrasound images acquired with a uniform standardized protocol. Performance will be assessed against CT-confirmed diagnosis.	From enrollment to the completion of imaging data collection at 16 months

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Operator-level variability in acquisition performance using the standardized protocol	Comparison of diagnostic level images acquisition time across operators performing scans with the same standardized protocol. Results will evaluate reproducibility and ease of training.	At enrollment time (T0)
Operator-level variability in acquisition performance using the standardized protocol	Comparison of transtemporal window acquisition success rate across operators performing scans with the same standardized protocol. Results will evaluate reproducibility and ease of training.	At enrollment time (T0)

Collaborators and Investigators

• Sponsor: Hospital Universitari Vall d'Hebron Research Institute

Publications and helpful links

General Publications

• Simonetti R, Canals P, Gonzalez Riveros JD, Alanis-Bernal M, Pancorbo O, Rodriguez-Luna D. Feasibility of an AI-assisted transcranial duplex sonography protocol for early detection of intracerebral haemorrhage: the HYPER-AI-SCAN single-centre prospective study. BMJ Open. 2025 Nov 19;15(11):e102903. doi: 10.1136/bmjopen-2025-102903.

Study record dates

Study Major Dates

• Study Start (Actual): March 14, 2025
• Primary Completion (Estimated): December 1, 2026
• Study Completion (Estimated): June 1, 2027

Study Registration Dates

• First Submitted: December 7, 2025
• First Submitted That Met QC Criteria: December 19, 2025
• First Posted (Actual): January 6, 2026

Study Record Updates

• Last Update Posted (Actual): January 6, 2026
• Last Update Submitted That Met QC Criteria: December 19, 2025
• Last Verified: December 1, 2025

More Information

Terms related to this study

• Keywords: stroke; Artificial Intelligence; Ischemic Stroke; Diagnostic Accuracy; Deep Learning; Feasibility Study; Machine Learning; Transcranial Doppler; transcranial ultrasound; Hemorrhagic Stroke; Neurosonology; intracerebral hemorrage; Portable Ultrasound; Prehospital Care
• Additional Relevant MeSH Terms: Cerebrovascular Disorders; Brain Diseases; Central Nervous System Diseases; Nervous System Diseases; Vascular Diseases; Cardiovascular Diseases; Pathologic Processes; Hemorrhage; Intracranial Hemorrhages; Pathological Conditions, Signs and Symptoms; Ischemic Stroke; Hemorrhagic Stroke; Stroke; Cerebral Hemorrhage
• Other Study ID Numbers: PR(AG)015/2025

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: IPD will not be shared because the study collects sensitive clinical and imaging information, and no formal data-sharing agreements or secure access frameworks are currently in place. Sharing policies may be reconsidered if a structured multicenter collaboration with appropriate data protections is established.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No