US scientists develop quick-diagnosis stroke-assessment app
Scientists at Pennsylvania State University and Houston Methodist Hospital have developed a helpful smartphone app for patients who suspect that they’ve experienced a stroke. When someone has suffered a stroke, it’s imperative that they receive medical attention as soon as possible –however, these patients sometimes lose precious time waiting for assessment or scans to confirm the malady.
To counter this, the scientists started on their experimental app by recruiting over 80 patients with stroke symptoms who were asked to perform a speech test while being video-recorded on an iPhone. The data set generated was used to train an artificial intelligence (AI)-based algorithm.
“This is one of the first works that is enabling AI to help with stroke diagnosis in emergency settings,” said Penn State’s Associate Professor Sharon Huang.
The app was found to be 79% accurate at detecting stroke victims – it identified abnormalities in speech and facial muscular movements specific to the patients who were found to have indeed suffered a stroke, in just four minutes. The app is claimed to already be in line with the accuracy of tests currently performed in emergency rooms, but could be more accurate if the technology is developed further.
“The acquisition of facial data in natural settings makes our work robust and useful for real-world clinical use, and ultimately empowers our method for remote diagnosis of stroke and self-assessment.”
Category: Features, Technology & Devices
