Mobile app shows potential for real-time stroke detection

February 9, 2023 1 minute read Source/Disclosure sauce: Raychev RI, et al. Abstract WMP120: Development of smartphone-enabled machine learning algorithms for autonomous stroke detection. Venue: International Stroke Congress. Dallas; February 8-10, 2023. Disclosure: Raychev reports receiving research grants for current research, support from Modest and the Society of Vascular and Interventional Neurology, and ownership of Modest and Spartan Micro. See research for relevant financial disclosures of all other authors. Add topic to email alert Get an email when new articles are posted Enter your email address to receive an email when new articles are posted . " data-actionsubscribe> subscribe Your request could not be processed. Please try again later.If this issue persists, please contact us customerservice@slackinc.com. back to helio A smartphone-enabled machine learning algorithm could be as effective as a neurologist in identifying signs and symptoms of acute stroke, according to preliminary research presented at the International Stroke Conference. "Many stroke patients do not arrive at the hospital in time for thrombectomy treatment. This is one reason why it is important to recognize stroke symptoms and call 9-1-1 immediately." Radoslav I. Raychev, MD, FAHA, A clinical professor of neurology at the University of California, Los Angeles said in a related press release: Preliminary results from a recent study suggest that a new smartphone app could be as effective as neurologists at detecting signs of stroke. Source: Adobe Stock Raychev et al developed FAST.AI. FAST.AI is a smartphone application designed to recognize stroke using machine learning algorithms to identify typical symptoms such as facial asymmetry, upper extremity weakness, and speech changes. Researchers analyzed data from 269 individuals diagnosed acute stroke Patients (median age, 71 years, 41% female) admitted to four major urban stroke centers in Eastern Europe between July 2021 and July 2022. Elements of the algorithm include 68 facial landmark points to measure asymmetry, a grasp-agnostic classifier to detect arm weakness, and an anomalous or slurred speech detection. Includes a frequency analysis component for The researchers performed all the tests within 72 hours of his hospitalization and compared each machine learning output to the neurologist's clinical impressions. According to the results, analysis of 18,311 face images demonstrated significant sensitivity (99.42%), specificity (93.67%) and accuracy (97.11%) in detection. facial asymmetrythe results of 43 motion trajectories detected arm weakness with a sensitivity of 71.42%, a specificity of 72.41%, and an accuracy of 72.09%. The researchers also reported that a preliminary analysis of the voice-altering algorithm confirmed that it worked well enough to detect anomalies. "Initial results confirm that the app identified acute symptoms of stroke as accurately as neurologists, and will help us improve the app's accuracy in detecting stroke signs and symptoms." Raychev said in a release. reference: Add topic to email alert Get an email when new articles are posted Enter your email address to receive an email when new articles are posted . " data-actionsubscribe> subscribe Your request could not be processed. Please try again later.If this issue persists, please contact us customerservice@slackinc.com. back to helio International Stroke Conference Sources 1/ https://Google.com/ 2/ https://www.healio.com/news/neurology/20230209/mobile-app-shows-promise-for-realtime-stroke-detection The mention sources can contact us to remove/changing this article