Abstract 234: Current Stroke Solutions Using Artificial Intelligence: A Review of the Literature

IntroductionRecently, artificial intelligence (AI) has emerged as a tool for improving stroke diagnosis, enhancing critical decision‐making, and improving acute ischemic stroke (AIS) care. AI‐based platforms such as RapidAI, Brainomix® and Viz.ai, amongst others, have been studied to improve image analysis, detect large vessel occlusions (LVO), and predict patient outcomes in a timely manner. However, there is limited data pertaining to the impact of these AI platforms on real‐world patient care and management. The objective of this literature review is to evaluate the effectiveness and accuracy of AI platforms in facilitating the treatment of AIS.MethodsFollowing the PRISMA guidelines, a comprehensive systematic review was conducted using PubMed, Embase, Web of Science, and Scopus databases. Studies that meet the inclusion criteria were included. The final selection comprised studies that provided detailed analyses of AI tools, focusing on their sensitivity, specificity, accuracy, and comparative effectiveness.ResultsA total of 31 studies were included of which 29 studies primarily focused on detecting AIS or LVO, and 2 studies explored the use of AI in hemorrhagic strokes. AI tools including Viz.ai, RapidAI, and Brainomix®, demonstrated great utility in stroke management. These tools contributed to significant reductions in door‐to‐puncture times, enhanced accuracy in estimating core and penumbra volumes, and provided reliable assessments of ASPECT scores and the presence of intracranial hemorrhage. RapidAI was noted for its ability to rapidly identify LVOs. Viz.ai showed high accuracy in detecting both AIS and LVO, with sensitivity and specificity comparable to expert human interpretation. Brainomix® offered advantages in evaluating stroke severity and predicting outcomes, thereby aiding in the decision‐making process for intravenous thrombolysis and endovascular thrombectomy.ConclusionIntegrating AI tools in stroke care proves to be a valuable tool for aid diagnostic and management measures for greater accuracy and faster decision‐making, leading to improved patient outcomes. Considering the advancement in technology, AI‐based platforms are escalating to become vital assets for personalized care, providing a new level of hand‐tailored and expedited stroke management in the near future.