Online anomaly detection for long-term ECG monitoring using wearable devices
作者：
Highlights：
• A novel solution to perform online ECG monitoring and detect anomalous heartbeats.
• We learn dictionaries yielding sparse representations to describe normal heartbeats.
• We track heart-rate variations by transforming the user-specific dictionaries.
• Transformations are user-independent and learned from a publicly available dataset.
• Our solution is optimized to perform online ECG monitoring on wearable devices.
摘要
•A novel solution to perform online ECG monitoring and detect anomalous heartbeats.•We learn dictionaries yielding sparse representations to describe normal heartbeats.•We track heart-rate variations by transforming the user-specific dictionaries.•Transformations are user-independent and learned from a publicly available dataset.•Our solution is optimized to perform online ECG monitoring on wearable devices.
论文关键词：Online and long-term ECG monitoring,Anomaly detection,Domain adaptation,Wearable devices,Sparse representations
论文评审过程：Received 24 June 2018, Revised 11 November 2018, Accepted 17 November 2018, Available online 23 November 2018, Version of Record 13 December 2018.
论文官网地址：https://doi.org/10.1016/j.patcog.2018.11.019
