Doctor Aurelio Vega Astudillo has introduced the Rapid AI software platform at Hospital Universitario Puerta de Hierro Majadahonda to accelerate critical stroke treatment decisions.
Vega, who serves as the head of the Neuroradiology Section at the university hospital located in Majadahonda near Madrid, explained that acting quickly is essential for patients experiencing a cardiovascular event such as an ischemic stroke.
An ischemic stroke occurs when a blood clot blocks an artery supplying oxygen-rich blood to the brain, leading to rapid cellular damage. Vega emphasized that speed is decisive in stroke intervention because every minute of delay results in the loss of 1.9 million neurons. This reality underpins modern medical protocols, which prioritize reducing the elapsed time from a patient's arrival at the emergency department to image acquisition and treatment delivery under the medical dictum that time is brain.
According to Vega, neuroradiology plays a fundamental role in diagnosing stroke cases. He noted that current treatment strategies for ischemic stroke rely almost entirely on the diagnostic information provided by medical imaging. These images enable specialized medical teams to determine whether a patient is a suitable candidate for intravenous thrombolysis using clot-dissolving medications, endovascular mechanical thrombectomy, or conservative medical management.
Diagnostic imaging and computed tomography protocols
Following an initial clinical assessment by an attending neurologist, neuroradiologists conduct specialized neuroimaging examinations to determine the exact nature and extent of the brain injury. While magnetic resonance imaging is occasionally utilized in specific scenarios, such as evaluating pediatric stroke patients, computed tomography scans remain the primary diagnostic tool in acute emergency settings.
Vega explained that neuroradiologists systematically obtain three distinct types of imaging examinations during a comprehensive computed tomography evaluation. The first is a simple non-contrast CT scan, which evaluates the structural state of the brain parenchyma to check for hemorrhage or established tissue damage.
The second examination is an Angio-CT scan, an angiography study that visualizes the cerebral arterial network. This scan identifies the exact location of the occluded cerebral artery and evaluates collateral blood circulation around the affected area.
The third component is a CT perfusion study, which measures blood flow dynamics across brain tissue. Vega pointed out that diagnostic accuracy has advanced significantly following the introduction of modern helical and multislice CT scanners that capture high-resolution images at rapid acquisition speeds.
These hardware capabilities are augmented by advanced software programs capable of generating three-dimensional Angio-CT vascular reconstructions. Specialized perfusion software allows neuroradiologists to evaluate tissue viability by distinguishing brain tissue that is already necrotized from surrounding ischemic tissue that remains viable and can still be saved through an efficient thrombectomy procedure.
Artificial intelligence integration with Rapid AI
To optimize diagnostic efficiency and reduce processing times, Hospital Universitario Puerta de Hierro Majadahonda recently integrated the Rapid AI platform into its acute stroke protocol. The platform is widely recognized as one of the leading artificial intelligence software systems for acute stroke management in hospitals worldwide.
Vega clarified that the artificial intelligence platform is not designed to replace the clinical judgment of neuroradiologists or neurologists. Instead, the technology serves to accelerate image processing and streamline complex data analysis, facilitating faster clinical decisions during the critical early minutes of care.
Under the automated workflow, as soon as a simple CT scan, Angio-CT, or CT perfusion scan is performed in the radiology suite, the digital images are automatically transmitted to the secure Rapid AI server for analysis.
Within one to three minutes of transmission, the software processes the radiological data and generates quantitative maps and numerical measurements. These results are distributed instantaneously to the multidisciplinary stroke team via hospital computer workstations and mobile phones.
Automated scan parameters and clinical alerts
The Rapid AI software evaluates multiple clinical parameters to assist physicians in selecting the appropriate therapeutic path. The Rapid CTA and Rapid LVO modules specialize in detecting large vessel occlusions, such as blockages in the internal carotid artery or middle cerebral artery. Upon detecting a major vascular blockage, the software sends immediate automated alerts to notify stroke team members.
The Rapid CTP perfusion module calculates key volume metrics regarding cerebral tissue damage. It measures the volume of the irreversible infarct core where tissue has died, the volume of the ischemic penumbra where tissue is compromised but salvageable, and the mismatch between the two volumes.
Vega noted that calculating the mismatch between core and penumbra tissue is vital for determining whether a mechanical thrombectomy will benefit the patient, particularly in cases involving an extended therapeutic window.
Additionally, the Rapid Aspects module automatically evaluates non-contrast CT images to calculate an Aspects score. This quantitative score highlights specific anatomical regions of the brain that have already suffered irreversible infarction.
Extended treatment windows and network coordination
Vega outlined several practical advantages associated with adopting the artificial intelligence software. By decreasing the time required between initial scan acquisition and therapeutic decision making, the platform facilitates the prompt activation of interventional neuroradiology teams who perform catheter procedures.
The automated delivery of imaging data also enhances communication and coordination among different hospitals within a regional stroke care network, allowing remote specialists to review scan findings simultaneously.
Furthermore, Vega explained that the platform allows medical teams to evaluate stroke patients for mechanical thrombectomy even up to 24 hours after the initial onset of symptoms. This extended treatment timeframe is applicable for patients who fulfill the specific imaging criteria established in the landmark Dawn and Defuse-3 clinical trials.
While praising its reliability, Vega cautioned that the software is not flawless. The system can produce occasional false positive or false negative results, which is why official medical guidelines emphasize that Rapid AI remains a supportive tool that cannot substitute for expert interpretation by neuroradiologists and neurologists.
Patient recovery rates and future neurointerventional advances
The Rapid AI system is currently deployed in more than 2,500 hospitals spanning over 100 countries. It includes multiple software modules that have received regulatory clearance from health agencies, including the United States Food and Drug Administration, for various acute stroke applications.
Vega reported that the combination of rapid automated diagnostic software and rapid technological progress in micro-catheters used to physically extract blood clots from cerebral arteries has substantially improved patient prognoses. Today, medical teams are able to successfully recover more than 30 percent of acute stroke patients who previously faced severe, lasting neurological impairment.
Looking toward the future, Vega stated that the field of neurointervention is currently undergoing its most significant transformation since mechanical thrombectomy was first introduced into clinical practice.
Over the next five to ten years, Vega anticipates major technological developments across eight key areas: real-time artificial intelligence integrated directly into procedural tools to assist clinicians during surgery, endovascular surgical robotics, smart micro-catheters, advanced imaging modalities, novel devices for treating cerebral aneurysms, new pharmacological treatments for stroke, personalized medicine tailored to individual patient profiles, and augmented reality systems for surgical guidance.
