Real-time frequency-independent single-Lead and single-beat myocardial infarction detection

作者：

Highlights：

• Long-Short Term Memory (LSTM) for real-time sampling frequency independent myocardial infarction (MI) detection

• Inter-patient myocardial infarction detection from single heartbeat

• Very widely available lead, Lead, II is used.

• Classification performance independent of signal sampling frequency from 202 Hz to 2.8 kHz

• Real-time performance on signals with sampling frequencies up to 500 Hz

• Trained over the newly released PTB-XL database

• Tested over multiple independent datasets to asses expected performance on novel data and the effects of domain-shift

摘要

•Long-Short Term Memory (LSTM) for real-time sampling frequency independent myocardial infarction (MI) detection•Inter-patient myocardial infarction detection from single heartbeat•Very widely available lead, Lead, II is used.•Classification performance independent of signal sampling frequency from 202 Hz to 2.8 kHz•Real-time performance on signals with sampling frequencies up to 500 Hz•Trained over the newly released PTB-XL database•Tested over multiple independent datasets to asses expected performance on novel data and the effects of domain-shift

论文关键词：Cardiovascular disease,Electrocardiograms,Frequency Independence,Long Short-Term Memory Neural Network,Myocardial infarction,Real-time processing

论文评审过程：Received 17 February 2021, Revised 29 July 2021, Accepted 21 September 2021, Available online 1 October 2021, Version of Record 15 October 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102179