Researchers at the University of Utah's Scientific Computing and Imaging Institute say they have used machine learning to teach computers to read cardiac electrical signals and interpret electrocardiograms with greater accuracy.
The team employed machine learning to spot changes in the cardiac signal that indicates the first symptoms of a heart attack by isolating the electrical signals from the heart and analyzing changes before, during, and after simulated attacks. The computer then interprets these signals and categorizes the data.
The two categories isolated by the computer are "having a heart attack" and "not having a heart attack."
In comparison with human observers, the computer can ascertain the onset of a heart attack 10% faster, and also is 32% more accurate at identifying the early signs of a heart attack.
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found