Inter-patient ECG classification with i-vector based unsupervised patient adaptation

作者：

Highlights：

• Apply unsupervised patient adaptation for creating patient-specific ECG classifiers.

• Use i-vectors as compact representation of patient-specific information.

• Apply information-theoretic measure to rate the decision of ECG classifiers.

• A new two-stage adaptation approach that automates personalized ECG classification.

摘要

•Apply unsupervised patient adaptation for creating patient-specific ECG classifiers.•Use i-vectors as compact representation of patient-specific information.•Apply information-theoretic measure to rate the decision of ECG classifiers.•A new two-stage adaptation approach that automates personalized ECG classification.

论文关键词：Unsupervised adaptation,Patient-specific i-vectors,Arrhythmia,ECG classification,DNN adaptation

论文评审过程：Received 28 May 2021, Revised 18 July 2022, Accepted 3 August 2022, Available online 10 August 2022, Version of Record 17 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118410