Periodic Reporting for period 2 - Deepstroke-Acc (Paradigm shift in stroke: accurate diagnosis and evaluation of all treatment options in any hospital, at any time)
Berichtszeitraum: 2023-11-01 bis 2025-01-31
Acute stroke is a medical emergency where faster admission and treatment by specialized stroke teams significantly improve recovery. However, stroke care inequalities across Europe are evident. Only about 30% of patients receive specialized stroke unit (SU) care, which improves clinical outcomes but is primarily available in major hospitals. As a result, SU access varies widely within and between countries, often favoring urban areas over rural ones. While stroke mortality has declined over the past two decades, survival rates still depend heavily on geographic location.
Today, stroke treatment decisions rely on expert physicians interpreting advanced medical imaging. Standard (non-contrast CT) and advanced imaging (CT angiography [CTA] and CT perfusion [CTP]) use the same equipment, which is available in most hospitals. However, CTA/CTP requires skilled practitioners for both image acquisition and interpretation, as contrast agents must be injected into the bloodstream. Due to a global shortage of trained specialists, over 50% of hospitals worldwide cannot perform CTA imaging 24/7, and only 5% can conduct CTP imaging continuously.
Methinks AI is a Barcelona-based company dedicated to providing universal and timely medical assistance to enable life-saving treatments worldwide. Our primary focus is stroke care. Our flagship AI product, Methinks Stroke Suite, received CE Mark certification in January 2023. This innovative solution expedites time-to-treatment in any hospital by analyzing standard medical imaging, such as non-contrast CT scans. In stroke cases, "time is brain," meaning every minute counts in determining the risk of disability or death.
Methinks Stroke Suite is an AI-powered radiological triage and notification system that uses advanced algorithms to detect suspected Large Vessel Occlusions (LVO) in Non-Contrast Computed Tomography (NCCT) and Computed Tomography Angiography (CTA) images. It also identifies Intracerebral Hemorrhages (ICH) on NCCT scans. Seamlessly integrating into clinical workflows, it leverages widely available CT scans in hospitals worldwide to enable early, rapid, and precise triaging of acute stroke patients.
Regarding the FDA, only two companies have obtained clearance so far. We compared their submissions with the results of our validation studies and clearly outperformed them in sensitivity and specificity. Additionally, our study includes one more occlusion type (M2). These results remain confidential for now but will be released soon.
Our optimized product, technology, and AI models have also been validated in both retrospective and prospective studies across the EU and the US, achieving excellent results, all surpassing the thresholds required by the FDA for certification.
We have developed a new Methinks App, integrated into Microsoft Teams and available on the Azure Marketplace through the MACC program. Our app features a viewer that facilitates stroke identification, as well as a communication system to share results, chat, and video call with clinicians within the same hospital or across hospitals. This system improves stroke care coordination between the spoke (where initial analysis is done) and the hub (larger hospitals capable of advanced treatments, such as thrombectomy).
We surveyed clinicians using our new app in prospective studies, and physician satisfaction has been very high. The solution received an average rating of 4.6/5 for ease of use, 4.2/5 for productivity improvement, and 3.9/5 for analysis speed.
All of this has allowed us to validate our product-market fit, which was further confirmed just after the project’s conclusion with the signing of a distribution agreement with Medtronic and our first direct sales. These two key milestones also confirmed our business and revenue model, as well as the market entry strategy developed as part of this EIC Accelerator project.