ESOC 2023: Artificial intelligence tool outperforms human emergency call managers in identifying strokes, new study shows
Receive notifications of the latest news, articles and much more!
MUNICH, May 23, 2023 /PRNewswire/ — Researchers in Denmark have developed a new artificial intelligence (AI) framework to address the number of heart attacks that go unrecognized by humans handling emergency calls. The framework outperformed 911 callers in stroke recognition for both sexes and in all age groups studied, indicating its potential as a complementary tool for early and accurate stroke identification in the future. .
The retrospective study, presented today at the 2023 European Stroke Organization (ESOC) Conference, was based on the Danish Stroke Registry and a data set of more than 1.5 million calls made to the Stroke Services. Copenhagen Emergency Physicians between 2015 and 2020, including more than 7,000 stroke-related calls. The researchers used this data to train an AI framework to transcribe the audio of the call and then predict stroke risk based on the transcribed text.
The results, evaluated in 2021 calls, revealed that the AI framework performed more effectively than emergency call managers in identifying stroke cases. The AI framework achieved a recall (sensitivity) of 63.0% and an accuracy (positive predictive value) of 24.9%, resulting in an F1 score of 35.7. In contrast, emergency call managers had a recall of 52.7% and an accuracy of 17.1%, resulting in an F1 score of 25.8.
The doctor. Jonathan Wenstrup, one of the lead authors of the study from Copenhagen University Hospital, commented: “As one of the first points of contact for patients seeking medical assistance, emergency call managers play a critical role in facilitating early and accurate recognition. of a stroke. Many cases of stroke may go undetected at this stage, leading to treatment delays that can have life-threatening consequences for patients.”
Across Europe, stroke is the second leading cause of death and one of the leading causes of disability in adults, affecting more than 1 million people each year. As the population continues to grow and age, the number of people suffering a stroke is projected to increase by 27% between 2017 and 2047 in the EU. Despite this, many strokes can be prevented and, if treated early, the likelihood of a positive outcome can be greatly improved.
“By implementing this cost-effective new support tool, we can improve stroke identification and ensure more patients receive appropriate and timely care, ultimately improving patient outcomes,” Dr. Wenstrup added.
View original content:https://www.prnewswire.com/news-releases/esoc-2023-new-artificial-intelligence-framework-301830979.html