Home›Neurology› Study protocol for evaluating AI-based decision support in acute ischemic stroke management
Study protocol for evaluating AI-based decision support in acute ischemic stroke managementNew AI System Predicts Stroke Recovery Before Treatment Starts
medRxivPublished April 27, 2026Study authors: Rubiera, M.; Bendszus, M.; Leker, R. R.; Hilbert, A.; Werren, I.; Lopez-Ramos, L. M.; Ayesta, M.; Ng…DOI ↗Editorial oversight: Dr. Ji-eun Park, MD · Brain, Mind & Pain
AI-generated summary of the cited source, checked by automated accuracy review.
How we work
Share
Key Takeaway
Note that this protocol evaluates the feasibility of AI-based decision support for acute ischemic stroke treatment.
This study protocol outlines a prospective, multicenter, observational study designed to evaluate the implementation of the VALIDATE-CDSS, an AI-based clinical decision support system. The research will take place across three tertiary stroke centers, focusing on consecutive patients presenting with acute ischemic stroke.
The primary objective of the study is to assess the real-world feasibility and usability of the VALIDATE-CDSS within the hyperacute stroke workflow. The system is designed to generate individualized predictions of 3-month functional outcomes, measured by the modified Rankin Scale, across four specific treatment strategies: intravenous thrombolysis, endovascular thrombectomy, combined therapy, or no reperfusion.
Secondary outcomes of the protocol include evaluating the predictive performance of the AI, the agreement between model-suggested and actual treatments, and the incremental value of the system as data availability increases. The researchers also plan to assess potential bias across predefined subgroups.
While the protocol establishes a framework for testing AI-driven personalized treatment selection, the study aims to provide the first prospective real-world evidence on the clinical potential of this specific decision support tool in acute ischemic stroke management.
Imagine waking up with sudden weakness on one side of your body. Every minute counts when a stroke strikes. Doctors must choose between drugs or surgery quickly. This split-second choice can change a life forever. Families wait in the hallway while the clock ticks down. The pressure on medical teams is immense during this crisis.
Why Stroke Decisions Are So Hard
Stroke is a medical emergency that blocks blood flow to the brain. Current treatments work well, but picking the right one is tough. Doctors rely on experience and quick scans to guide them. Sometimes, the best path is not clear to the human eye. Missing the window for treatment can lead to permanent damage. This is why new tools are needed to help them decide.
How AI Maps Your Brain Damage
Think of the AI as a navigator for a complex road trip. It maps the brain's damage using detailed images from the hospital. The system looks at scans and history to guess the outcome. It compares four treatment options for each person in the room. It predicts recovery chances at three months for everyone involved. This helps doctors see the long-term result of their choices.
The Shadow Mode Safety Test
Three hospitals are testing this system in real time right now. They use a "shadow mode" to keep patients safe during the trial. The AI watches but does not change the plan for anyone. Doctors still make the final call on treatment for their patients. This tool will not replace your doctor's judgment. It acts as a second pair of eyes on the data. This ensures no human error slips through the cracks.
What Patients Gain From This Tool
The team aims to show if the AI matches real decisions. They want to see if it predicts recovery accurately across groups. Better predictions mean better care for everyone involved in the study. It could help choose between drugs or surgery more confidently. The system updates its guess at three key moments in care. This gives doctors a fresh look at the situation.
But there is a catch. The system is still being tested.
Experts say this could reduce errors in high-pressure moments. It adds a layer of data to the decision process. This helps ensure no option is missed during the rush. It brings technology into the room without taking control away.
Patients might get better care if the tool works well. It offers a clearer picture of what recovery looks like. Families can understand the risks better before signing consent forms. Knowing the odds helps people prepare for the weeks ahead.
The study is small and focused on specific centers only. It is still early in the research process for this tech. We do not know if it works everywhere yet. More data is needed to prove it is safe for all.
More data will come as the trial continues over time. Approval takes time to ensure safety for all patients. This could change stroke care in the near future. Researchers will share results in open journals for everyone.
Introduction Despite the proven benefits of reperfusion therapies in acute ischemic stroke, treatment decisions in the hyperacute phase remain complex and are rarely supported by individualized outcome predictions. Artificial intelligence (AI)-based clinical decision support systems (CDSS) offer potential real-time prognostic estimates, but prospective evidence of their feasibility and performance in routine clinical workflows is limited. Our aim is to prospectively evaluate real-time feasibility, usability, and predictive performance of an AI-based CDSS (VALIDATE-CDSS) for individualized outcome prediction in acute stroke care. Methods and analysis Prospective, multicenter, observational study enrolling consecutive patients with acute ischemic stroke presenting to three tertiary stroke centers. Clinical management will follow standard practice at the discretion of treating physicians. In parallel, a dedicated researcher will collect patient data in real time and input them into the VALIDATE-CDSS using a mobile application, operating in shadow mode without influencing clinical decisions. The system will generate individualized predictions of 3-month functional outcome (modified Rankin Scale) for four treatment strategies (intravenous thrombolysis, endovascular thrombectomy, combined therapy, or no reperfusion) at three sequential time points: baseline clinical data, non-contrast CT, and CT angiography. The primary outcome is the real-world feasibility and usability of the VALIDATE-CDSS in the hyperacute stroke workflow. Secondary outcomes include predictive performance, agreement between model-suggested and actual treatments, incremental value with increasing data availability, and assessment of potential bias across predefined subgroups. This study will provide prospective real-world evidence on the implementation and clinical potential of AI-based decision support for personalized treatment selection in acute ischemic stroke Ethics and dissemination Patient enrollment began after approval from the ethics committees of all participating centers. Results will be disseminated through peer-reviewed open-access journals and conference presentations. Following open science principles, anonymized data and metadata will be made publicly available in the Zenodo repository upon study completion. Trial registration: ClinicalTrials.gov (NCT05622539).
