QRS complexes and T waves localization in multi-lead ECG signals based on deep learning and electrophysiology knowledge

作者：

Highlights：

• Deep learning and DTAA method along with electrophysiology knowledge are presented.

• CNN and LSTM based on U-net can express spatial and temporal correlation features.

• DTAA method and electrophysiology knowledge reduced missed and false detections.

• The proposed method is more competitive than most state-of-the-art methods.

摘要

•Deep learning and DTAA method along with electrophysiology knowledge are presented.•CNN and LSTM based on U-net can express spatial and temporal correlation features.•DTAA method and electrophysiology knowledge reduced missed and false detections.•The proposed method is more competitive than most state-of-the-art methods.

论文关键词：Localization,U-Net,CNN,LSTM,Dynamic threshold adaptive adjustment,Electrophysiology knowledge

论文评审过程：Received 26 January 2022, Revised 8 March 2022, Accepted 1 April 2022, Available online 7 April 2022, Version of Record 8 April 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117187