A novel Domain Adaptive Residual Network for automatic Atrial Fibrillation Detection

作者：

Highlights：

• A domain adaptive residual network is proposed for atrial fibrillation Detection.

• It extracts deep features without manual feature extraction and feature selection.

• The feature extractor and classifier utilize residual block for good performance.

• The multi-kernel maximum mean discrepancy is incorporated in the training process.

• It improves accuracy by 4.50% on average and the F1 score by 4.28% on average.

摘要

•A domain adaptive residual network is proposed for atrial fibrillation Detection.•It extracts deep features without manual feature extraction and feature selection.•The feature extractor and classifier utilize residual block for good performance.•The multi-kernel maximum mean discrepancy is incorporated in the training process.•It improves accuracy by 4.50% on average and the F1 score by 4.28% on average.

论文关键词：Atrial Fibrillation Detection,ECG,Domain adaptation,Residual Block

论文评审过程：Received 3 April 2020, Revised 13 May 2020, Accepted 7 June 2020, Available online 9 June 2020, Version of Record 26 June 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106122